The effects of sunlight exposure on mortality: a systematic review of epidemiological studies

Systematic review; Sunlight; Ultraviolet radiation; Mortality; Cancer; Cardiovascular disease ALL Metrics - Views Downloads How to cite this article Parkhouse T, Spiga F, Rhodes LE et al. The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.13980.1) NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article. Export Citation Sciwheel EndNote Ref. Manager Bibtex ProCite Sente Select a format first ▬ ✚ Systematic Review [version 1; peer review: 4 approved with reservations, 2 not approved] Thomas Parkhouse https://orcid.org/0000-0001-6773-5376 1, Francesca Spiga1, Lesley E Rhodes2,3, [...] Sarah Dawson1,4, Katie E Webster https://orcid.org/0009-0002-7997-4133 1, Deborah M Caldwell https://orcid.org/0000-0001-8014-7480 1, Julian P T Higgins1,4Thomas Parkhouse https://orcid.org/0000-0001-6773-5376 1, Francesca Spiga1, [...] Lesley E Rhodes2,3, Sarah Dawson1,4, Katie E Webster https://orcid.org/0009-0002-7997-4133 1, Deborah M Caldwell https://orcid.org/0000-0001-8014-7480 1, Julian P T Higgins1,4 PUBLISHED 18 Jun 2025 Author details Author details 1 NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, England, BS8 2PS, UK 2 Division of Musculoskeletal and Dermatological Sciences, School of Biological Sciences,, The University of Manchester, Manchester, England, M13 9PL, UK 3 Dermatology Research Centre, Salford Royal Hospital, Manchester Academic Health Science Centre, Northern Care Alliance NHS Foundation Trust, Salford, England, UK 4 NIHR Applied Research Collaboration West (ARC West), University Hospitals Bristol and Weston NHS Foundation Trust, Bristol, England, UK 2 Division of Musculoskeletal and Dermatological Sciences, School of Biological Sciences,, The University of Manchester, Manchester, England, M13 9PL, UK 3 Dermatology Research Centre, Salford Royal Hospital, Manchester Academic Health Science Centre, Northern Care Alliance NHS Foundation Trust, Salford, England, UK 4 NIHR Applied Research Collaboration West (ARC West), University Hospitals Bristol and Weston NHS Foundation Trust, Bristol, England, UK Thomas Parkhouse Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Francesca Spiga Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Lesley E Rhodes Roles: Investigation, Methodology, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Original Draft Preparation, Writing – Review & Editing Sarah Dawson Roles: Investigation, Methodology, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Review & Editing Katie E Webster Roles: Investigation, Methodology, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Review & Editing Deborah M Caldwell Roles: Funding Acquisition, Investigation, Methodology, Supervision, Writing – Review & Editing Roles: Funding Acquisition, Investigation, Methodology, Supervision, Writing – Review & Editing Julian P T Higgins Roles: Funding Acquisition, Investigation, Methodology, Project Administration, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Funding Acquisition, Investigation, Methodology, Project Administration, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing OPEN PEER REVIEW REVIEWER STATUS Current sun safety advice focuses on minimizing exposure to sunlight, due to the relationship between ultraviolet radiation and skin cancer. However, sunlight also has beneficial effects, and there are calls for guidance to reflect these alongside the harmful effects. To examine the net effect of harmful and beneficial aspects, we aimed to determine the association between sunlight exposure and all-cause mortality. Additionally, we examined cause-specific mortality and whether the associations varied according to skin type/colour or ethnicity. We conducted a systematic review, searching MEDLINE, Embase, Web of Science and the Cochrane Central Register of Controlled Trials (Nov 2023) for reports of epidemiological studies in the general population investigating the effect of any measure of long-term sun exposure on all-cause, cardiovascular-related, or cancer-related mortality. We conducted a narrative synthesis of the findings and assessed risk of bias using the ROBINS-E tool. PROSPERO: CRD42023474157. The search identified 73 eligible articles. Methods of measuring sunlight exposure comprised radiation (e.g., ultraviolet radiation levels), proxy measures of radiation (e.g., latitude) and behaviour associated with sunlight exposure (e.g., frequency of sunbathing). The evidence was inconclusive. While most studies of skin cancer mortality found a higher risk associated with more exposure to sunlight, many studies of other cancers reported lower risks associated with more exposure to sunlight. Evidence for all-cause mortality was mixed, as were findings for cardiovascular mortality. Results were subject to high risk of bias, largely due to the likelihood of uncontrolled confounding and the use of indirect measures of sunlight exposure. There were insufficient data regarding any differential effects of sunlight on mortality for those of different skin types/colours or ethnicity. Findings from observational epidemiological studies of the association between sunlight exposure and mortality are too variable to provide a strong rationale for changes to sun protection guidance. The effect of sunlight on death rate: a systematic review of the evidence Sunlight causes damage to our skin. Being sunburnt can increase the chance of getting skin cancer. Therefore, sun safety advice in the UK is mainly focused on avoiding the harmful impacts of sunlight. Organizations providing advice on sun exposure say to stay in the shade between 11 am and 3 pm and wear covering clothing. There are also positive effects of sunlight exposure such as vitamin D production in your body. Vitamin D is important for your bones. Vitamin D or other effects of sunlight may also help reduce your risk of developing some cancers and cardiovascular disease. People of different skin types react to sunlight in different ways. People with darker skin need more exposure to sunlight than people with lighter skin to produce the same amount of vitamin D, and they also have less risk of developing sun-related skin cancer. We gathered all the available studies that have measured both sunlight exposure and mortality (death) rates. We were interested in the overall number of deaths (by any cause), as well as deaths specifically caused by cardiovascular disease or cancer. Our findings are mixed. Exposure to sunlight has been reported both to increase and to decrease your risk of dying. Alongside its harmful effect on skin cancer, sunlight may help prevent other types of cancer. However, there were problems in how the studies were done, so we can’t be certain about the findings. From the information available, there is not strong enough evidence to alter sun exposure advice. Systematic review; Sunlight; Ultraviolet radiation; Mortality; Cancer; Cardiovascular disease Corresponding Author(s) Thomas Parkhouse ([email protected]) Grant information: This project was funded by the National Institute for Health and Care Research (NIHR) under its Evidence Synthesis Programme (grant reference number: NIHR161983). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Copyright: © 2025 Parkhouse T et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. How to cite: Parkhouse T, Spiga F, Rhodes LE et al. The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.13980.1) First published: 18 Jun 2025, 5:51 (https://doi.org/10.3310/nihropenres.13980.1) Latest published: 28 Nov 2025, 5:51 (https://doi.org/10.3310/nihropenres.13980.2) The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There is a newer version of this article available. of this article available. Current sun safety advice is primarily focused on minimizing exposure to sunlight. For example, in the United Kingdom (UK) this includes Cancer Research UK recommendations to keep to the shade between 11 am and 3 pm and to cover up with clothes1, advice that is echoed by the National Health Service (NHS)2 and the National Institute for Health and Care Excellence (NICE)3, and in the USA by the American Academy of Dermatology (AAD) guidance on sunlight exposure4. The reason behind this advice is the well-established relationship between exposure to ultraviolet (UV) radiation (UVR) and skin cancer5. It has been estimated that 86% of melanoma cases are attributable to UVR6, and the risk of developing basal cell carcinoma (BCC), the most common form of non-melanoma skin cancer (NMSC), is 2.12 times greater with every five sunburns experienced as an adult7. Cutaneous squamous cell carcinoma (cSCC), which is capable of metastasizing, is the other common type of NMSC, and shows an increasing incidence in the UK and various European countries8,9. There are calls to alter sun exposure advice10. The harmful effects of sunlight exposure should be considered alongside any beneficial effects. For example, sunlight exposure is usually the body’s main source of vitamin D11, which is linked to various health benefits, such as reducing cancer risk and improving immune system functioning, in addition to its established benefit in musculoskeletal health12–14. Other mechanisms may also contribute to health. Evidence suggests that sunlight converts nitric oxide metabolites, stored in the skin, to nitric oxide15, which may help to reduce blood pressure, amongst other actions that may be beneficial to cardiovascular health16. Furthermore, sun protection advice often fails to consider different skin types fully. People of different skin types react to UVR in different ways, resulting in different needs. A recent review found little evidence of an association between melanoma incidence and UVR in people with skin of colour17, suggesting that sunlight exposure may not be a risk factor for melanoma in those with darker skin, as it is for those with lighter skin18. Given the various risks and benefits associated with sunlight exposure, there is a need to examine the overall effect on population mortality to help people find the right balance between gaining the benefits of sunlight exposure whilst minimizing the risks. We sought to examine the global evidence on the association between sunlight exposure and mortality. Our main aims were to estimate the effect of exposure to sunlight on all-cause mortality, cardiovascular-related mortality and cancer-related mortality. We additionally examined evidence on whether the effects of exposure to sunlight on mortality vary according to skin type/colour or ethnicity. We undertook a systematic review following principles outlined in the Centre for Reviews and Dissemination (CRD) guidance for undertaking reviews in health care19 and the Cochrane Handbook for Systematic Reviews of Interventions20. We formed a public advisory group comprising four individuals with different skin types (three with Afro-Caribbean ancestry and one with very pale skin). We obtained input into the protocol from this group, although their ability to influence it was limited since the project was commissioned. They explained that people with darker skin can be very concerned about the harms of sunlight exposure as well as the possibility of vitamin D deficiency, reinforcing the need to examine data by subgroups defined by skin colour and type. They also made it clear that more refined guidance would be useful for people with all types of skin. They strongly supported the project and advised us that the findings would be of great interest. They further advised us on how we might disseminate the findings to those responsible for national guidelines on sun exposure; they felt that their health providers (e.g. general practitioners) were not sufficiently informed to provide targeted advice. Members of the public helped us write a lay summary of our plans for our website in advance of the project start (https://bristol-esg.org/projects-2/sunlight-exposure/) and a member of the public advisory group helped us write a plain language summary to disseminate the results. Criteria for inclusion of articles in the review were: (i) reporting a primary epidemiological study with a cohort design (including randomized trials of interventions to alter relevant behavioural exposures), a case-control design or an ecological design; (ii) in a general population; (iii) using any measure(s) of long-term sunlight exposure; and (iv) reporting outcome data on all-cause mortality, cancer-specific mortality or cardiovascular disease (CVD) mortality. To perform a comprehensive assessment, we used the exposure term ‘sunlight’ broadly to include measures of radiation (such as sunlight, UVR and UVB); proxies for radiation measures (such as latitude and geographical location); and behaviours associated with sun exposure (such as episodes of sunburn and recreational sun exposure). We excluded: (i) studies of artificial sources of UVR exposure, such as sunbed use and prescribed phototherapy and (ii) studies restricted to people with pre-existing disease (e.g. reporting survival rates in people diagnosed with melanoma). We searched MEDLINE, Embase, Web of Science and the Cochrane Central Register of Controlled Trials to November 2023 using relevant controlled vocabularies, text‐words and search syntax appropriate to each resource. See Appendix S1 in the extended data21 for the full search strategy. Additionally, we performed forwards and backwards citation searches of included articles identified in the search and scanned the reference lists of relevant systematic reviews. We also checked for any relevant retraction statements or errata of included studies. Two reviewers independently assessed all reports for eligibility. Data were extracted using standardized data extraction forms developed in Microsoft Excel, which had first been piloted on a small sample of articles and adapted as necessary. We collected the following data: study design (nested case-control, case-control, cohort, ecological, trial, quasi-experimental), funding sources (public, industry, mixed), study location, sex, age, ethnicity/race, skin type/colour, occupation, inclusion criteria, method/definition of sunlight exposure, period of exposure (childhood, adolescence, adulthood), length of exposure (specific, lifetime) and target condition (all-cause, cancer, CVD). We extracted summary data relating to the association between sun exposure and mortality overall and for CVD and cancer-related causes. This was reported differently across the studies included in this review, and included odds ratios (OR), risk ratios (RR), hazard ratios (HR), regression coefficients, correlation coefficients and narrative reporting. Our main outcomes were death from any cause (all-cause mortality), death due to any cancer (all-cancer mortality) and death due to any cardiovascular cause22 (all-CVD mortality). We also examined death due to skin cancer (melanoma and NMSC) and the five most common causes of cancer-related mortality in the UK according to Cancer Research UK23: breast, prostate, lung, bowel and pancreatic cancer; as well as specific causes of CVD. We assessed risk of bias in results using the Risk Of Bias In Non-randomized Studies of Exposure (ROBINS-E) tool for observational studies24. Risk-of-bias assessments were not carried out on articles that reported results as correlations, mortality rates or narratively because the ROBINS-E tool aims to assess risk of bias in estimates of causal effects. Two reviewers independently assessed all reports considered potentially relevant for inclusion, extracted descriptive data and results of included articles, and assessed the risk of bias in results. Any disagreements were resolved by consensus or discussion with a third reviewer. Due to the diversity in the types and units of measurement of the sunlight exposure measures, we did not consider it appropriate to conduct meta-analyses. Instead, we produced a narrative synthesis of the findings for each mortality outcome and examined any comparisons across people of different skin type/colour or ethnicity. Articles reporting results as a relative risk (e.g., RR, HR; or data sufficient to derive these) were grouped by type of exposure and presented in forest plots, with estimates inverted as appropriate to illustrate findings using consistent direction of effect. Where an article reported only separate risk estimates for males and females, we combined these using fixed-effect meta-analysis to provide a result that additionally controlled for sex/gender. Where results were presented in forms other than relative risks for two or more studies, we listed these in tables. Where such results were reported only separately for males and females these were treated as one analysis result (to be consistent with the combined relative risk results), but were reported as separate results in the tables. We address certainty in the evidence through consideration of each of the five GRADE domains in the discussion. The database searches identified 8501 records, and 5181 records were identified using other methods. After examination of potentially relevant full text articles, 73 articles25–97 met the criteria for inclusion in the review (Figure 1). For a comprehensive summary of the characteristics of these 73 articles, see Table S1 in the extended data21. Where multiple articles had used the same cohort, same exposure type and reported the same outcome, we classed these as ‘overlapping’. In these cases, we chose a main article as the source of data. The selection of data was based on an algorithm that included the type of analysis performed (ratio data preferred to linear regression or correlation coefficients), adjustment within analysis (most relevant confounders controlled for preferred), population (whole population preferred to specific subgroups), follow-up time (longer preferred), number of units of observation (e.g. counties preferred to states) and number of participants. As the majority of the included studies were ecological, we applied the same process to national datasets. We considered articles using national mortality data for the same country and date range to be using the same cohort, and used the same selection algorithm to select a main article in such cases. Table S2 in the extended data21 provides a review of all overlapping studies. After the selection of main articles, this resulted in 55 articles being included in the narrative synthesis. All subsequent analysis is conducted on these 55 articles. Eight of the articles reported data on all-cause mortality, eight on all-CVD mortality, and 17 on all-cancer mortality. Eleven described cohort studies, three case-control studies and 41 ecological studies. Twenty-four (44%) of the articles were conducted in the USA, five were conducted worldwide (including between 34 and 172 countries), four in China, and three each in Japan, Spain and Sweden. Other articles were conducted in the USA and Canada, the UK, and Switzerland (all n = 2); as well as Australia, Chile, Europe-wide, France, Italy, New Zealand and Turkey (all n = 1). We provide descriptions of the exposures in Table S3 in the extended data21. The majority of the articles (n = 42, 76%) measured a single exposure; ten (18%) measured two exposures, two (4%) measured three exposures and one measured four exposures. Thus, we reported details of 72 exposures in total. Of those, 40 were measures of radiation, 17 were proxy measures of radiation and 15 were measures of behaviours associated with sunlight exposure. Radiation measures included solar radiation, UVR, UVA and UVB, DNA- and erythemal-weighted UVB and UVR, UV Index, solar incidence, insolation, irradiation, sunlight hours and sunshine. Among the proxy measures for radiation, 16 were latitude (e.g. residential at county or state level) and one was geographic location of deployment during war (tropical vs European). Measures of sunlight exposure-related behaviour included four of occupational exposure, four of recreational exposure and one of both occupational and recreational exposure combined; three used skin damage, two used NMSC mortality rates and one used melanoma mortality rates. Of the 55 included main articles, 34 (62%) reported information on participants’ skin type/colour or ethnicity. Most were conducted in only White participants, whilst two only contributed data for Black participants. In six articles, the authors reported that skin type/colour or ethnicity was mixed. Of these, four included Black and White participants (although, where proportions were reported, the populations were predominately White). One included Black, White and Hispanic participants, and one included Black, White, Hispanic, Asian and Native American participants. Below we provide our findings for our primary outcomes: all-cause, all-CVD and all-cancer mortality, both overall and by type of exposure. Subsequently, we provide a brief summary of the findings for our secondary, cause-specific mortality outcomes. We report the full risk-of-bias assessments, including justifications for judgments, in Table S4 in the extended data21. Overview. In total, eight articles investigated the effect of sunlight on all-cause mortality. Six used a cohort design and two had an ecological design. One article reported findings for two exposures (nine analyses reported across the eight articles). The findings were mixed (see Figure 2 for reported relative risks). Five analysis results52,77,89,97 were in the direction of a beneficial association between sunlight and all-cause mortality. Four analysis results47,67,76,83 were in the direction of a harmful effect of sunlight. Radiation. Three articles looked at the effect of radiation on all-cause mortality (Figure 2). The results of all three articles were considered to be at high risk of bias. Goggins et al.52 conducted an ecological study of the Hong Kong population showing that a 10 W/m2 increase in solar radiation was associated with a 10% decrease in all-cause mortality (RR = 0.90, 95% confidence interval (CI) 0.85 to 0.94). Lin et al.76 measured the residential address of cohort participants across six states and two metropolitan areas in the USA to determine average July erythemal UVR, defined as biological damage per square meter (BD/m2). This measure was subsequently split into quartiles. When comparing the second quartile with the first, there was no evidence of a difference in all-cause mortality (RR = 1.00, 95% CI 0.98 to 1.03; Figure 3). However, when comparing the third and fourth quartiles with the first there were 8% and 6% increases in all-cause mortality, respectively. In addition to the articles included in Figure 2, Fu and Wang47 examined the relationship between the average hours of daily sunshine in China and all-cause mortality using regression. They found that an increase of 0.1 to 0.2 in the average daily sunshine duration rate (equivalent to 2.9 hours of additional daily sunshine) was associated with an increase in all-cause mortality rate the following year (β = 11.509, 95% CI 1.87 to 21.15; high risk of bias). Proxy for radiation. Two articles looked at the effect of proxy radiation exposures on all-cause mortality (Figure 2). The results of both articles were considered to be at high risk of bias. Stevenson et al.89 examined the effect of latitude in the UK. They found that more southerly latitudes were associated with lower all-cause mortality, when compared with residences 300km further north (HR = 0.94, 95% CI 0.92 to 0.96). Page et al.83 examined a cohort of WWII veterans who were deployed to either Pacific or European battlefronts, arguing that those who were deployed to the Pacific area would have experienced higher levels of sun exposure than those in Europe. They observed a small increase in risk associated with being deployed in the Pacific compared with Europe, however the wide confidence intervals were compatible with both a benefit and harm (odds ratio = 1.03, 95% CI 0.89 to 1.19). Behavioural. Three articles examined the effect of sunlight exposure behaviour (Figure 2). He et al.67 looked at the effect of physician-assessed actinic skin damage. The findings suggested that greater sun exposure, as indicated by skin damage, is associated with an increase in the risk of mortality. When comparing those with severe skin damage with those who were assessed to have no damage, there was a 45% increase in risk (HR = 1.45, 95% CI 1.22 to 1.72; high risk of bias). The association does not appear to be driven by smoking (an important common cause of skin damage and CVD), which was controlled for in the analysis. Lindqvist et al.77 measured self-reported sun exposure in a cohort of Swedish women. The evidence suggested that there may be a decrease in mortality with increased sun exposure. Those who reported the highest level of sun seeking behaviour had a 38% decreased risk of dying from any cause compared with those who reported the lowest level (HR = 0.62, 95% CI 0.50 to 0.80; high risk of bias). This finding is supported by Yang et al.97 in another cohort of Swedish women, though their results were considered to be at very high risk of bias. Those who reported spending one week or more annually on sunbathing vacations had a 30% decreased risk of mortality compared with those who never went on sunbathing vacations (HR = 0.70, 95% CI 0.60 to 0.90). They also reported an association between higher number of sunburns experienced and lower all-cause mortality, however the wide confidence intervals were compatible with both a benefit and harm (HR = 0.90, 95% CI 0.70 to 1.20). Overview. In total, eight articles investigated the effect of sunlight on overall CVD mortality, with one reporting findings for two exposure types (thirteen analyses reported across the eight articles). Six of the articles had a cohort design and two were ecological. There were mixed findings (Figure 3). Six analysis results26,52,78,89,97 suggested that higher levels of sunlight are associated with lower risk of CVD mortality. In contrast, five analyses26,67 suggested that higher levels of sunlight are associated with a higher risk of CVD mortality. Two analyses40,76 produced mixed findings. Radiation. Three articles looked at the effect of radiation, all were considered to be at high risk of bias (Figure 3). Al-Hamdan et al.26 performed a national ecological study in the USA. They found that a 100 kJ/m2 increase in solar radiation was associated with a 1% higher risk of dying from CVD for White participants, a 3% higher risk for Black, Hispanic and Asian participants, but a 3% lower risk for Native American participants. Goggins et al.’s52 Hong Kong-based study found that an increase in solar radiation of 10 W/m2 was associated with a lower risk of dying from CVD (RR = 0.87, 95% CI 0.78 to 0.97). In Lin et al.’s76 dose response analysis of average July residential erythemal UVR exposure (BD/m2), there were mixed findings. Risk of CVD mortality increased with dose. However, across all levels of comparison, the confidence intervals were compatible with both a benefit and harm, or with a null effect. Proxy for radiation. Stevenson et al.89 found evidence to suggest a beneficial effect of sunlight exposure, measured via latitude in the UK. Compared with northerly latitudes, residences 300 km further south were associated with a 9% lower risk of CVD mortality (HR = 0.91, 95% CI 0.86 to 0.95; high risk of bias). Behavioural. Four articles examined the effect of sunlight exposure behaviour (Figure 3). Lindqvist’s et al.78 Swedish cohort study found that those who had self-reported high recreational sun exposure were less at risk of dying from CVD than those who reported no exposure (HR = 0.45, 95% CI 0.31 to 0.67; high risk of bias). Yang et al.97 found that people who reported spending one or more weeks a year on sunbathing vacations were half as likely to die from CVD than those who never spent time on sunbathing vacations (HR = 0.50, 95% CI 0.30 to 0.80). Additionally, there was an association between higher frequency of sunburning and lower CVD mortality, however the wide confidence intervals indicate uncertainty in this finding (HR = 0.30, 95% CI 0.10 to 1.10). Furthermore, the results reported in this article were considered to be at very high risk of bias. In contrast, the findings in He et al.67 suggested a harmful effect of sunlight. They found that greater physician-assessed actinic skin damage was associated with a higher risk of CVD mortality. Those whose skin damage was considered severe had a 64% increased risk of CVD mortality compared with those with no actinic skin damage (HR = 1.64, 95% CI 1.29 to 2.10; high risk of bias). As shown for all-cause mortality, the association does not appear to be driven by smoking, which was controlled for in the analysis. Donneyong et al.40 looked at the self-reported frequency of outdoor recreational activity. The results were mixed, with wide confidence intervals that were compatible with both a benefit and harm, and considered to be at high risk of bias. Overview. In total, 17 articles looked at the effect of sunlight exposure on all-cancer mortality, with some reporting multiple exposure types, outcomes and/or date ranges (22 analyses reported across the 17 articles). Twelve of the articles had an ecological design and five had a cohort design. The majority of analysis results (20 analyses) were in the direction of a beneficial association between sunlight and all-cancer mortality. However, two analyses67,76 suggested there may be a harmful effect of sunlight. See Figure 4 for reported relative risks and Table 1 for results reported in other formats. | Study | Location | Exposure | Unit of analysis | Specific Outcome | Subgroup | Analysis | Results | Direction of effect | Risk of bias | |---|---|---|---|---|---|---|---|---|---| | Apperly28 | USA and Canada | Radiation (solar radiation) | Solar radiation Index | All-cancer | n/a | Correlation | r = −0.63 | Benefit | n/a | | Behavioural (occupational exposure to sunlight) | % of farmers per state population | All-cancer | n/a | Correlation | r = −0.68 | Benefit | n/a | || | Camara and Brandao34 | Worldwide | Radiation (solar incidence) | kWh/m2 / day | All-cancer | n/a | Mortality rate | “92.48/100,000 in countries with high sunlight incidence 124.85/100,000 in countries with low sunlight incidence” (p < 0.05) | Benefit | n/a | | Ezzati et al.42 | USA | Radiation (insolation) | Annual average solar radiation | All-cancer (smoking- related) | Males | Regression | β = −0.00029 (95% CI −0.00054 to −0.000031) | Benefit | Some concerns | | Females | Regression | β = −0.00033 (95% CI −0.00051 to −0.00015) | Benefit | Some concerns | ||||| | All-cancer (non- smoking related | Males | Regression | β = −0.00032 (95% CI −0.00057 to −0.000064) | Benefit | Some concerns | |||| | Females | Regression | β = −0.00079 (95% CI −0.001 to −0.00055) | Benefit | Some concerns | ||||| | Grant and Garland56 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | All- cancer** | White males | Regression | β = −0.65 (p < 0.001) | Benefit | Some concerns | | White females | Regression | β = −0.85 (p < 0.001) | Benefit | Some concerns | ||||| | Grant and Garland56 (1970–1994) | USA | Radiation (UVB) | kJ/m2 | All- cancer** | White males | Regression | β = −0.54 (p < 0.001) | Benefit | Some concerns | | White females | Regression | β = −0.82 (p < 0.001) | Benefit | Some concerns | ||||| | Grant57 | USA | Radiation (UVB) | kJ/m2 | All- cancer** | Black males | Regression | β = −0.34 (p = 0.01) | Benefit | Very high | | Black females | Regression | β = −0.50 (p < 0.001) | Benefit | Very high | ||||| | Grant59 | China | Proxy for radiation (latitude) | Degree of latitude* | All-cancer | Females | Regression | β = 0.75 (p < 0.001) | Benefit | High | | Grant62 | France | Proxy for radiation (latitude) | Degree of latitude squared* | All-cancer | Male | Correlation | r = 0.8 (p < 0.001) | Benefit | High | | Female | Correlation | r = 0.78 (p < 0.001) | Benefit | High | ||||| | Grant64 | USA (California) | Proxy for radiation (latitude) | Degree of latitude* | All-cancer | n/a | Regression | β = 0.47 (p = 0.009) | Benefit | High | | Behavioural (NMSC mortality) | Mortality rate | All-cancer | n/a | Regression | β = −0.69 (p < 0.001) | Benefit | High | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. **not including lung cancer. Abbreviations: b: regression beta coefficient; CI: confidence interval; n/a: not applicable; NMSC: non-melanoma skin cancer; r: correlation coefficient; RoB: risk of bias; UVB: ultraviolet B radiation. Radiation. Ten articles looked at the effect of radiation on all-cancer mortality. The majority (n = 9) found a potentially beneficial effect of sunlight, with higher levels of radiation associated with lower levels of cancer mortality (Figure 4 and Table 1). In Altug and Kilçiksiz27, a national ecological study in Turkey, the results indicated that a one (unspecified) unit increase in sunlight duration was associated with a 56% reduction in the risk of cancer mortality (RR = 0.44, 95% CI 0.32 to 0.60; high risk of bias). Chen et al.35 performed a national ecological study in China, looking at the effect of UVB exposure on cancer mortality. The findings suggested that a 10 mW/m2 increase in average daily UVB irradiance was associated with a 4% lower risk of mortality (rate ratio = 0.96, 95% CI 0.95 to 0.97; high risk of bias). Similarly, Fukuda et al.’s48 ecological study in Japan found that, per MJ/m2 of solar radiation, there was a 1.3% reduction in cancer mortality risk (RR = 0.99, 95% CI 0.98 to 0.99; some concerns over risk of bias). The findings in Grant and Garland56 and Grant57 suggest a similar association. Higher levels of UVB were found to be associated with lower cancer mortality rates across both Black and White males and females between 1970 and 1994 (some concerns over risk of bias). A relationship between higher levels of radiation and lower cancer mortality was also observed in Apperly28, Camara and Brandao34, Ezzati et al.42 (some concerns over risk of bias) and Grant and Garland56 (1950–1969; some concerns over risk of bias). Goggins et al.52 found an association between higher solar radiation and lower cancer mortality, however the confidence intervals were compatible with a null effect (RR = 0.93, 95% CI 0.85 to 1; high risk of bias). In contrast, Lin et al.76 found that increasing levels of residential erythemal UVR (BD/m2) were associated with an increased risk of cancer mortality. Comparing the fourth with the first quartile showed a 6% increase in the risk of mortality (HR = 1.06, 95% CI 1.02 to 1.11; some concerns over risk of bias). Proxy for radiation. Four articles looked at the effect of proxy radiation exposures on cancer mortality, with all finding a potentially beneficial effect of sunlight (Figure 4 and Table 1). The findings of all four of the articles were assessed to be at high risk of bias. Stevenson et al.89 reported that, compared with UK northerly latitudes, residences 300 km further south had a 7% reduction in mortality risk (HR = 0.93, 95% CI 0.90 to 0.96). Grant59 conducted a national ecological study in China. They found an association between higher latitudes (i.e., lower levels of sunlight) and higher cancer mortality (β = 0.75, p < 0.001). A similar relationship was found by Grant in France62 and California64. Behavioural. Five articles studied the effect of sunlight exposure behaviour on cancer mortality (Figure 4 and Table 1). He et al.67 found that those with greater physician-assessed actinic skin damage had a greater risk of cancer mortality. Participants whose skin damage was considered severe had a 78% higher risk of mortality compared with those with no actinic skin damage (HR = 1.78, 95% CI 1.20 to 2.63; high risk of bias). In contrast, Grant’s64 ecological study in California, using population-level NMSC mortality rates as a proxy for sunlight exposure, found a relationship between higher NMSC mortality rates and lower overall cancer mortality (β = −0.69, p < 0.001; high risk of bias). Apperly28 looked at the proportion of the population engaged in agricultural work and reported a relationship between higher proportions of agricultural workers and lower cancer mortality. However, no margins of error were reported with this estimate. Two articles, Lindqvist et al.78 and Yang et al.97, found that recreational sunbathing behaviour was associated with lower cancer mortality. However, in both cases there were wide confidence intervals which were compatible with both a beneficial and harmful effect of sunlight exposure (Figure 4). Furthermore, they were considered to be at high and very high risk of bias, respectively. Skin cancers. In total, 23 articles investigated the effect of sunlight on skin cancer mortality with some reporting multiple exposure types, outcomes or date ranges (46 analyses reported across the 23 articles; 28 analyses measuring effect on melanoma, 16 measuring effect on NMSC, and two measuring effect on both combined). Seventeen of the articles were ecological, three had a cohort design and three used a case-control design. There were 25 analyses looking at the effect of radiation, 16 on the effect of proxy radiation exposures and five looked at the effect of sunlight exposure behaviour. Most analysis results (n=34) suggested that higher levels of sunlight were associated with a higher risk of both melanoma and NMSC mortality (Figure 5 and Table 2). However, the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma39 and NMSC mortality25; two suggested an association between higher levels of UVB and lower melanoma mortality51,90; one reported a weak association between higher annual insolation and lower melanoma mortality25 and one reported an association between higher NMSC mortality rates and lower melanoma mortality58. The remaining six analyses either produced mixed results45,58,96, had no direction reported44, showed little to no effect72 or had very wide confidence intervals that were compatible with both a benefit and harm83. | Study | Location | Exposure | Unit of analysis | Specific Outcome | Subgroup | Analysis | Results | Direction of effect | Risk of bias | |---|---|---|---|---|---|---|---|---|---| | Alcalá Ramírez del Puerto et al.25 | Spain | Radiation (irradiation) | NR | Melanoma | n/a | Correlation | r = 0.16 (p > 0.05) | Harm | n/a | | NMSC | n/a | Correlation | r = 0.40 (p > 0.05) | Harm | n/a | |||| | Radiation (insolation) | NR | Melanoma | n/a | Correlation | r = −0.113 (p > 0.05) | Benefit | n/a | || | NMSC | n/a | Correlation | r = 0.266 (p > 0.05) | Harm | n/a | |||| | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | n/a | Correlation | r = 0.22 (p > 0.05) | Harm | n/a | || | NMSC | n/a | Correlation | r = −0.40 (p < 0.01) | Benefit | n/a | |||| | Elwood et al.41 | USA & Canada | Radiation (Epidemiological index of UVR) | NR | Melanoma | Males | Regression | β = 0.039 (p < 0.001) | Harm | High | | Females | Regression | β = 0.022 (p < 0.001) | Harm | High | ||||| | NMSC | Males | Regression | β = 0.045 (p < 0.001) | Harm | High | |||| | Females | Regression | β = 0.017 (p < 0.001) | Harm | High | ||||| | Proxy for radiation (latitude) | Degree of latitude* | All skin cancer | Males | Regression | β = −0.12 (SE = 0.011) | Harm | High | || | Females | Regression | β = −0.055 (SE = 0.009) | Harm | High | ||||| | Fleischer and Fleischer44 | USA | Radiation (solar radiation) | kJ/m2 | Melanoma | n/a | Narrative | “No associations were demonstrated between solar energy and cancer mortality for melanoma of the skin (p = 0.60)” | NR | n/a | | Garland et al.51 | Worldwide | Radiation (UVA) | 1 photon flux per nanometer | Melanoma | Males | Regression | β = 0.00012 (SE = 0.000050) | Harm | High | | Females | Regression | β = 0.00004 (SE = 0.00003) | Harm | High | ||||| | Radiation (UVB) | 1 photon flux per nanometer | Melanoma | Males | Regression | β = −0.00146 (SE = 0.00094) | Benefit | High | || | Females | Regression | β = −0.00106 (SE = 0.00053) | Benefit | High | ||||| | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | Males | Correlation | r = −0.08 (p > 0.05) | Mixed (gender) | n/a | | Females | Correlation | r = 0.28 (p > 0.05) | Mixed (gender) | n/a | ||||| | NMSC | Males | Correlation | r = −0.5 (p < 0.01) | Harm | n/a | |||| | Females | Correlation | r = −0.33 (p 0.05) | Benefit | n/a | || | Females | Correlation | r = −0.43 (p < 0.01) | Benefit | n/a | ||||| | Grant61 (1950–1969) | USA | Proxy for radiation (latitude) | Degree of latitude* | NMSC | White males | Regression | β = −0.66 (p < 0.001) | Harm | High | | White females | Regression | β = −0.41 (p = 0.002) | Harm | High | ||||| | Grant61 (1970–1994) | USA | Proxy for radiation (latitude) | Degree of latitude* | NMSC | White males | Regression | β = −0.37 (p = 0.001) | Harm | High | | Grant63 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | Melanoma | White males | Regression | β = 0.47 (p < 0.001) | Harm | High | | White females | Regression | β = 0.46 (p < 0.001) | Harm | High | ||||| | NMSC | White males | Regression | β = 0.34 (p < 0.001) | Harm | High | |||| | White females | Regression | β = 0.21 (p < 0.001) | Harm | High | ||||| | Grant64 | USA (California) | Behavioural (NMSC mortality) | Mortality rate | Melanoma | White males | Regression | β = 0.46 (p = 0.03) | Harm | High | | Lee74 (1950–1959) | USA | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | White males | Regression | β = −0.039 (95% CI −0.050 to −0.028) | Harm | High | | White females | Regression | β = −0.037 (95% CI −0.053 to −0.021) | Harm | High | ||||| | Lee74 (1960–1969) | USA | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | White males | Regression | β = −0.037 (95% CI −0.047 to −0.028) | Harm | High | | White females | Regression | β = −0.030 (95% CI −0.039 to −0.021) | Harm | High | ||||| | Lee74 (1970–1979) | USA | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | White males | Regression | β = −0.026 (95% CI −0.036 to −0.016) | Harm | High | | White females | Regression | β = −0.023 (95% CI −0.032 to −0.013) | Harm | High | ||||| | Lee74 (1988–1992) | USA | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | White males | Regression | β = −0.014 (95% CI −0.023 to −0.005) | Harm | High | | White females | Regression | β = −0.009 (95% CI −0.019 to −0.002) | Harm | High | ||||| | Rivas et al.85 | Chile | Proxy for radiation (latitude) | Degree of latitude* | Melanoma | n/a | Correlation | r = −0.88 | Harm | n/a | | NMSC | n/a | Correlation | r = −0.53 | Harm | n/a | |||| | Takahashi et al.90 | Japan | Radiation (UVB) | 10 J/m2 /year | Melanoma | Males | Correlation | r = −0.155 (p = 0.35) | Benefit | n/a | | Females | Correlation | r = −0.09 (p = 0.58) | Benefit | n/a | ||||| | NMSC | Males | Correlation | r = 0.268 (p = 0.09) | Harm | n/a | |||| | Females | Correlation | r = 0.282 (p = 0.08) | Harm | n/a | ||||| | Wu and Weinstock96 | USA | Radiation (UV index) | UV index zone 2 vs UV index zone 1 | Keratinocyte carcinoma | White males | Narrative | “White male KC mortality rate was found to be higher in sun Zone 2 (p = 0.004)” | Mixed (gender) | n/a | | White females | Narrative | “There was no statistical difference between sun zones for White women (p = 0.379)” | Mixed (gender) | n/a | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; CI: confidence interval; n/a: not applicable; KC: keratinocyte carcinoma; NMSC: non-melanoma skin cancer; NR: not reported; r: correlation coefficient; RoB: risk of bias; SE: standard error; UV: ultraviolet radiation. UVA: ultraviolet A radiation; UVB: ultraviolet B radiation. Breast cancer mortality. In total, 11 articles looked at the effect of sunlight on breast cancer mortality, with some reporting multiple exposure types and/or date ranges (17 analyses reported, across the 11 articles). Nine articles had an ecological design and one each used cohort and case-control designs. There were ten analysis results examining the effect of radiation, four looked at proxy measures of radiation and three looked at behavioural measures of exposure. Overall, the evidence suggested that sunlight may provide a protective effect (Figure 6 and Table 3). Thirteen analysis results from Boscoe and Schymura30, Chen et al.35, Freedman et al.46, Grant and Garland56, and Grant53,57,58,62 indicated that higher levels of sunlight were associated with reduced breast cancer mortality. Three analyses from Lin et al.76, Fukuda et al.48, and Grant58 suggested that there may be a harmful association between sunlight and breast cancer mortality, and one from Fleischer and Fleischer44 was reported without a direction of effect. | Study | Location | Exposure | Unit of analysis | Specific Outcome | Subgroup | Analysis | Results | Direction of effect | Risk of bias | |---|---|---|---|---|---|---|---|---|---| | Fleischer and Fleischer44 | USA | Radiation(solar radiation) | kJ/m2 | Breast cancer | n/a | Narrative | “No associations were demonstrated between solar energy and cancer mortality for breast cancer (p = 0.40)” | Not reported | n/a | | Grant53 | Worldwide | Proxy for radiation (latitude) | Degree of latitude* | Breast cancer | n/a | Correlation | r = 0.66 (p < 0.001) | Benefit | n/a | | Grant and Garland56 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | Breast cancer | White females | Regression | β = −0.59 (p < 0.001) | Benefit | Some concerns | | Grant and Garland56 (1970–1994) | USA | Radiation (UVB) | kJ/m2 | Breast cancer | White males | Regression | β = −0.71 (p = 0.006) | Benefit | Some concerns | | White females | Regression | β = −0.71 (p < 0.001) | Benefit | Some concerns | ||||| | Grant57 | USA | Radiation (UVB DNA) | kJ/m2 | Breast cancer | Black females | Regression | Β = −0.38 (p = 0.006) | Benefit | Very high | | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Breast cancer | Females | Correlation | r = 0.15 (p > 0.05) | Benefit | n/a | | Behavioural (NMSC mortality and melanoma mortality) | NMSC mortality rate | Breast cancer | Females | Correlation | r = −0.38 (p < 0.01) | Benefit | n/a | || | Melanoma mortality rate | Breast cancer | Females | Correlation | r = 0.30 (p < 0.05) | Harm | n/a | ||| | Grant62 (1992) | France | Proxy for radiation (latitude) | Degree of latitude* | Breast cancer | Females | Correlation | r = 0.69 (p = 0.001) | Benefit | n/a | | Grant62 (1998–2000) | France | Proxy for radiation (latitude) | Degree of latitude squared* | Breast cancer | Females | Correlation | r = 0.66 (p = 0.001) | Benefit | n/a | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; CI: confidence interval; n/a: not applicable; NMSC: non-melanoma skin cancer; r: correlation coefficient; RoB: risk of bias; UVB: ultraviolet B radiation. Prostate cancer mortality. There were 15 articles looking at the effect of sunlight on prostate cancer mortality, with some reporting multiple exposure types and/or date ranges (25 analyses reported across the 15 articles). There were 12 articles with an ecological design, two used a cohort and one used a case-control design. The findings were mixed (Figure 7 and Table 4). Fourteen analysis results looked at the effect of radiation, five looked at proxy radiation measures and six examined the effect of sunlight exposure behaviour. Most results suggested there may be a beneficial effect of sunlight on prostate cancer mortality (14 analyses). However, six results from Lin et al.76, Grant and Garland56, Freedman et al.46, John et al.69 and Grant58, indicated that sunlight may have a harmful effect on prostate cancer mortality. Three results from Grant58, Grant and Garland56 and Mizoue80, found little evidence of an effect; one result from Colli and Grant37 produced very wide confidence intervals compatible with both a benefit and harm and one article44 did not report the direction of effect. | Study | Location | Exposure | Unit of analysis | Outcome | Subgroup | Analysis | Results | Direction of effect | Risk of bias | |---|---|---|---|---|---|---|---|---|---| | Colli and Colli36 | Worldwide | Radiation (UV index) | UV index unit | Prostate cancer | Males | Regression | β = −0.81 (95% CI −1.292 to −0.322) | Benefit | High | | Proxy for radiation (latitude) | Degree of latitude* | Prostate cancer | Males | Regression | β = 0.14 (95% CI 0.044 to 0.24) | Benefit | High | || | Colli and Grant37 | USA | Radiation (UV index) | UV index unit | Prostate cancer | Black males | Regression | β = 0.2 (95% CI −2.4 to 2.8) | Harm | High | | Fleischer and Fleischer44 | USA | Radiation (solar radiation) | kJ/m2 | Prostate cancer | n/a | Narrative | “No associations were demonstrated between solar energy and cancer mortality for prostate cancer (p = 0.90)” | NR | n/a | | Grant55 | Worldwide | Radiation (UVB) | MJ/m2 / year | Prostate cancer | Predominantly Caucasian populations | Narrative | Inverse relationship between UV and mortality t = -5.8, p < 0.001 | Benefit | n/a | | Grant and Garland56 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | Prostate cancer | White males | Regression | β = 0.02 (p = 0.94) | Harm | Some concerns | | Proxy for radiation (latitude) | Degree of latitude* | Prostate cancer | White males | Regression | β = 0.52 (p = 0.09) | Benefit | High | || | Grant and Garland56 (1970–1994) | USA | Radiation (UVB) | kJ/m2 | Prostate cancer | White males | Regression | β = 0.38 (p = 0.04) | Harm | Some concerns | | Proxy for radiation (latitude) | Degree of latitude* | Prostate cancer | White males | Regression | β = 0.27 (p = 0.12) | Benefit | High | || | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Prostate cancer | Males | Correlation | r = 0.06 (p > 0.05) | Benefit | n/a | | Behavioural (NMSC mortality; melanoma mortality) | NMSC mortality rate | Prostate cancer | Males | Correlation | r = −0.21 (p > 0.05) | Benefit | n/a | || | Melanoma mortality rate | Prostate cancer | Males | Correlation | r = 0.52 (p < 0.01) | Harm | n/a | ||| | Grant62 | France | Proxy for radiation (latitude) | Degree of latitude squared* | Prostate cancer | Males | Correlation | r = 0.68 (p = 0.001) | Benefit | n/a | | Grant64 | USA (California) | Behavioural (NMSC mortality) | NMSC mortality rate | Prostate cancer | White males | Regression | β = −0.62 (p = 0.005) | Benefit | High | | Mizoue80 | Japan | Radiation (solar radiation) | KWh/ Hour/day | Prostate cancer | Males | Correlation | r = −0.07 (p > 0.05) | Benefit | n/a | | Santos Arrontes et al.86 | Spain | Radiation (sunlight) | Hours of sun exposure/year | Prostate cancer | Males | Narrative | “Mortality from prostate cancer presented statistically significant differences, being . . . lower in the areas with the greatest number of hours of sunshine per year (p = 0.041). | Benefit | n/a | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; CI: confidence interval; n/a: not applicable; NMSC: non-melanoma skin cancer; NR: not reported r: correlation coefficient; RoB: risk of bias; UV: ultraviolet radiation; UVB: ultraviolet B radiation. Lung cancer mortality. There were 10 articles investigating the effect of sunlight on lung cancer mortality, with some reporting multiple exposure types (12 analyses reported across the 10 articles). Nine articles had an ecological design and one had a cohort design. There were six analyses looking at the effect of radiation, three looked at proxy radiation measures and three looked at sunlight exposure behaviour. The findings were mixed (Figure 8 and Table 5). The majority of analyses (n = 7), reported in Chen et al.35, Fukuda et al.48, Fleischer and Fleischer44, Grant57,59,64, and Garland et al.50 found that higher levels of sunlight were associated with a decreased risk of lung cancer mortality. However, three analyses in Lin et al.76 and Grant58 suggested that there may be a harmful effect of sunlight. The findings of two analyses from two articles58,62 were mixed, with different findings reported for males and females. | Study | Location | Exposure | Unit of analysis | Outcome | Subgroup | Analysis | Results | Direction of effect | RoB | |---|---|---|---|---|---|---|---|---|---| | Fleischer and Fleischer44 | USA | Radiation (solar radiation) | kJ/m2 | Lung cancer | n/a | Narrative | “Associations were demonstrated between increasing solar energy and decreasing cancer incidence for lung cancer (p < 0.001)” | Benefit | n/a | | Garland et al.50 | USA | Radiation (solar radiation) | Calories/cm² | Lung, trachea and bronchus cancer | Females | Correlation | r = −0.19 (p = 0.28) | Benefit | n/a | | Grant57 | USA | Radiation (UVB DNA) | kJ/m2 | Lung cancer | Black males | Regression | β = −0.52 (p = 0.003) | Benefit | Very high | | Black females | Regression | β = −0.29 (p = 0.08) | Benefit | Very high | ||||| | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Lung cancer | Males | Correlation | r = −0.36 (p 0.05) | Harm | n/a | ||||| | Behavioural (NMSC mortality; melanoma mortality) | NMSC mortality rate | Lung cancer | Males | Correlation | r = 0.02 (p > 0.05) | Mixed (gender) | n/a | || | Females | Correlation | r = −0.31 (p < 0.05) | Mixed (gender) | n/a | ||||| | Melanoma mortality rate | Lung cancer | Males | Correlation | r = 0.33 (p < 0.05) | Harm | n/a | ||| | Females | Correlation | r = 0.36 (p < 0.05) | Harm | n/a | ||||| | Grant59 | China | Proxy for radiation (latitude) | Degree of latitude* | Lung cancer | 35–64 year old males | Regression | β = 0.43 (p = 0.002) | Benefit | High | | Grant62 | France | Proxy for radiation (latitude) | Degree of latitude squared* | Lung cancer | Males | Correlation | r = 0.54 (p = 0.01) | Benefit | n/a | | Females | Narrative | Not significant | Mixed (gender) | n/a | ||||| | Grant64 | USA (California) | Behavioural (NMSC mortality) | NMSC mortality rate | Lung cancer | White males | Regression | β = −0.38 (p = 0.11) | Mixed (gender) | High | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; n/a: not applicable; NMSC: non-melanoma skin cancer; NR: not reported r: correlation coefficient; RoB: risk of bias; UVB DNA: DNA-weighted UVB radiation. Bowel cancer mortality. There were 14 articles measuring the effect of sunlight on bowel cancer mortality, with some reporting multiple exposure types, outcomes and/or date ranges (31 analyses reported across the 14 articles). There were 11 articles with an ecological design, two used a cohort and one had case-control design. Sixteen analysis results examined radiation measures, eight looked at proxy radiation measures and seven looked at the effect of exposure behaviour measures. Overall, the majority of analysis results (n = 22) suggested that higher levels of sunlight were associated with a decreased risk of bowel cancer mortality (Figure 9 and Table 6). However, four analyses from Grant58 and Veach et al.94 indicated that there may be a harmful association between sunlight and mortality. Three analyses from Grant58, Lin et al.76 and Freedman et al.46 produced mixed findings, with conflicting results found either across dose levels or between gender. The result in Page et al.83 produced very wide confidence intervals that were compatible with both a benefit and a harm, whilst in Fleischer and Fleischer44 the direction of effect was not reported. | Study | Location | Exposure | Unit of analysis | Outcome | Subgroup | Analysis | Results | Direction of effect | RoB | |---|---|---|---|---|---|---|---|---|---| | Fleischer and Fleischer44 | USA | Radiation (solar radiation) | kJ/m2 | Colon/rectum cancer | n/a | Narrative | “No associations were demonstrated between solar energy and cancer mortality for colon/ rectum cancer (p = 0.12)” | NR | n/a | | Grant and Garland56 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | Colon cancer | White males | Regression | β = −0.63 (p < 0.001) | Benefit | Some concerns | | White females | Regression | β = −0.70 (p < 0.001) | Benefit | Some concerns | ||||| | Rectum cancer | White males | Regression | β = −0.62 (p < 0.001) | Benefit | Some concerns | |||| | White females | Regression | β = −0.65 (p < 0.001) | Benefit | Some concerns | ||||| | Grant and Garland56 (1970–1994) | USA | Radiation (UVB) | kJ/m2 | Colon cancer | White males | Regression | β = −0.71 (p < 0.001) | Benefit | Some concerns | | White females | Regression | β = −0.76 (p < 0.001) | Benefit | Some concerns | ||||| | Rectum cancer | White males | Regression | β = −0.75 (p < 0.001) | Benefit | Some concerns | |||| | White females | Regression | β = −0.70 (p < 0.001) | Benefit | Some concerns | ||||| | Grant57 | USA | Radiation (UVB DNA) | kJ/m2 | Colon cancer | Black males | Regression | β = −0.37 (p = 0.03) | Benefit | Very high | | Black females | Regression | β = −0.05 (p = 0.76) | Benefit | Very high | ||||| | Rectum cancer | Black males | Regression | β = −0.38 (p = 0.02) | Benefit | Very high | |||| | Black females | Regression | β = −0.02 (p = 0.94) | Benefit | Very high | ||||| | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Colon cancer | Males | Correlation | r = 0.19 (p > 0.05) | Mixed (gender) | n/a | | Females | Correlation | r = −0.006 (p > 0.05) | Mixed (gender) | n/a | ||||| | Rectum cancer | Males | Correlation | r = 0.61 (p < 0.01) | Benefit | n/a | |||| | Females | Correlation | r = 0.33 (p < 0.05) | Benefit | n/a | ||||| | Behavioural (NMSC mortality; melanoma mortality) | NMSC mortality rate | Colon cancer | Males | Correlation | r = −0.30 (p < 0.05) | Benefit | n/a | || | Females | Correlation | r = −0.31 (p < 0.05) | Benefit | n/a | ||||| | Rectum cancer | Males | Correlation | r = −0.46 (p < 0.01) | Benefit | n/a | |||| | Females | Correlation | r = −0.35 (p < 0.05) | Benefit | n/a | ||||| | Melanoma mortality rate | Colon cancer | Males | Correlation | r = 0.47 (p < 0.01) | Harm | n/a | ||| | Females | Correlation | r = 0.43 (p < 0.01) | Harm | n/a | ||||| | Rectum cancer | Males | Correlation | r = 0.31 (p 0.05) | Harm | n/a | ||||| | Grant59 | China | Proxy for radiation (latitude) | Degree of latitude* | Colorectal cancer | 35–64 years old males | Regression | β = 0.41 (p = 0.005) | Benefit | High | | Grant62 (1992) | France | Proxy for radiation (latitude) | Degree of latitude* | Colorectal cancer | Males | Correlation | r = 0.53 (p = 0.01) | Benefit | n/a | | Females | Correlation | r = 0.46 (p = 0.04) | Benefit | n/a | ||||| | Grant62 (1998–2000) | France | Proxy for radiation (latitude) | Degree of latitude squared* | Colorectal cancer | Males | Correlation | r = 0.49 (p = 0.02) | Benefit | n/a | | Females | Correlation | r = 0.65 (p = 0.001) | Benefit | n/a | ||||| | Grant64 | USA (California) | Proxy for radiation (latitude) | Degree of latitude* | Colon cancer | White males | Regression | β = 0.47 (p = 0.01) | Benefit | High | | Rectum cancer | White males | Regression | β = 0.48 (p = 0.007) | Benefit | High | |||| | Behavioural (NMSC mortality) | NMSC mortality rate | Colon cancer | White males | Regression | β = −0.64 (p = 0.002) | Benefit | High | || | Rectum cancer | White males | Regression | β = −0.48 (p = 0.009) | Benefit | High | |||| | Veach et al.94 | USA | Radiation (solar radiation) | Weber/m2 | Colorectal cancer | Black males | Regression | β = 0.003 (p < 0.001) | Harm | Some concerns | | White males | Regression | β = −0.001 (p = 0.001) | Benefit | Some concerns | ||||| | Hispanic males | Regression | β = 0.001 (p = 0.033) | Harm | Some concerns | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; n/a: not applicable; NMSC: non-melanoma skin cancer; NR: not reported r: correlation coefficient; RoB: risk of bias; UVB: ultraviolet B radiation; UVB DNA: DNA-weighted UVB radiation. Pancreatic cancer mortality. There were seven articles measuring the effect of sunlight on pancreatic cancer mortality, with some reporting multiple exposure types or date ranges (11 analyses reported across the seven articles). Six articles had an ecological design and one used a cohort. There were seven analyses examining the effect of radiation, two articles looked at proxy for radiation measures and two looked at sunlight exposure behaviours. The majority of analyses (n = 9) suggested that higher levels of sunlight are associated with a decreased risk of pancreatic cancer mortality, as reported in Boscoe and Schymura30, Fukuda et al.48, Lin et al.76, Neale et al.81, Grant and Garland56, and Grant58. However, one analysis in Grant58 indicated there may be a harmful association between melanoma mortality rates and pancreatic cancer mortality. One analysis in Fleischer and Fleischer44 was reported narratively, without a direction of effect (Figure 10 and Table 7). | Study | Location | Exposure | Unit of analysis | Outcome | Subgroup | Analysis | Results | Direction of effect | RoB | |---|---|---|---|---|---|---|---|---|---| | Fleischer and Fleischer44 | USA | Radiation (solar radiation) | kJ/m2 | Pancreas cancer | n/a | Narrative | “No associations were demonstrated between solar energy and cancer mortality for pancreatic cancer (p = 0.07)” | NR | n/a | | Grant and Garland56 (1950–1969) | USA | Radiation (UVB) | kJ/m2 | Pancreas cancer | White males | Regression | β = −0.39 (p = 0.02) | Benefit | Some concerns | | White females | Regression | β = −0.74 (p < 0.001) | Benefit | Some concerns | ||||| | Grant and Garland56 (1970–1994) | USA | Radiation (UVB) | kJ/m2 | Pancreas cancer | White males | Regression | β = −0.46 (p = 0.005) | Benefit | Some concerns | | White females | Regression | β = −0.34 (p = 0.06) | Benefit | Some concerns | ||||| | Grant58 | Spain | Proxy for radiation (latitude) | Degree of latitude* | Pancreas cancer | Males | Correlation | r = 0.55 (p < 0.01) | Benefit | n/a | | Females | Correlation | r = 0.40 (p < 0.01) | Benefit | n/a | ||||| | Behavioural (NMSC mortality; melanoma mortality) | NMSC mortality rate | Pancreas cancer | Males | Correlation | r = −0.35 (p < 0.05) | Benefit | n/a | || | Females | Correlation | r = −0.35 (p 0.05) | Harm | n/a | ||| | Females | Correlation | r = 0.45 (p < 0.01) | Harm | n/a | Note. *An increase in latitude indicates a decrease in sunlight exposure. Therefore, a positive relationship between latitude and mortality suggests a protective effect of sunlight. Abbreviations: b: regression beta coefficient; n/a: not applicable; NMSC: non-melanoma skin cancer; NR: not reported r: correlation coefficient; RoB: risk of bias; UVB: ultraviolet B radiation. Cause-specific CVD mortality. Six articles looked at cause-specific CVD mortality (7 analyses across the 6 articles); four with an ecological design and two using a cohort design. All seven analyses examined radiation measures. Four results looked at the effect on heart disease mortality and three looked at the effect on stroke mortality. The findings were mixed (Figure 11 and Table 8). Two analyses suggested a beneficial effect of radiation on heart disease mortality42,87. However, three analyses reported a harmful effect on stroke mortality42,52,76. One produced mixed results43, and one did not report a direction of effect34 (Figure 11 and Table 8). | Study | Location | Exposure | Unit of analysis | Outcome | Subgroup | Analysis | Results | Direction of effect | RoB | |---|---|---|---|---|---|---|---|---|---| | Camara and Brandao34 | Worldwide | Radiation (Solar incidence) | kWh/m−2 / day−1 | Coronary heart disease | n/a | Narrative | No significant difference between high and low sunlight incidence countries (p > 0.05) | NR | n/a | | Ezzati et al.42 | USA | Radiation (insolation) | NR | Ischaemic heart disease | 45+ year old males | Regression | β = −0.00023 (95% CI −0.0011 to 0.00061) | Benefit | Some concerns | | 45+ year old females | Regression | β = −0.00014 (95% CI −0.00099 to 0.00072) | Benefit | Some concerns | ||||| | Stroke | 45+ year old males | Regression | β = 0.00046 (95% CI 0.0002 to 0.00072) | Harm | Some concerns | |||| | 45+ year old females | Regression | β = 0.00062 (95% CI 0.00025 to 0.00099) | Harm | Some concerns | ||||| | Scarborough et al.87 | UK | Radiation (sunshine) | 1000s hours/ year | Coronary heart disease | Males | Regression | β = −27.3 (p < 0.05) | Benefit | Some concerns | | Females | Regression | β = −14.3 (p < 0.05) | Benefit | Some concerns | We sought to investigate differences in effect between people with different skin types/colours or ethnicity. However, information to allow this was limited. Most articles reported findings that examined the whole population (62%), and several limited their population to White people only (24%). One study in the USA examined all-CVD mortality by ethnicity subgroup26, reporting slightly higher mortality risk associated with sunlight exposure for White, Black, Hispanic and Asian people, but a slightly lower risk among Native Americans. Also in the USA, Boscoe and Schymura30 found that residing along the southern border (erythemally-weighted UVB exposure of roughly 1540 kJ/m2/year) was associated with a decreased risk of breast cancer mortality compared with residing along the northern border (roughly 650 kJ/m2/year) among both White women (RR = 0.87, 95% CI 0.85 to 0.88) and Black women (RR = 0.90, 95% CI 0.86 to 0.94). Pennello et al.84 found a harmful relationship between UVB and both melanoma and NMSC mortality for both Black and White people. Finally, Veach et al.94 found that higher solar radiation (Weber/m2 by state) was associated with higher risk of bowel cancer mortality in Black (β = 0.003, p < 0.001) and Hispanic men (β = 0.001, p = 0.033), but lower risk in White men (β = −0.001, p = 0.001). It was sometimes possible to make indirect comparisons across people of different skin types/colours or ethnicity. Grant and Garland56 and Grant57 reported data for the White population and the Black population, respectively, in the USA between 1970 and 1994. The results were suggestive of a beneficial effect of sunlight on all-cancer, breast cancer and bowel cancer mortality for both Black and White people. Boscoe and Schymura30 reported that higher levels of sunlight exposure were associated with a decreased risk of prostate cancer in White men in the USA between 1993 and 2002 (RR = 0.85, 95% CI 0.84 to 0.87); whilst conversely, Colli and Grant37 observed a small positive association between higher winter UV Index and prostate cancer mortality among Black men in the USA between 1992 and 2001, though the confidence intervals were compatible with both benefit and harm (β = 0.20; 95% CI −2.4 to 2.8). The evidence identified by this review provides a mixed message about the association between sunlight exposure and mortality risk. Eight articles reported data for our primary outcome, with half having results in the direction of a beneficial association and half with results in the direction of a harmful association between sunlight and all-cause mortality. Mixed results were also found for all-CVD mortality, while a majority reported that higher levels of sunlight were associated with lower risk of all-cancer mortality. There was considerable uncertainty in the results across all outcomes. As expected, most articles looking at skin cancer found that higher levels of sunlight were associated with higher levels of both melanoma and NMSC mortality. In contrast, most articles examining the five cancers with the highest UK mortality rate (breast, prostate, lung, bowel and pancreatic cancer) found that higher levels of sunlight were associated with lower risks of mortality. However, for each of these specific cancers the evidence was not fully consistent, with some findings also suggesting a harmful effect of sunlight. There were also mixed findings when looking at specific causes of CVD mortality (heart disease and stroke). As with the primary outcomes, there was considerable uncertainty in the findings. Many of the associations we observed between higher sunlight exposure and lower risk of all-cause and all-CVD mortality came from studies conducted in higher latitude countries (specifically, the UK and Sweden). On the other hand, many of the studies finding associations between higher sunlight exposure and higher risk of these mortality outcomes were conducted in the USA. These observations are consistent with the possibility that, whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index10. However, this is not a conclusion we can reach with confidence from the data, given the limitations of the evidence base. Furthermore, a potentially beneficial effect of sunlight on all-cause and all-CVD mortality was observed in Hong Kong52, the location closest to the equator amongst all those included in the analyses. We intended to examine the extent to which the effects of sunlight exposure on mortality would vary according to skin type/colour or ethnicity. Evidence to allow this investigation was limited. In the four studies that reported results by these subgroups, the findings suggested some differences. For example, in the USA, the Native American population were found to have a more beneficial association between sunlight exposure and all-CVD mortality, compared with other ethnicities26. Though there were also findings suggesting similar results between those of different skin type/colour or ethnicity, such as the beneficial association between UVB and breast cancer mortality found for both Black and White women30. Around a quarter of the main articles included in this review were restricted to White populations. Around two thirds reported on the whole population, though given that a large number of these were conducted in Europe and North America, it is likely that the populations in those articles were predominantly White as well. In order to gain a more complete picture of the relationship between sunlight exposure and mortality, further studies investigating the impact of skin type/colour or ethnicity are warranted. This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types. While it is well-established that sunlight increases the risk of skin cancer, particularly through UVR damage to skin cell DNA, the mechanisms through which sunlight may affect non-skin cancer risk and mortality are unclear. Sunlight exposure of the skin is usually the body’s major source of vitamin D11 and experimental studies show that vitamin D can potentially slow or prevent the development of cancer by several cellular mechanisms. These include promoting cellular differentiation and cell death, reducing cancer cell division and tumour blood vessel formation, inhibiting tumour progression and metastasis98, and stimulating the immune response to cancer cells99. Moreover, vitamin D receptors are present in many organs and cell types100. It is therefore conceivable that some of the beneficial effects of sunlight on mortality may be mediated by vitamin D. However, in randomized trials, vitamin D supplements had little to no effect on the risk of developing cancers (both overall and at specific sites) as well as on CVD and all-cause mortality101. It is possible, therefore, that the suggested beneficial effects of sunlight on cancer mortality may involve vitamin D-independent pathways. The radiation measured by studies in this review included solar radiation (encompassing UV, visible and infrared radiation) and ambient UVR (UVB and UVA), while many focused primarily on UVB, which encompasses the wavelengths initiating vitamin D synthesis in the skin, as well as being principally responsible for direct DNA damage in skin cells. However, it is now recognized that there is a range of potential benefits of solar radiation on health besides vitamin D synthesis102. The radiation responsible for these effects may include UVB and/or UVA, and possibly other types of radiation. For example, UVA and UVB are reported to regulate release of the vasodilator nitric oxide from skin cells, potentially protecting against CVD15,16. While UVB is generally more potent than UVA in effecting local skin immunomodulation, both may influence systemic immunity100. Therefore, it is warranted to consider a broader range of solar radiation and its effects with respect to mortality, than UVB alone. Strengths of our review include our pre-specified review methodology with a comprehensive search and adoption of systematic review methods widely considered to reduce biases and human error. We included a wide range of measures of sunlight exposure, including measurements of radiation, proxy measures of radiation, and behaviour associated with sunlight exposure. Since we might expect the nature of confounding to be different for different exposures (e.g. for geographical level versus individual level exposure measures), our approach builds in the idea of ‘triangulating’ analyses that have different potential biases103. Our comprehensive approach enabled enquiry into an increasingly debated area of public health, i.e. the benefits and risks of sunlight exposure, through assessment of associations with all-cause mortality, and separately with all-cancer mortality and all-CVD mortality. We found a large number of articles and have presented the results systematically by outcome, with consideration of risk of bias, consistency of findings and imprecision. There was considerable potential for bias in the results of the included studies. None of the results included in this review were judged to be at low risk of bias, and, in most cases, we judged the results to be at high risk of bias. The main sources of bias were a lack of control over important confounders, and the measurement of sunlight exposure. There is also a high risk of publication bias, with the potential for statistically significant associations (potentially in either direction) being considered more attractive for publication. Concerns about indirectness (or applicability) of the evidence arose primarily from the measures of exposure. The measures of radiation encompassed heterogenous methodology and measurement units, ranging from satellite-derived data and ground-based measurements of radiation, to mean annual sunshine hours. Some of these measures can provide reliable estimates of ambient sunlight levels at a given location, but do not provide information about individual-level or skin exposure, which are influenced by sun exposure behaviour and sun protection measures (e.g. use of clothing, sunscreens and shade). Some of the findings were imprecise, with wide margins of error which were often compatible with a higher and lower risk of mortality being associated with higher sunlight exposure. There was also inconsistency in the findings. For every outcome included in the review, we found results suggesting both a beneficial and a harmful effect of sunlight. In conclusion, evidence from existing observational epidemiological studies of the association between sunlight exposure and mortality is inconclusive. While most studies of skin cancer mortality demonstrate a higher risk associated with more exposure to sunlight, many studies of other cancers have reported lower risks associated with more exposure to sunlight. Evidence for cardiovascular mortality is mixed. Perhaps because of a variety of effects that sunlight can have, or possibly because of potential biases in the studies available, findings for overall mortality are too variable to provide a rationale for changes to sun protection guidance. Figshare: Dataset for “The effects of sunlight exposure on mortality: a systematic review of epidemiological studies” https://doi.org/10.6084/m9.figshare.28660109104. The project contains the following underlying data: Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0). Figshare: Supplementary information for ““The effects of sunlight exposure on mortality: a systematic review of epidemiological studies” https://doi.org/10.6084/m9.figshare.2906965721 This project contains the following extended data: - Appendix S1. Search strategies. - Table S1. Characteristics of included studies. - Table S2. Overlapping data and main article selection. - Table S3. Description of exposures measured in articles included in analysis. - Table S4. Risk-of-bias assessments. Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0). Figshare: PRISMA checklist for "The effects of sunlight exposure on mortality: a systematic review of epidemiological studies" https://doi.org/10.6084/m9.figshare.28937123105. Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0). We thank Monika Halicka and Christie Cabral (University of Bristol) for their contributions to writing the protocol for this review. We thank Ann R Webb (Professor of Atmospheric Radiation, University of Manchester) for her advice regarding the atmospheric radiation dosimetry. LER acknowledges the support of the NIHR Manchester Biomedical Research Centre (NIHR203308). We thank members of our public advisory panel: Sharon Bernard, Joy Bramwell, Vinette Jones and Robin Clay. Permission was obtained to include the names and affiliations of those acknowledged. Faculty Opinions recommendedReferences - 1. Cancer Research UK: Sun safety. 2021. Reference Source - 2. NHS: Sunscreen and sun safety. 2022. Reference Source - 3. NICE: Sunlight exposure: risks and benefits (NG34). 2016. Reference Source - 4. AAD: Practice safe sun. 2024. Reference Source - 5. IARC:A review of human carcinogens. Part D: Radiation. IARC Monographs on the Evaluation of Carcingenic Risks to Humans. 100D. Lyon, France: IARC, 2012. - 6. Brown KF, Rumgay H, Dunlop C, et al.: The fraction of cancer attributable to modifiable risk factors in England, Wales, Scotland, Northern Ireland, and the United Kingdom in 2015. Br J Cancer. 2018; 118(8): 1130–41. PubMed Abstract | Publisher Full Text | Free Full Text - 7. Lashway SG, Worthen ADM, Abuasbeh JN, et al.: A meta-analysis of sunburn and basal cell carcinoma risk. Cancer Epidemiol. 2023; 85: 102379. PubMed Abstract | Publisher Full Text - 8. Kwiatkowska M, Ahmed S, Ardern-Jones M, et al.: An updated report on the incidence and epidemiological trends of keratinocyte cancers in the United Kingdom 2013-2018. Skin Health Dis. 2021; 1(4): e61. PubMed Abstract | Publisher Full Text | Free Full Text - 9. Olsen CM, Pandeya N, Ragaini BS, et al.: International patterns and trends in the incidence of melanoma and cutaneous squamous cell carcinoma, 1989-2020. Br J Dermatol. 2024; 190(4): 492–500. PubMed Abstract | Publisher Full Text - 10. Weller RB: Sunlight: time for a rethink? J Invest Dermatol. 2024; 144(8): 1724–32. PubMed Abstract | Publisher Full Text - 11. Saraff V, Shaw N: Sunshine and vitamin D. Arch Dis Child. 2016; 101(2): 190–2. PubMed Abstract | Publisher Full Text - 12. Grant WB, Holick MF: Benefits and requirements of vitamin D for optimal health: a review. Altern Med Rev. 2005; 10(2): 94–111. PubMed Abstract - 13. Raymond-Lezman JR, Riskin SI: Benefits and risks of sun exposure to maintain adequate vitamin D levels. Cureus. 2023; 15(5): e38578. PubMed Abstract | Publisher Full Text | Free Full Text - 14. Wimalawansa SJ: Infections and autoimmunity-the immune system and vitamin D: a systematic review. Nutrients. 2023; 15(17): 3842. PubMed Abstract | Publisher Full Text | Free Full Text - 15. Pelegrino MT, Paganotti A, Seabra AB, et al.: Photochemistry of nitric oxide and S-nitrosothiols in human skin. Histochem Cell Biol. 2020; 153(6): 431–41. PubMed Abstract | Publisher Full Text | Free Full Text - 16. Liu D, Fernandez BO, Hamilton A, et al.: UVA irradiation of human skin vasodilates arterial vasculature and lowers blood pressure independently of nitric oxide synthase. J Invest Dermatol. 2014; 134(7): 1839–46. PubMed Abstract | Publisher Full Text - 17. Lopes F, Sleiman MG, Sebastian K, et al.: UV exposure and the risk of cutaneous melanoma in skin of color: a systematic review. JAMA Dermatol. 2021; 157(2): 213–9. PubMed Abstract | Publisher Full Text - 18. Raimondi S, Suppa M, Gandini S: Melanoma epidemiology and sun exposure. Acta Derm Venereol. 2020; 100(11): adv00136. PubMed Abstract | Publisher Full Text | Free Full Text - 19. Centre for Reviews and Dissemination: Systematic Reviews: CRD’s guidance for undertaking reviews in health care. York: University of York, 2009. Reference Source - 20. Higgins JPT, Thomas J, Chandler J, et al.: Cochrane handbook for systematic reviews of interventions. version 6.5 (updated August 2024). Cochrane, 2024. Reference Source - 21. Parkhouse T, Spiga F, Rhodes LE, et al.: Supplementary information for “The effects of sunlight exposure on mortality: a systematic review of epidemiological studies”. figshare. 2025. http://www.doi.org/10.6084/m9.figshare.29069657 - 22. World Health Organisation: International Classification of Diseases Eleventh Revision (ICD-11). Geneva: World Health Organization, 2022. Reference Source - 23. Cancer Research UK: Cancer mortality for common cancers. Reference Source - 24. Higgins JPT, Morgan RL, Rooney AA, et al.: A tool to assess Risk Of Bias In Non-randomized follow-up Studies of Exposure effects (ROBINS-E). Environ Int. 2024; 186: 108602. PubMed Abstract | Publisher Full Text | Free Full Text - 25. Alcala Ramirez del Puerto A, Hernandez-Rodriguez JC, Sendin-Martin M, et al.: Skin cancer mortality in Spain: adjusted mortality rates by province and related risk factors. Int J Dermatol. 2023; 62(6): 776–82. PubMed Abstract | Publisher Full Text - 26. Al-Hamdan AZ, Preetha PP, Al-Hamdan MZ, et al.: Reconnoitering the linkage between Cardiovascular Disease mortality and long-term exposures to outdoor environmental factors in the USA using remotely-sensed data. J Environ Sci Health A Tox Hazard Subst Environ Eng. 2018; 53(9): 809–818. PubMed Abstract | Publisher Full Text - 27. Altug F, Kilçiksiz S: The spatial factors and regional concentration of cancer-related deaths in Turkey. J Geogr-Cografya Dergisi. 2020; (41): 1–12. Publisher Full Text - 28. Apperly FL: The relation of solar radiation to cancer mortality in North America. Cancer Res. 1941; 1(3): 191–195. Reference Source - 29. Baker-Blocker A: Ultraviolet radiation and melanoma mortality in the United States. Environ Res. 1980; 23(1): 24–8. PubMed Abstract | Publisher Full Text - 30. Boscoe FP, Schymura MJ: Solar Ultraviolet-B exposure and cancer incidence and mortality in the United States, 1993–2002. BMC Cancer. 2006; 6: 264. PubMed Abstract | Publisher Full Text | Free Full Text - 31. Boz S, Berlin C, Kwiatkowski M, et al.: A prospective cohort analysis of residential radon and UV exposures and malignant melanoma mortality in the Swiss population. Environ Int. 2022; 169: 107437. PubMed Abstract | Publisher Full Text - 32. Bulliard JL, Cox B, Elwood JM: Latitude gradients in melanoma incidence and mortality in the non-Maori population of New Zealand. Cancer Causes Control. 1994; 5(3): 234–40. PubMed Abstract | Publisher Full Text - 33. Bulliard JL: Site-specific risk of cutaneous malignant melanoma and pattern of Sun exposure in New Zealand. Int J Cancer. 2000; 85(5): 627–32. PubMed Abstract | Publisher Full Text - 34. Câmara AB, Brandão IA: Study of the relationship between vitamin D deficiency, sunlight incidence and skeletal/extra skeletal diseases. Acta Sci Health Sci. 2020; 42(1): e50599. Publisher Full Text - 35. Chen W, Clements M, Rahman B, et al.: Relationship between cancer mortality/incidence and ambient Ultraviolet B irradiance in China. Cancer Causes Control. 2010; 21(10): 1701–9. PubMed Abstract | Publisher Full Text - 36. Colli JL, Colli A: International comparisons of prostate cancer mortality rates with dietary practices and sunlight levels. Urol Oncol. 2006; 24(3): 184–94. PubMed Abstract | Publisher Full Text - 37. Colli JL, Grant WB: Solar Ultraviolet B radiation compared with prostate cancer incidence and mortality rates in United States. Urology. 2008; 71(3): 531–5. PubMed Abstract | Publisher Full Text - 38. Colli JL, Kolettis PN: Bladder cancer incidence and mortality rates compared to ecologic factors among states in America. Int Urol Nephrol. 2010; 42(3): 659–65. PubMed Abstract | Publisher Full Text - 39. Crocetti E, Buzzoni C, Chiarugi A, et al.: Relationship between latitude and melanoma in Italy. ISRN Oncol. 2012; 2012: 864680. PubMed Abstract | Publisher Full Text | Free Full Text - 40. Donneyong MM, Taylor KC, Kerber RA, et al.: Is outdoor recreational activity an independent predictor of cardiovascular disease mortality - NHANES III? Nutr Metab Cardiovasc Dis. 2016; 26(8): 735–42. PubMed Abstract | Publisher Full Text - 41. Elwood JM, Lee JA, Walter SD, et al.: Relationship of melanoma and other skin cancer mortality to latitude and ultraviolet radiation in the United States and Canada. Int J Epidemiol. 1974; 3(4): 325–32. PubMed Abstract | Publisher Full Text - 42. Ezzati M, Horwitz MEM, Thomas DSK, et al.: Altitude, life expectancy and mortality from ischaemic heart disease, stroke, COPD and cancers: national population-based analysis of US counties. J Epidemiol Community Health. 2012; 66(7): e17. PubMed Abstract | Publisher Full Text - 43. Faeh D, Moser A, Panczak R, et al.: Independent at heart: persistent association of altitude with ischaemic heart disease mortality after consideration of climate, topography and built environment. J Epidemiol Community Health. 2016; 70(8): 798–806. PubMed Abstract | Publisher Full Text - 44. Fleischer AB, Fleischer SE: Solar radiation and the incidence and mortality of leading invasive cancers in the United States. Dermatoendocrinol. 2016; 8(1): e1162366. PubMed Abstract | Publisher Full Text | Free Full Text - 45. Freedman DM, Zahm SH, Dosemeci M: Residential and occupational exposure to sunlight and mortality from non-Hodgkin's lymphoma: composite (threefold) case-control study. BMJ. 1997; 314(7092): 1451–5. PubMed Abstract | Publisher Full Text | Free Full Text - 46. Freedman DM, Dosemeci M, McGlynn K: Sunlight and mortality from breast, ovarian, colon, prostate, and non-melanoma skin cancer: a composite death certificate based case-control study. Occup Environ Med. 2002; 59(4): 257–62. PubMed Abstract | Publisher Full Text | Free Full Text - 47. Fu Y, Wang W: Association between provincial sunshine duration and mortality rates in China: panel data study. Heliyon. 2023; 9(5): e15862. PubMed Abstract | Publisher Full Text | Free Full Text - 48. Fukuda Y, Nakaya T, Nakao H, et al.: Multilevel analysis of solar radiation and cancer mortality using ecological data in Japan. Biosci Trends. 2008; 2(6): 235–40. PubMed Abstract - 49. Garland CF, Garland FC: Do sunlight and vitamin D reduce the likelihood of colon cancer. Int J Epidemiol. 1980; 9(3): 227–31. PubMed Abstract | Publisher Full Text - 50. Garland FC, Garland CF, Gorham ED, et al.: Geographic variation in breast cancer mortality in the United States: a hypothesis involving exposure to solar radiation. Prev Med. 1990; 19(6): 614–22. PubMed Abstract | Publisher Full Text - 51. Garland CF, Garland FC, Gorham ED: Epidemiologic evidence for different roles of ultraviolet A and B radiation in melanoma mortality rates. Ann Epidemiol. 2003; 13(6): 395–404. PubMed Abstract | Publisher Full Text - 52. Goggins WB, Chan EYY, Yang CY, et al.: Associations between mortality and meteorological and pollutant variables during the cool season in two Asian cities with sub-tropical climates: Hong Kong and Taipei. Environ Health. 2013; 12: 59. PubMed Abstract | Publisher Full Text | Free Full Text - 53. Grant WB: An ecologic study of dietary and solar Ultraviolet-B links to breast carcinoma mortality rates. Cancer. 2002; 94(1): 272–81. PubMed Abstract | Publisher Full Text - 54. Grant WB: An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar Ultraviolet-B radiation. Cancer. 2002; 94(6): 1867–75. PubMed Abstract | Publisher Full Text - 55. Grant WB: A multicountry ecologic study of risk and risk reduction factors for prostate cancer mortality. Eur Urol. 2004; 45(3): 271–9. PubMed Abstract | Publisher Full Text - 56. Grant WB, Garland CF: The association of solar Ultraviolet B (UVB) with reducing risk of cancer: multifactorial ecologic analysis of geographic variation in age-adjusted cancer mortality rates. Anticancer Res. 2006; 26(4A): 2687–99. PubMed Abstract - 57. Grant WB: Lower vitamin-D production from solar Ultraviolet-B irradiance may explain some differences in cancer survival rates. J Natl Med Assoc. 2006; 98(3): 357–64. PubMed Abstract | Free Full Text - 58. Grant WB: An ecologic study of cancer mortality rates in Spain with respect to indices of solar UVB irradiance and smoking. Int J Cancer. 2007; 120(5): 1123–8. PubMed Abstract | Publisher Full Text - 59. Grant WB: Does solar ultraviolet irradiation affect cancer mortality rates in China? Asian Pac J Cancer Prev. 2007; 8(2): 236–42. PubMed Abstract - 60. Grant WB: An ecological study of cancer mortality rates including indices for dietary iron and zinc. Anticancer Res. 2008; 28(3B): 1955–63. PubMed Abstract - 61. Grant WB: Air pollution in relation to U.S. cancer mortality rates: an ecological study; likely role of carbonaceous aerosols and polycyclic aromatic hydrocarbons. Anticancer Res. 2009; 29(9): 3537–45. PubMed Abstract - 62. Grant WB: An ecological study of cancer incidence and mortality rates in France with respect to latitude, an index for vitamin D production. Dermatoendocrinol. 2010; 2(2): 62–7. PubMed Abstract | Publisher Full Text | Free Full Text - 63. Grant WB: An ecological study of cancer mortality rates in the United States with respect to solar Ultraviolet-B doses, smoking, alcohol consumption, and urban/rural residence. Dermatoendocrinol. 2010; 2(2): 68–76. PubMed Abstract | Publisher Full Text | Free Full Text - 64. Grant WB: An ecological study of cancer mortality rates in California, 1950–64, with respect to solar UVB and smoking indices. Dermatoendocrinol. 2012; 4(2): 176–82. PubMed Abstract | Publisher Full Text | Free Full Text - 65. Hanchette CL, Schwartz GG: Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer. 1992; 70(12): 2861–9. PubMed Abstract | Publisher Full Text - 66. Hartge P, Devesa SS, Grauman DJ, et al.: Non-Hodgkin's lymphoma and sunlight. J Natl Cancer Inst. 1996; 88(5): 298–300. PubMed Abstract | Publisher Full Text - 67. He W, Zhu F, Ma X, et al.: Actinic skin damage and mortality--the First National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. PLoS One. 2011; 6(5): e19907. PubMed Abstract | Publisher Full Text | Free Full Text - 68. Jemal A, Devesa SS, Fears TR, et al.: Cancer surveillance series: changing patterns of cutaneous malignant melanoma mortality rates among whites in the United States. J Natl Cancer Inst. 2000; 92(10): 811–8. PubMed Abstract | Publisher Full Text - 69. John EM, Koo J, Schwartz GG: Sun exposure and prostate cancer risk: evidence for a protective effect of early-life exposure. Cancer Epidemiol Biomarkers Prev. 2007; 16(6): 1283–6. PubMed Abstract | Publisher Full Text - 70. Kinoshita S, Wagatsuma Y, Okada M: Geographical distribution for malignant neoplasm of the pancreas in relation to selected climatic factors in Japan. Int J Health Geogr. 2007; 6: 34. PubMed Abstract | Publisher Full Text | Free Full Text - 71. Langford IH, Bentham G, McDonald AL: Mortality from non-Hodgkin lymphoma and UV exposure in the European Community. Health Place. 1998; 4(4): 355–64. PubMed Abstract | Publisher Full Text - 72. Langford IH, Bentham G, McDonald AL: Multi-level modelling of geographically aggregated health data: a case study on malignant melanoma mortality and UV exposure in the European Community. Stat Med. 1998; 17(1): 41–57. PubMed Abstract | Publisher Full Text - 73. Lee JA, Scotto J: Melanoma: linked temporal and latitude changes in the United States. Cancer Causes Control. 1993; 4(5): 413–8. PubMed Abstract | Publisher Full Text - 74. Lee JA: Declining effect of latitude on melanoma mortality rates in the United States. A preliminary study. Am J Epidemiol. 1997; 146(5): 413–7. PubMed Abstract | Publisher Full Text - 75. Lefkowitz ES, Garland CF: Sunlight, vitamin D, and ovarian cancer mortality rates in US women. Int J Epidemiol. 1994; 23(6): 1133–6. PubMed Abstract | Publisher Full Text - 76. Lin SW, Wheeler DC, Park Y, et al.: Prospective study of ultraviolet radiation exposure and mortality risk in the United States. Am J Epidemiol. 2013; 178(4): 521–33. PubMed Abstract | Publisher Full Text | Free Full Text - 77. Lindqvist PG, Epstein E, Landin-Olsson M, et al.: Avoidance of sun exposure is a risk factor for all-cause mortality: results from the Melanoma in Southern Sweden cohort. J Intern Med. 2014; 276(1): 77–86. PubMed Abstract | Publisher Full Text - 78. Lindqvist PG, Epstein E, Nielsen K, et al.: Avoidance of sun exposure as a risk factor for major causes of death: a competing risk analysis of the Melanoma in Southern Sweden cohort. J Intern Med. 2016; 280(4): 375–87. PubMed Abstract | Publisher Full Text - 79. Lindqvist PG, Epstein E, Landin-Olsson M, et al.: Women with fair phenotypes seem to confer a survival advantage in a low UV milieu. A nested matched case control study. PLoS One. 2020; 15(1): e0228582. PubMed Abstract | Publisher Full Text | Free Full Text - 80. Mizoue T: Ecological study of solar radiation and cancer mortality in Japan. Health Phys. 2004; 87(5): 532–8. PubMed Abstract | Publisher Full Text - 81. Neale RE, Youlden DR, Krnjacki L, et al.: Latitude variation in pancreatic cancer mortality in Australia. Pancreas. 2009; 38(4): 387–90. PubMed Abstract | Publisher Full Text - 82. Paffenbarger RS Jr, Wing AL, Hyde RT: Characteristics in youth predictive of adult-onset malignant lymphomas, melanomas, and leukemias: brief communication. J Natl Cancer Inst. 1978; 60(1): 89–92. PubMed Abstract | Publisher Full Text - 83. Page WF, Whiteman D, Murphy M: A comparison of melanoma mortality among WWII veterans of the Pacific and European theaters. Ann Epidemiol. 2000; 10(3): 192–5. PubMed Abstract | Publisher Full Text - 84. Pennello G, Devesa S, Gail M: Association of surface Ultraviolet B radiation levels with melanoma and nonmelanoma skin cancer in United States blacks. Cancer Epidemiol Biomarkers Prev. 2000; 9(3): 291–7. PubMed Abstract - 85. Rivas M, Rojas E, Calaf GM, et al.: Association between non-melanoma and melanoma skin cancer rates, vitamin D and latitude. Oncol Lett. 2017; 13(5): 3787–92. PubMed Abstract | Publisher Full Text | Free Full Text - 86. Santos Arrontes D, García González JI, Martín Muñoz MP, et al.: [Will the climate change affect the mortality from prostate cancer?] Arch Esp Urol. 2007; 60(2): 119–23. PubMed Abstract | Publisher Full Text - 87. Scarborough P, Allender S, Rayner M, et al.: Contribution of climate and air pollution to variation in coronary heart disease mortality rates in England. PLoS One. 2012; 7(3): e32787. PubMed Abstract | Publisher Full Text | Free Full Text - 88. Schwartz GG, Hanchette CL: UV, latitude, and spatial trends in prostate cancer mortality: all sunlight is not the same (United States). Cancer Causes Control. 2006; 17(8): 1091–101. PubMed Abstract | Publisher Full Text - 89. Stevenson AC, Clemens T, Pairo-Castineira E, et al.: Higher ultraviolet light exposure is associated with lower mortality: an analysis of data from the UK Biobank cohort study. medRxiv. 2023. Publisher Full Text - 90. Takahashi K, Pan G, Feng Y, et al.: Regional correlation between estimated UVB levels and skin cancer mortality in Japan. J Epidemiol. 1999; 9(6 Suppl): S123–8. PubMed Abstract | Publisher Full Text - 91. Taksler GB, Cutler DM, Giovannucci E, et al.: Ultraviolet index and racial differences in prostate cancer incidence and mortality. Cancer. 2013; 119(17): 3195–203. PubMed Abstract | Publisher Full Text - 92. Uehara M, Takahashi K, Hoshuyama T, et al.: Geographical correlation between ambient UVB level and mortality risk of leukemia in Japan. Environ Res. 2003; 92(2): 78–84. PubMed Abstract | Publisher Full Text - 93. van Wijngaarden E, Savitz DA: Occupational sunlight exposure and mortality from non-Hodgkin lymphoma among electric utility workers. J Occup Environ Med. 2001; 43(6): 548–53. PubMed Abstract | Publisher Full Text - 94. Veach E, Xique I, Johnson J, et al.: Race matters: analyzing the relationship between colorectal cancer mortality rates and various factors within respective racial groups. Front Public Health. 2014; 2: 239. PubMed Abstract | Publisher Full Text | Free Full Text - 95. Vienneau D, de Hoogh K, Hauri D, et al.: Effects of radon and UV exposure on skin cancer mortality in Switzerland. Environ Health Perspect. 2017; 125(6): 067009. PubMed Abstract | Publisher Full Text | Free Full Text - 96. Wu W, Weinstock MA: Trends of keratinocyte carcinoma mortality rates in the United States as reported on death certificates, 1999 through 2010. Dermatol Surg. 2014; 40(12): 1395–401. PubMed Abstract | Publisher Full Text - 97. Yang L, Lof M, Veierod MB, et al.: Ultraviolet exposure and mortality among women in Sweden. Cancer Epidemiol Biomarkers Prev. 2011; 20(4): 683–90. PubMed Abstract | Publisher Full Text - 98. Feldman D, Krishnan AV, Swami S, et al.: The role of vitamin D in reducing cancer risk and progression. Nat Rev Cancer. 2014; 14(5): 342–57. PubMed Abstract | Publisher Full Text - 99. Fleet JC, Burcham GN, Calvert RD, et al.: 1alpha, 25 Dihydroxyvitamin D (1,25(OH)2D) inhibits the T cell suppressive function of myeloid derived suppressor cells (MDSC). J Steroid Biochem Mol Biol. 2020; 198: 105557. PubMed Abstract | Publisher Full Text | Free Full Text - 100. Hart PH, Norval M, Byrne SN, et al.: Exposure to ultraviolet radiation in the modulation of human diseases. Annu Rev Pathol. 2019; 14: 55–81. PubMed Abstract | Publisher Full Text - 101. O'Connor EA, Evans CV, Ivlev I, et al.: Vitamin and mineral supplements for the primary prevention of cardiovascular disease and cancer: updated evidence report and systematic review for the US Preventive Services Task Force. JAMA. 2022; 327(23): 2334–47. PubMed Abstract | Publisher Full Text - 102. Neale RE, Lucas RM, Byrne SN, et al.: The effects of exposure to solar radiation on human health. Photochem Photobiol Sci. 2023; 22(5): 1011–47. PubMed Abstract | Publisher Full Text | Free Full Text - 103. Lawlor DA, Tilling K, Davey Smith G: Triangulation in aetiological epidemiology. Int J Epidemiol. 2016; 45(6): 1866–86. PubMed Abstract | Publisher Full Text | Free Full Text - 104. Parkhouse T, Spiga F, Rhodes LE, et al.: Dataset for "The effects of sunlight exposure on mortality: a systematic review of epidemiological studies". [Dataset], figshare. 2025. http://www.doi.org/10.6084/m9.figshare.28660109 - 105. Parkhouse T, Spiga F, Rhodes LE, et al.: PRISMA checklist for "The effects of sunlight exposure on mortality: a systematic review of epidemiological studies". figshare. 2025. http://www.doi.org/10.6084/m9.figshare.28937123 Author details Author details 1 NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, England, BS8 2PS, UK 2 Division of Musculoskeletal and Dermatological Sciences, School of Biological Sciences,, The University of Manchester, Manchester, England, M13 9PL, UK 3 Dermatology Research Centre, Salford Royal Hospital, Manchester Academic Health Science Centre, Northern Care Alliance NHS Foundation Trust, Salford, England, UK 4 NIHR Applied Research Collaboration West (ARC West), University Hospitals Bristol and Weston NHS Foundation Trust, Bristol, England, UK 2 Division of Musculoskeletal and Dermatological Sciences, School of Biological Sciences,, The University of Manchester, Manchester, England, M13 9PL, UK 3 Dermatology Research Centre, Salford Royal Hospital, Manchester Academic Health Science Centre, Northern Care Alliance NHS Foundation Trust, Salford, England, UK 4 NIHR Applied Research Collaboration West (ARC West), University Hospitals Bristol and Weston NHS Foundation Trust, Bristol, England, UK Thomas Parkhouse Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Francesca Spiga Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Data Curation, Formal Analysis, Investigation, Methodology, Project Administration, Visualization, Writing – Original Draft Preparation, Writing – Review & Editing Lesley E Rhodes Roles: Investigation, Methodology, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Original Draft Preparation, Writing – Review & Editing Sarah Dawson Roles: Investigation, Methodology, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Review & Editing Katie E Webster Roles: Investigation, Methodology, Writing – Review & Editing Roles: Investigation, Methodology, Writing – Review & Editing Deborah M Caldwell Roles: Funding Acquisition, Investigation, Methodology, Supervision, Writing – Review & Editing Roles: Funding Acquisition, Investigation, Methodology, Supervision, Writing – Review & Editing Julian P T Higgins Roles: Funding Acquisition, Investigation, Methodology, Project Administration, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing Roles: Funding Acquisition, Investigation, Methodology, Project Administration, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing Competing interests LER is a member of the Committee on Medical Effects of Radiation in the Environment (COMARE), the European Society for Photodermatology (ESPD) board, and the UN Environment Programme’s Environmental Effects Assessment Panel (UNEP EEAP). She collaborates and performs clinical trials with Clinuvel Pharmaceuticals Ltd and Mitsubishi Tanabe Pharma America Inc. No other authors have any competing interests to declare. Grant information This project was funded by the National Institute for Health and Care Research (NIHR) under its Evidence Synthesis Programme (grant reference number: NIHR161983). The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Article Versions (2) Copyright © 2025 Parkhouse T et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. metrics VIEWS $counts.viewCount downloads Citations CITE how to cite this article Parkhouse T, Spiga F, Rhodes LE et al. The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.13980.1) NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article. track receive updates on this article Track an article to receive email alerts on any updates to this article. Current Reviewer Status: ? Key to Reviewer Statuses VIEW HIDE ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested Approved with reservations A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit. Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions Version 1 VERSION 1 PUBLISHED 18 Jun 2025 Views 0 How to cite this report: McKenzie R, Liley B and Liley J. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36176) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36176 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36176 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 09 Sep 2025 Richard McKenzie, National Institute of Water and Atmospheric Research Lauder Atmospheric Research Station (Ringgold ID: 563277), Lauder, Otago, New Zealand; Earth Sciences, New Zealand, Lauder, Otago, New Zealand Ben Liley, Earth Sciences New Zealand, Lauder, Otago, New Zealand James Liley, Mathematical Sciences, University of Durham, Durham, UK Approved with Reservations VIEWS 0 This is a carefully conducted and well-written review that attempts to assess the mortality risk caused by overexposure to UV in sunlight, compared with the latter’s potential benefit in lowering mortality rates. The risk is widely understood to be from ... Continue reading We confirm that we have read this submission and believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however we have significant reservations, as outlined above. Close This is a carefully conducted and well-written review that attempts to assess the mortality risk caused by overexposure to UV in sunlight, compared with the latter’s potential benefit in lowering mortality rates. The risk is widely understood to be from skin cancers, especially melanoma, whereas the benefit has been argued for other cancers and cardio-vascular disease (CVD). The inconclusive result is unsurprising, given the difficulty of measurement, the time lags involved, and confounding factors. The inhomogeneity of data sources between studies must also have been challenging. The authors conclude that no changes in public messaging are justified. Although that may be quite controversial, we find the data presented is a useful guide to current knowledge on the risks and benefits. It also provides a possible stepping stone for future work. We therefore recommend to satisfactorily addressing the concerns about the conclusions raised below. Major Points 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Minor Point: P9, near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. The inconclusive result is unsurprising, given the difficulty of measurement, the time lags involved, and confounding factors. The inhomogeneity of data sources between studies must also have been challenging. The authors conclude that no changes in public messaging are justified. Although that may be quite controversial, we find the data presented is a useful guide to current knowledge on the risks and benefits. It also provides a possible stepping stone for future work. We therefore recommend to satisfactorily addressing the concerns about the conclusions raised below. Major Points 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Minor Point: P9, near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? Yes - Are the conclusions drawn adequately supported by the results presented in the review? Partly

1. Nessvi S, Johansson L, Jopson J, Stewart A, et al.: Association of 25‐Hydroxyvitamin D3 Levels in Adult New Zealanders with Ethnicity, Skin Color and Self‐Reported Skin Sensitivity to Sun Exposure. Photochemistry and Photobiology. 2011; 87 (5): 1173-1178 Publisher Full TextCompeting Interests: No competing interests were disclosed. Reviewer Expertise: Richard McKenzie is an internationally recognised expert in the measurement, understanding, and implications of solar UV radiation. He wrote the initial review, and sought further insight from Ben Liley, a fellow atmospheric scientist with particular expertise in the analysis of UV dosimeter measurements of personal exposure. James Liley also previously contributed to the latter, but is now a medically-qualified biostatistician. He has contributed to the questions on statistics, and shared insight into the public health context of the study CITE HOW TO CITE THIS REPORT McKenzie R, Liley B and Liley J. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36176) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36176 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36176 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below. 1. None of the studies cited were able to directly ... Continue reading Many thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below. 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. Response: We agree with the point raised here – no measures of exposure were direct and there were various issues with the proxy radiation measures. These issues are addressed in the discussion. However, our intention with this review was to be comprehensive and thorough in attempting to assess the effect of sunlight, therefore we felt it was beneficial to include these measures, though grouped separately in order to maintain that distinction. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. Response: The ROBINS-E tool assesses both authorial good practice and fundamental bias. We have added a brief description of the tool in the methods, addressing the point raised here as well as listing the seven domains the tool covers: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? Response: Looking at the primary outcomes – as the reviewers point out, no study has a RoB assessment lower than high for all-cause and all-CVD. So discounting all ‘high’ and ‘very high’ RoB results would leave no results for these two primary outcomes. For all-cancer, this would leave three studies in the forest plot and three in the table – the results of which are similar to those currently presented (i.e., most results in the direction of a beneficial effect). Two papers were assessed to be of very high RoB. For the Yang paper, this was due to it being a proxy measure of sunlight (frequency of annual sunburns) that was self-reported and had the potential for differential measurement error (e.g., those who never sunburnt less likely to mis-report frequency than those regularly sunburnt). For the Grant paper (ref 57), it was due to missing data being an issue for both the outcome and confounders (they excluded states with <40 deaths from lung cancer – a proxy for smoking). 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. Response: We have increased the discussion around the applicability of the measures of exposure in the limitations section of the discussion. This includes mention of the issue you raise here regarding the lack of ambient UV received by individuals, as well as the validity of latitude when considering skin cancer. We also go on to discuss the difficulty of disentangling the effect of sunlight on the skin from the effect of associated confounders, given the lack of individual-level sunlight exposure measures: “This issue is further compounded for the proxy radiation measures, e.g., latitude. For example, a studies using personal dosimeters to investigate individual-level sun exposure in northern and southern latitudes in New Zealand showed that, regardless of latitude, people received less than 2% of ambient UVR exposure (Scragg et al, 2016 ). A finding which is echoed by a study in the UK showing that White people were exposed to ~2% ambient UVR, whilst those with darker skin types were exposed to even less, ~1% (Kift et al, 2013). As such, the validity of latitude as a proxy measure of radiation, especially when considering skin cancer, may be limited and results should be interpreted with caution. Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. Response: We added the sentence “Overall, the current available evidence does not indicate a large effect of sunlight on mortality” to the conclusion. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. Response: We have increased discussion on the issues surrounding skin type/ethnicity and the lack of available data on this. We have added a new paragraph discussing the potential interaction between location and skin type (i.e., in low-UV countries, sunlight may benefit those with lighter skin, whilst in high-UV countries those with darker skin might benefit): “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Later, we have added a couple of sentences to highlight your point here that skin cancer was the only outcome with clear risks outweighing benefits, and that melanoma mortality is lower for those with darker skin: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance” Finally, we have added reference to the recent sun exposure guidance in Australia: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. Response: We agree, there is scope for various future research projects – both utilising the data we have extracted here, as well as novel primary research. We have now increased our discussion of suggestions for future work, before the concluding paragraph. One follow-up study we recommend is to develop a common unit of exposure measurement that the current radiation-based measures could be converted to – this would allow, at least, for statistical synthesis of the results, and would likely open the door to other useful projects. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Response: Again, we agree that in principle the results presented in the review do not necessarily rule out the possibility that mortality is not a sufficient basis for sun safety advice. Although, the safety guidance is of course not only informed by mortality outcomes. There are many other outcomes that could, and should, be considered, such as incidence and quality of life. Unfortunately they were beyond the scope of the current review. However, as mentioned in reply to the point above, we have now increased discussion around future research ideas. As part of the suggested idea of converting results to a common unit of exposure, we mention that doing so could also prove useful for investigating other outcomes, such as incidence of disease. Minor point P9 - near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. Response: We thank the reviewers for highlighting a typo here – that should have read 1,000 kJ/m2. This has been corrected. In the case of all studies, we kept the units of measurement as reported in the paper.Many thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below.Competing Interests: No competing interests were disclosed. Close 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. Response: We agree with the point raised here – no measures of exposure were direct and there were various issues with the proxy radiation measures. These issues are addressed in the discussion. However, our intention with this review was to be comprehensive and thorough in attempting to assess the effect of sunlight, therefore we felt it was beneficial to include these measures, though grouped separately in order to maintain that distinction. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. Response: The ROBINS-E tool assesses both authorial good practice and fundamental bias. We have added a brief description of the tool in the methods, addressing the point raised here as well as listing the seven domains the tool covers: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? Response: Looking at the primary outcomes – as the reviewers point out, no study has a RoB assessment lower than high for all-cause and all-CVD. So discounting all ‘high’ and ‘very high’ RoB results would leave no results for these two primary outcomes. For all-cancer, this would leave three studies in the forest plot and three in the table – the results of which are similar to those currently presented (i.e., most results in the direction of a beneficial effect). Two papers were assessed to be of very high RoB. For the Yang paper, this was due to it being a proxy measure of sunlight (frequency of annual sunburns) that was self-reported and had the potential for differential measurement error (e.g., those who never sunburnt less likely to mis-report frequency than those regularly sunburnt). For the Grant paper (ref 57), it was due to missing data being an issue for both the outcome and confounders (they excluded states with <40 deaths from lung cancer – a proxy for smoking). 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. Response: We have increased the discussion around the applicability of the measures of exposure in the limitations section of the discussion. This includes mention of the issue you raise here regarding the lack of ambient UV received by individuals, as well as the validity of latitude when considering skin cancer. We also go on to discuss the difficulty of disentangling the effect of sunlight on the skin from the effect of associated confounders, given the lack of individual-level sunlight exposure measures: “This issue is further compounded for the proxy radiation measures, e.g., latitude. For example, a studies using personal dosimeters to investigate individual-level sun exposure in northern and southern latitudes in New Zealand showed that, regardless of latitude, people received less than 2% of ambient UVR exposure (Scragg et al, 2016 ). A finding which is echoed by a study in the UK showing that White people were exposed to ~2% ambient UVR, whilst those with darker skin types were exposed to even less, ~1% (Kift et al, 2013). As such, the validity of latitude as a proxy measure of radiation, especially when considering skin cancer, may be limited and results should be interpreted with caution. Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. Response: We added the sentence “Overall, the current available evidence does not indicate a large effect of sunlight on mortality” to the conclusion. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. Response: We have increased discussion on the issues surrounding skin type/ethnicity and the lack of available data on this. We have added a new paragraph discussing the potential interaction between location and skin type (i.e., in low-UV countries, sunlight may benefit those with lighter skin, whilst in high-UV countries those with darker skin might benefit): “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Later, we have added a couple of sentences to highlight your point here that skin cancer was the only outcome with clear risks outweighing benefits, and that melanoma mortality is lower for those with darker skin: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance” Finally, we have added reference to the recent sun exposure guidance in Australia: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. Response: We agree, there is scope for various future research projects – both utilising the data we have extracted here, as well as novel primary research. We have now increased our discussion of suggestions for future work, before the concluding paragraph. One follow-up study we recommend is to develop a common unit of exposure measurement that the current radiation-based measures could be converted to – this would allow, at least, for statistical synthesis of the results, and would likely open the door to other useful projects. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Response: Again, we agree that in principle the results presented in the review do not necessarily rule out the possibility that mortality is not a sufficient basis for sun safety advice. Although, the safety guidance is of course not only informed by mortality outcomes. There are many other outcomes that could, and should, be considered, such as incidence and quality of life. Unfortunately they were beyond the scope of the current review. However, as mentioned in reply to the point above, we have now increased discussion around future research ideas. As part of the suggested idea of converting results to a common unit of exposure, we mention that doing so could also prove useful for investigating other outcomes, such as incidence of disease. Minor point P9 - near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. Response: We thank the reviewers for highlighting a typo here – that should have read 1,000 kJ/m2. This has been corrected. In the case of all studies, we kept the units of measurement as reported in the paper. COMMENTS ON THIS REPORT - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below. 1. None of the studies cited were able to directly ... Continue reading Many thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below. 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. Response: We agree with the point raised here – no measures of exposure were direct and there were various issues with the proxy radiation measures. These issues are addressed in the discussion. However, our intention with this review was to be comprehensive and thorough in attempting to assess the effect of sunlight, therefore we felt it was beneficial to include these measures, though grouped separately in order to maintain that distinction. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. Response: The ROBINS-E tool assesses both authorial good practice and fundamental bias. We have added a brief description of the tool in the methods, addressing the point raised here as well as listing the seven domains the tool covers: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? Response: Looking at the primary outcomes – as the reviewers point out, no study has a RoB assessment lower than high for all-cause and all-CVD. So discounting all ‘high’ and ‘very high’ RoB results would leave no results for these two primary outcomes. For all-cancer, this would leave three studies in the forest plot and three in the table – the results of which are similar to those currently presented (i.e., most results in the direction of a beneficial effect). Two papers were assessed to be of very high RoB. For the Yang paper, this was due to it being a proxy measure of sunlight (frequency of annual sunburns) that was self-reported and had the potential for differential measurement error (e.g., those who never sunburnt less likely to mis-report frequency than those regularly sunburnt). For the Grant paper (ref 57), it was due to missing data being an issue for both the outcome and confounders (they excluded states with <40 deaths from lung cancer – a proxy for smoking). 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. Response: We have increased the discussion around the applicability of the measures of exposure in the limitations section of the discussion. This includes mention of the issue you raise here regarding the lack of ambient UV received by individuals, as well as the validity of latitude when considering skin cancer. We also go on to discuss the difficulty of disentangling the effect of sunlight on the skin from the effect of associated confounders, given the lack of individual-level sunlight exposure measures: “This issue is further compounded for the proxy radiation measures, e.g., latitude. For example, a studies using personal dosimeters to investigate individual-level sun exposure in northern and southern latitudes in New Zealand showed that, regardless of latitude, people received less than 2% of ambient UVR exposure (Scragg et al, 2016 ). A finding which is echoed by a study in the UK showing that White people were exposed to ~2% ambient UVR, whilst those with darker skin types were exposed to even less, ~1% (Kift et al, 2013). As such, the validity of latitude as a proxy measure of radiation, especially when considering skin cancer, may be limited and results should be interpreted with caution. Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. Response: We added the sentence “Overall, the current available evidence does not indicate a large effect of sunlight on mortality” to the conclusion. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. Response: We have increased discussion on the issues surrounding skin type/ethnicity and the lack of available data on this. We have added a new paragraph discussing the potential interaction between location and skin type (i.e., in low-UV countries, sunlight may benefit those with lighter skin, whilst in high-UV countries those with darker skin might benefit): “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Later, we have added a couple of sentences to highlight your point here that skin cancer was the only outcome with clear risks outweighing benefits, and that melanoma mortality is lower for those with darker skin: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance” Finally, we have added reference to the recent sun exposure guidance in Australia: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. Response: We agree, there is scope for various future research projects – both utilising the data we have extracted here, as well as novel primary research. We have now increased our discussion of suggestions for future work, before the concluding paragraph. One follow-up study we recommend is to develop a common unit of exposure measurement that the current radiation-based measures could be converted to – this would allow, at least, for statistical synthesis of the results, and would likely open the door to other useful projects. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Response: Again, we agree that in principle the results presented in the review do not necessarily rule out the possibility that mortality is not a sufficient basis for sun safety advice. Although, the safety guidance is of course not only informed by mortality outcomes. There are many other outcomes that could, and should, be considered, such as incidence and quality of life. Unfortunately they were beyond the scope of the current review. However, as mentioned in reply to the point above, we have now increased discussion around future research ideas. As part of the suggested idea of converting results to a common unit of exposure, we mention that doing so could also prove useful for investigating other outcomes, such as incidence of disease. Minor point P9 - near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. Response: We thank the reviewers for highlighting a typo here – that should have read 1,000 kJ/m2. This has been corrected. In the case of all studies, we kept the units of measurement as reported in the paper.Many thanks to the reviewers for these helpful and insightful suggestions. We have attempted to address each point below.Competing Interests: No competing interests were disclosed. Close 1. None of the studies cited were able to directly compare UV exposure with mortality rates. UV dosimeters to measure personal exposure have been available for a relatively short time, and it in any case remains impractical to measure total doses over long periods of time, let alone lifetimes. As a results, the studies all use proxies for UV exposure. The best of these use ambient UV (measured or calculated), while others use measures as simple as latitude of residence as a proxy. These last are particularly troublesome given modern holiday habits where much of the annual exposure may occur in places at much lower latitudes. Consequently, in many if not most cases it was considered that there was a very high or high risk of bias. Response: We agree with the point raised here – no measures of exposure were direct and there were various issues with the proxy radiation measures. These issues are addressed in the discussion. However, our intention with this review was to be comprehensive and thorough in attempting to assess the effect of sunlight, therefore we felt it was beneficial to include these measures, though grouped separately in order to maintain that distinction. 2. The repeated statement that some of the papers are considered biased deserves more elaboration. Further info on how this bias is assessed would be useful for readers who aren’t familiar with the ‘ROBINS-E’ tool. In particular, we would find it helpful to have an impression of whether ROBINS-E primarily assesses authorial good practice, or fundamental bias; that is, if study 1 has a large unavoidable bias in which the authors did everything they could to mitigate it, and study 2 is probably minimally biased but has poor reporting and acknowledgement of potential confounders, which study would be flagged as high- or low- bias risk? This is particularly relevant in this context, since studies using (e.g.) latitude as a proxy for UV exposure will generally be more confounded/biased than (potential) studies using a direct measurement, even with authorial best practice. Response: The ROBINS-E tool assesses both authorial good practice and fundamental bias. We have added a brief description of the tool in the methods, addressing the point raised here as well as listing the seven domains the tool covers: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. None of the “all-cause mortality’ or ‘CVD Mortality’ results had bias risks lower than ‘High’. What is the difference between ‘High’ and ‘Very high’ risk? Of the three main categories, only the ‘All Cancer Mortality’ results had what look to be acceptably low bias risks. Our inclination would be to discount all studies for which the risk of bias is ‘High’ or ’Very high’. Perhaps the authors could attempt to address how that may change their conclusions? In particular (especially given point 2 above) are some types of studies in this area inevitably susceptible to bias? Response: Looking at the primary outcomes – as the reviewers point out, no study has a RoB assessment lower than high for all-cause and all-CVD. So discounting all ‘high’ and ‘very high’ RoB results would leave no results for these two primary outcomes. For all-cancer, this would leave three studies in the forest plot and three in the table – the results of which are similar to those currently presented (i.e., most results in the direction of a beneficial effect). Two papers were assessed to be of very high RoB. For the Yang paper, this was due to it being a proxy measure of sunlight (frequency of annual sunburns) that was self-reported and had the potential for differential measurement error (e.g., those who never sunburnt less likely to mis-report frequency than those regularly sunburnt). For the Grant paper (ref 57), it was due to missing data being an issue for both the outcome and confounders (they excluded states with <40 deaths from lung cancer – a proxy for smoking). 4. Nearly 40 percent of studies considered used proxies like latitude. These may be more valid for assessing non-skin-cancer risks, which may depend on lack of winter exposure because, whatever the behavioural advice, populations confined to mid and high latitudes in winter will never be able to access sufficient UV from sunlight to maintain optimal levels of vitamin D. However, the validity of latitude as a proxy is less obvious in the case of skin cancer. Our own research has shown that populations in both northern and southern New Zealand receive less than 5 percent of the available UV for most of the year (see here (1)). Given that small proportion, behavioural differences probably far outweigh effects of latitudinal gradients, especially in summer when the excess exposures probably occur. This should be mentioned. Response: We have increased the discussion around the applicability of the measures of exposure in the limitations section of the discussion. This includes mention of the issue you raise here regarding the lack of ambient UV received by individuals, as well as the validity of latitude when considering skin cancer. We also go on to discuss the difficulty of disentangling the effect of sunlight on the skin from the effect of associated confounders, given the lack of individual-level sunlight exposure measures: “This issue is further compounded for the proxy radiation measures, e.g., latitude. For example, a studies using personal dosimeters to investigate individual-level sun exposure in northern and southern latitudes in New Zealand showed that, regardless of latitude, people received less than 2% of ambient UVR exposure (Scragg et al, 2016 ). A finding which is echoed by a study in the UK showing that White people were exposed to ~2% ambient UVR, whilst those with darker skin types were exposed to even less, ~1% (Kift et al, 2013). As such, the validity of latitude as a proxy measure of radiation, especially when considering skin cancer, may be limited and results should be interpreted with caution. Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 5. While the collective studies do not provide strong evidence for a non-zero effect of UV exposure on mortality, CVD mortality, etc., they do provide moderate evidence for the absence of a large effect. The authors may wish to mention this. Response: We added the sentence “Overall, the current available evidence does not indicate a large effect of sunlight on mortality” to the conclusion. 6. Further to this, although the overall risks from sun exposure clearly outweigh the benefits in the case of melanoma, that does not appear to be true for other cancers or CVD. The regression studies (e.g., Table 1) report a (perhaps troublingly) large number of studies (particularly by Grant -re vitamin D) showing benefits from sun exposure – even for melanoma in a few of the studies. For other cancers, most notably breast cancer, sunlight exposure may be beneficial. But for all cancers and for CVD, it is not easy to see whether the benefits outweigh the risks. Given the much lower overall mortality rate from melanoma compared with that from CVD and all other cancers, it is indeed worth considering whether health messaging should be changed. Our main concern is that, if current guidelines do not stratify advice by skin colour, we feel that, on the basis of these results, they should do so. Given the low incidence of melanoma among dark skinned people, the relatively low levels of UV in the UK, and the larger fraction of the population who now fit into that category, it’s not clear from the present analysis whether total mortality burden from excessive sunlight exceeds that from insufficient exposure to sunlight. We do recognize the difficulty of changing public health messaging, and that a full analysis of information relevant to deciding policies is beyond scope, but the authors may wish to mention this. Response: We have increased discussion on the issues surrounding skin type/ethnicity and the lack of available data on this. We have added a new paragraph discussing the potential interaction between location and skin type (i.e., in low-UV countries, sunlight may benefit those with lighter skin, whilst in high-UV countries those with darker skin might benefit): “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Later, we have added a couple of sentences to highlight your point here that skin cancer was the only outcome with clear risks outweighing benefits, and that melanoma mortality is lower for those with darker skin: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance” Finally, we have added reference to the recent sun exposure guidance in Australia: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. We also note that some countries already advise supplementation of vitamin D over winter months. By adopting such a strategy, there’s less need to consider changing public health messaging of “minimizing exposure to sunlight”. However, as the authors remind us, deficiency in vitamin D is not the only winter issue. Low UV may affect other pathways as well in ways that aren’t yet fully understood. The authors also remind us that there’s little solid evidence for an association between vitamin-D status and mortality from cancer and CVD. Given the health implications raised here and the relative importance of the various forms of cancer and CVD for different skin type, we feel that the statement that ‘no change in health messaging is justified’ may be premature. Perhaps it would be better to recommend a follow-up study using the same data and attempt to estimate whether the overall burden of death from cancer and CVD increases or decreases as a function of sun exposure, and to assess whether the conclusion for white skinned people is the same as that for those with pigmented skins. Such a follow-up study should probably ignore studies for which the risk bias is high or very high. Response: We agree, there is scope for various future research projects – both utilising the data we have extracted here, as well as novel primary research. We have now increased our discussion of suggestions for future work, before the concluding paragraph. One follow-up study we recommend is to develop a common unit of exposure measurement that the current radiation-based measures could be converted to – this would allow, at least, for statistical synthesis of the results, and would likely open the door to other useful projects. 8. The question motivating this study, “Should health advice on sun exposure be changed?”, indeed seems to have no clear answer from the effects on years of lost life. By the same token, the question might have been, “Do studies of YLL , at least from the dominant causes of heart failure and cancer, inform what advice should be given on sun exposure?” That question would also be answered in the negative, but with the opposite effect. Some discussion of this might be warranted. Another question that might be addressed is whether changed mortality is on its own a sufficient basis for the advice. Many immune disorders are affected, in incidence or progression, by sun exposure; type 1 diabetes, multiple sclerosis, rheumatoid arthritis, coeliac disease, Graves’ disease, psoriasis and psoriatic arthritis, IBD, etc. Even if they do not result in early death, any of these can have a profound impact on quality of life. Using a measure like disability-adjusted life-years or health-adjusted life-years would presumably entail as much work again, and presumably another paper, but the possibility of doing so might usefully be raised here. Response: Again, we agree that in principle the results presented in the review do not necessarily rule out the possibility that mortality is not a sufficient basis for sun safety advice. Although, the safety guidance is of course not only informed by mortality outcomes. There are many other outcomes that could, and should, be considered, such as incidence and quality of life. Unfortunately they were beyond the scope of the current review. However, as mentioned in reply to the point above, we have now increased discussion around future research ideas. As part of the suggested idea of converting results to a common unit of exposure, we mention that doing so could also prove useful for investigating other outcomes, such as incidence of disease. Minor point P9 - near top. There may be confusion about units for radiation measurements in the cited papers. It’s reported that an increase of 100 kJ m-2 of solar radiation is associated with a 1% increased risk. Is that over a lifetime, or a year, or a day? In any of those cases, a measurable effect seems unlikely because in a single day, the total dose from sunlight can be over 20,000 kJ m-2, so the reported excess is less than 0.5% of peak values (or about 1% for peak winter values). Perhaps they meant UVB for which daily doses would be smaller by at least a factor of 500 (with peak noon values around 2 W m-2), or erythemal UV which would be smaller by a factor of at least 3500 (with peak noon values around 0.3 W m-2, UVI =12). Even for those quantities, a causative dose increase of 100 J m-2 seems rather small, given that erythema doses of 70 SED (7000 J m-2 can occur in a single day. The results of Chen later on the same page, which report increases in daily average UVB look overly sensitive given those noon values quoted above. It’s also puzzling to see that increased dose reported in average watts rather than joules. It would be more useful the percentage increase risks were provided as function of the percentage increase in UV exposure. Response: We thank the reviewers for highlighting a typo here – that should have read 1,000 kJ/m2. This has been corrected. In the case of all studies, we kept the units of measurement as reported in the paper. Views 0 How to cite this report: Heckman C. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36115) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36115 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36115 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 30 Aug 2025 Not Approved VIEWS 0 Important and timely topic and well-written paper. Title: Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. ... Continue reading I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above. Close Title: Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. ... Continue reading Important and timely topic and well-written paper. Title: Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample.

Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Lay Summary Intro

Results: p. 12, right paragraph – There seems to be an error in the reporting of the direction of the relationships.

Title: Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample.

Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Lay Summary - Could be clarified. The first two sentences make it seem that burns rather than UV exposure cause all skin cancers. - Sunscreen is not mentioned. - I think something along the lines of “limits of available data” would be more accurate than “problems in how the studies were done.” - A reminder at the end about sun protection would seem appropriate. Intro - Recommendations from the US National Cancer Institute might be more comparable to the other sources you note as opposed to the AAD. - I would include something about sunburns in childhood increasing risk for BCC too.

- If no randomized trials were included, I would not mention that design. If they were, please report something about them, specifically. - Summarize the bias criteria. - Briefly describe the GRADE domains. - For studies that didn’t report race or ethnicity: Would it be appropriate to consider the race and ethnic composition of some of the study countries during the time(s) of data collection? E.g., studies conducted in China likely included mostly Asian participants. Or to consider findings for whites vs. non-whites, though admittedly less than ideal. - Would it be appropriate to weight the findings based on sample size, level of bias, null findings, and/or something else?

p. 12, right paragraph – There seems to be an error in the reporting of the direction of the relationships.

- There seem to be contradictory statements in the first paragraph. - The US population of Native Americans is quite small, and reporting for that group is notoriously poor. Those findings should be considered with caution. - Expand on the bias issues. What are the major confounders? One would be physical activity. Assuming that was considered in some studies, it might be worth including as an additional reporting variable. - The review separated studies by measure of UV, but as mentioned, risk is determined by available UV (e.g., latitude) and exposure to UV (e.g., behavior). Can these types of measures be combined somehow for studies in a similar/same place/time? - Wondering if it would be useful to include the tables/figures on specific cancers (other than skin) as supplements? - It would seem that the pancreatic cancer findings could be emphasized since they were mostly consistent. - The recent guidelines by Australians regarding tailored recommendations should be discussed in the Intro and/or Discussion and why those may be valid or not as compared to the current review. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? I cannot comment. A qualified statistician is required. - Are the conclusions drawn adequately supported by the results presented in the review? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: Behavioral skin cancer prevention. My only statistical concern was how the male and female data were combined in some cases. CITE HOW TO CITE THIS REPORT Heckman C. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36115) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36115 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36115 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below. Title - Seeing the variety of types ... Continue reading Many thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below. Title - Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. Response: We understand your point, though as sunlight exposure was the underlying effect we were interested in, and we have defined what this term encompasses, we feel that this is an appropriate title. Abst - Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Response: Thank you. We have amended both the abstract and Fig 1 to make it clear that 55 articles were included in the narrative synthesis. Lay Summary 1. Could be clarified. The first two sentences make it seem that burns rather than UV exposure cause all skin cancers. Response: We have edited this sentence to make it more about UV exposure, whilst keeping it appropriate for a PLS: “Sunlight exposure of the skin can lead to the development of skin cancer” 2. Sunscreen is not mentioned. Response: We have now included sunscreen: “Organizations providing advice on sun exposure say to wear sunscreen and covering clothing, and to stay in the shade between 11am and 3pm.” 3. I think something along the lines of “limits of available data” would be more accurate than “problems in how the studies were done.” Response: We feel that limits of available data is a slightly different issue to the one we were referring to here (essentially RoB concerns). However, we have changed the wording to include the issue of data availability and to hopefully clarify the point: “However, there were issues with the amount of data available, as well as the quality of some of the data that was available…” 4. A reminder at the end about sun protection would seem appropriate. Response: We have added to the final sentence to include this: “Currently, there is not strong enough evidence to alter sun exposure advice and so people should continue to follow the guidance.” Intro 1. Recommendations from the US National Cancer Institute might be more comparable to the other sources you note as opposed to the AAD. Response: Thank you. We have changed this reference from the AAD to the NCI. 2. I would include something about sunburns in childhood increasing risk for BCC too. Response: We have added the statistic about childhood sunburns from the same review: “the risk of developing basal cell carcinoma (BCC), the most common form of non-melanoma skin cancer (NMSC), is 1.86 and 2.12 times greater with every five sunburns experienced as a child and as an adult, respectively”

1. If no randomized trials were included, I would not mention that design. If they were, please report something about them, specifically. Response: No RCTs were included but we did search for them. We feel it is important to highlight that our search terms included RCTs, even if we failed to find any that met the inclusion criteria. Therefore, in the section of the results discussing the characteristics of the included studies, we have added the sentence “We did not identify any eligible randomized controlled trials.” 2. Summarize the bias criteria. Response: We have provided some more information regarding the domains assessed with ROBINS-E: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. Briefly describe the GRADE domains. Response: We have provided a brief description of the five GRADE domains: “The GRADE domains assess the risk of bias in the included studies (risk of bias); how consistent the results are within each outcome (inconsistency); how applicable the evidence is to the review question (indirectness); how precise the estimates are (imprecision); and the possibility of selective publication of results (publication bias).” 4. For studies that didn’t report race or ethnicity: Would it be appropriate to consider the race and ethnic composition of some of the study countries during the time(s) of data collection? E.g., studies conducted in China likely included mostly Asian participants. Or to consider findings for whites vs. non-whites, though admittedly less than ideal. Response: Unfortunately, there was a lack of reporting of individual data regarding skin type/colour or ethnicity across the population level studies. Whilst we could use separate demographic information for countries/regions at the time of the study to infer race/ethnicity for some of the ecological studies, this is unfortunately beyond the scope of the current review and may not always be appropriate to perform as it would require broad assumptions. It also wouldn’t give us individual data, and therefore would not be appropriate to apply to case-control and cohort studies. 5. Would it be appropriate to weight the findings based on sample size, level of bias, null findings, and/or something else? Response: As there is no statistical synthesis, any weighting would be challenging. However, we have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes in order to assist in interpretation. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140).

p.12 - right paragraph – There seems to be an error in the reporting of the direction of the relationships. Response: We agree, this reporting was incorrect (and confusingly worded) – thank you for pointing this out. The text has now been corrected and hopefully the wording is clearer: “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.”

1. There seem to be contradictory statements in the first paragraph. Response: We have adapted the first sentence so that it now reads “a mixed message about the association between sunlight exposure and all-cause mortality risk.” 2. The US population of Native Americans is quite small, and reporting for that group is notoriously poor. Those findings should be considered with caution. Response: We have added a note of caution to this result: “Although this result should be treated with caution as the Native American population is relatively small.” 3. Expand on the bias issues. What are the major confounders? One would be physical activity. Assuming that was considered in some studies, it might be worth including as an additional reporting variable. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, addressing the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 4. The review separated studies by measure of UV, but as mentioned, risk is determined by available UV (e.g., latitude) and exposure to UV (e.g., behavior). Can these types of measures be combined somehow for studies in a similar/same place/time? Response: Although this would be an interesting extension to the findings, unfortunately there is insufficient individual data available to do this currently. 5. Wondering if it would be useful to include the tables/figures on specific cancers (other than skin) as supplements? Response: NIHR Open requires all information that is necessary to follow the methods and results of the review to be included in the paper. Therefore, we are unable to send these to the supplement. 6. It would seem that the pancreatic cancer findings could be emphasized since they were mostly consistent. Response: We have now added a sentence in the discussion to highlight the pancreatic cancer results: “In contrast, most articles examining the five cancers with the highest UK mortality rate (breast, prostate, lung, bowel and pancreatic cancer) found that higher levels of sunlight were associated with lower risks of mortality. In particular, for pancreatic cancer, where the available evidence was somewhat consistent in suggesting a beneficial effect of sunlight.” 7. The recent guidelines by Australians regarding tailored recommendations should be discussed in the Intro and/or Discussion and why those may be valid or not as compared to the current review. Response: We have added reference to this guidance in the discussion, in relation to the paragraph addressing the lack of available evidence on skin type/race and the need for more tailored guidance: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance. However, approximately a quarter of the main articles included in this review were restricted to White populations. Around two thirds reported on the whole population, though given that a large number of these were conducted in Europe and North America, it is likely that the populations in those articles were predominantly White as well. In order to gain a more complete picture of the relationship between sunlight exposure and mortality, further studies investigating the impact of skin type/colour or ethnicity are warranted. This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).”Many thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below.Competing Interests: No competing interests were disclosed. Close Title - Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. Response: We understand your point, though as sunlight exposure was the underlying effect we were interested in, and we have defined what this term encompasses, we feel that this is an appropriate title. Abst - Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Response: Thank you. We have amended both the abstract and Fig 1 to make it clear that 55 articles were included in the narrative synthesis. Lay Summary 1. Could be clarified. The first two sentences make it seem that burns rather than UV exposure cause all skin cancers. Response: We have edited this sentence to make it more about UV exposure, whilst keeping it appropriate for a PLS: “Sunlight exposure of the skin can lead to the development of skin cancer” 2. Sunscreen is not mentioned. Response: We have now included sunscreen: “Organizations providing advice on sun exposure say to wear sunscreen and covering clothing, and to stay in the shade between 11am and 3pm.” 3. I think something along the lines of “limits of available data” would be more accurate than “problems in how the studies were done.” Response: We feel that limits of available data is a slightly different issue to the one we were referring to here (essentially RoB concerns). However, we have changed the wording to include the issue of data availability and to hopefully clarify the point: “However, there were issues with the amount of data available, as well as the quality of some of the data that was available…” 4. A reminder at the end about sun protection would seem appropriate. Response: We have added to the final sentence to include this: “Currently, there is not strong enough evidence to alter sun exposure advice and so people should continue to follow the guidance.” Intro 1. Recommendations from the US National Cancer Institute might be more comparable to the other sources you note as opposed to the AAD. Response: Thank you. We have changed this reference from the AAD to the NCI. 2. I would include something about sunburns in childhood increasing risk for BCC too. Response: We have added the statistic about childhood sunburns from the same review: “the risk of developing basal cell carcinoma (BCC), the most common form of non-melanoma skin cancer (NMSC), is 1.86 and 2.12 times greater with every five sunburns experienced as a child and as an adult, respectively”

1. If no randomized trials were included, I would not mention that design. If they were, please report something about them, specifically. Response: No RCTs were included but we did search for them. We feel it is important to highlight that our search terms included RCTs, even if we failed to find any that met the inclusion criteria. Therefore, in the section of the results discussing the characteristics of the included studies, we have added the sentence “We did not identify any eligible randomized controlled trials.” 2. Summarize the bias criteria. Response: We have provided some more information regarding the domains assessed with ROBINS-E: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. Briefly describe the GRADE domains. Response: We have provided a brief description of the five GRADE domains: “The GRADE domains assess the risk of bias in the included studies (risk of bias); how consistent the results are within each outcome (inconsistency); how applicable the evidence is to the review question (indirectness); how precise the estimates are (imprecision); and the possibility of selective publication of results (publication bias).” 4. For studies that didn’t report race or ethnicity: Would it be appropriate to consider the race and ethnic composition of some of the study countries during the time(s) of data collection? E.g., studies conducted in China likely included mostly Asian participants. Or to consider findings for whites vs. non-whites, though admittedly less than ideal. Response: Unfortunately, there was a lack of reporting of individual data regarding skin type/colour or ethnicity across the population level studies. Whilst we could use separate demographic information for countries/regions at the time of the study to infer race/ethnicity for some of the ecological studies, this is unfortunately beyond the scope of the current review and may not always be appropriate to perform as it would require broad assumptions. It also wouldn’t give us individual data, and therefore would not be appropriate to apply to case-control and cohort studies. 5. Would it be appropriate to weight the findings based on sample size, level of bias, null findings, and/or something else? Response: As there is no statistical synthesis, any weighting would be challenging. However, we have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes in order to assist in interpretation. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140).

p.12 - right paragraph – There seems to be an error in the reporting of the direction of the relationships. Response: We agree, this reporting was incorrect (and confusingly worded) – thank you for pointing this out. The text has now been corrected and hopefully the wording is clearer: “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.”

1. There seem to be contradictory statements in the first paragraph. Response: We have adapted the first sentence so that it now reads “a mixed message about the association between sunlight exposure and all-cause mortality risk.” 2. The US population of Native Americans is quite small, and reporting for that group is notoriously poor. Those findings should be considered with caution. Response: We have added a note of caution to this result: “Although this result should be treated with caution as the Native American population is relatively small.” 3. Expand on the bias issues. What are the major confounders? One would be physical activity. Assuming that was considered in some studies, it might be worth including as an additional reporting variable. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, addressing the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 4. The review separated studies by measure of UV, but as mentioned, risk is determined by available UV (e.g., latitude) and exposure to UV (e.g., behavior). Can these types of measures be combined somehow for studies in a similar/same place/time? Response: Although this would be an interesting extension to the findings, unfortunately there is insufficient individual data available to do this currently. 5. Wondering if it would be useful to include the tables/figures on specific cancers (other than skin) as supplements? Response: NIHR Open requires all information that is necessary to follow the methods and results of the review to be included in the paper. Therefore, we are unable to send these to the supplement. 6. It would seem that the pancreatic cancer findings could be emphasized since they were mostly consistent. Response: We have now added a sentence in the discussion to highlight the pancreatic cancer results: “In contrast, most articles examining the five cancers with the highest UK mortality rate (breast, prostate, lung, bowel and pancreatic cancer) found that higher levels of sunlight were associated with lower risks of mortality. In particular, for pancreatic cancer, where the available evidence was somewhat consistent in suggesting a beneficial effect of sunlight.” 7. The recent guidelines by Australians regarding tailored recommendations should be discussed in the Intro and/or Discussion and why those may be valid or not as compared to the current review. Response: We have added reference to this guidance in the discussion, in relation to the paragraph addressing the lack of available evidence on skin type/race and the need for more tailored guidance: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance. However, approximately a quarter of the main articles included in this review were restricted to White populations. Around two thirds reported on the whole population, though given that a large number of these were conducted in Europe and North America, it is likely that the populations in those articles were predominantly White as well. In order to gain a more complete picture of the relationship between sunlight exposure and mortality, further studies investigating the impact of skin type/colour or ethnicity are warranted. This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” COMMENTS ON THIS REPORT - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below. Title - Seeing the variety of types ... Continue reading Many thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below. Title - Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. Response: We understand your point, though as sunlight exposure was the underlying effect we were interested in, and we have defined what this term encompasses, we feel that this is an appropriate title. Abst - Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Response: Thank you. We have amended both the abstract and Fig 1 to make it clear that 55 articles were included in the narrative synthesis. Lay Summary 1. Could be clarified. The first two sentences make it seem that burns rather than UV exposure cause all skin cancers. Response: We have edited this sentence to make it more about UV exposure, whilst keeping it appropriate for a PLS: “Sunlight exposure of the skin can lead to the development of skin cancer” 2. Sunscreen is not mentioned. Response: We have now included sunscreen: “Organizations providing advice on sun exposure say to wear sunscreen and covering clothing, and to stay in the shade between 11am and 3pm.” 3. I think something along the lines of “limits of available data” would be more accurate than “problems in how the studies were done.” Response: We feel that limits of available data is a slightly different issue to the one we were referring to here (essentially RoB concerns). However, we have changed the wording to include the issue of data availability and to hopefully clarify the point: “However, there were issues with the amount of data available, as well as the quality of some of the data that was available…” 4. A reminder at the end about sun protection would seem appropriate. Response: We have added to the final sentence to include this: “Currently, there is not strong enough evidence to alter sun exposure advice and so people should continue to follow the guidance.” Intro 1. Recommendations from the US National Cancer Institute might be more comparable to the other sources you note as opposed to the AAD. Response: Thank you. We have changed this reference from the AAD to the NCI. 2. I would include something about sunburns in childhood increasing risk for BCC too. Response: We have added the statistic about childhood sunburns from the same review: “the risk of developing basal cell carcinoma (BCC), the most common form of non-melanoma skin cancer (NMSC), is 1.86 and 2.12 times greater with every five sunburns experienced as a child and as an adult, respectively”

1. If no randomized trials were included, I would not mention that design. If they were, please report something about them, specifically. Response: No RCTs were included but we did search for them. We feel it is important to highlight that our search terms included RCTs, even if we failed to find any that met the inclusion criteria. Therefore, in the section of the results discussing the characteristics of the included studies, we have added the sentence “We did not identify any eligible randomized controlled trials.” 2. Summarize the bias criteria. Response: We have provided some more information regarding the domains assessed with ROBINS-E: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. Briefly describe the GRADE domains. Response: We have provided a brief description of the five GRADE domains: “The GRADE domains assess the risk of bias in the included studies (risk of bias); how consistent the results are within each outcome (inconsistency); how applicable the evidence is to the review question (indirectness); how precise the estimates are (imprecision); and the possibility of selective publication of results (publication bias).” 4. For studies that didn’t report race or ethnicity: Would it be appropriate to consider the race and ethnic composition of some of the study countries during the time(s) of data collection? E.g., studies conducted in China likely included mostly Asian participants. Or to consider findings for whites vs. non-whites, though admittedly less than ideal. Response: Unfortunately, there was a lack of reporting of individual data regarding skin type/colour or ethnicity across the population level studies. Whilst we could use separate demographic information for countries/regions at the time of the study to infer race/ethnicity for some of the ecological studies, this is unfortunately beyond the scope of the current review and may not always be appropriate to perform as it would require broad assumptions. It also wouldn’t give us individual data, and therefore would not be appropriate to apply to case-control and cohort studies. 5. Would it be appropriate to weight the findings based on sample size, level of bias, null findings, and/or something else? Response: As there is no statistical synthesis, any weighting would be challenging. However, we have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes in order to assist in interpretation. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140).

p.12 - right paragraph – There seems to be an error in the reporting of the direction of the relationships. Response: We agree, this reporting was incorrect (and confusingly worded) – thank you for pointing this out. The text has now been corrected and hopefully the wording is clearer: “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.”

1. There seem to be contradictory statements in the first paragraph. Response: We have adapted the first sentence so that it now reads “a mixed message about the association between sunlight exposure and all-cause mortality risk.” 2. The US population of Native Americans is quite small, and reporting for that group is notoriously poor. Those findings should be considered with caution. Response: We have added a note of caution to this result: “Although this result should be treated with caution as the Native American population is relatively small.” 3. Expand on the bias issues. What are the major confounders? One would be physical activity. Assuming that was considered in some studies, it might be worth including as an additional reporting variable. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, addressing the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 4. The review separated studies by measure of UV, but as mentioned, risk is determined by available UV (e.g., latitude) and exposure to UV (e.g., behavior). Can these types of measures be combined somehow for studies in a similar/same place/time? Response: Although this would be an interesting extension to the findings, unfortunately there is insufficient individual data available to do this currently. 5. Wondering if it would be useful to include the tables/figures on specific cancers (other than skin) as supplements? Response: NIHR Open requires all information that is necessary to follow the methods and results of the review to be included in the paper. Therefore, we are unable to send these to the supplement. 6. It would seem that the pancreatic cancer findings could be emphasized since they were mostly consistent. Response: We have now added a sentence in the discussion to highlight the pancreatic cancer results: “In contrast, most articles examining the five cancers with the highest UK mortality rate (breast, prostate, lung, bowel and pancreatic cancer) found that higher levels of sunlight were associated with lower risks of mortality. In particular, for pancreatic cancer, where the available evidence was somewhat consistent in suggesting a beneficial effect of sunlight.” 7. The recent guidelines by Australians regarding tailored recommendations should be discussed in the Intro and/or Discussion and why those may be valid or not as compared to the current review. Response: We have added reference to this guidance in the discussion, in relation to the paragraph addressing the lack of available evidence on skin type/race and the need for more tailored guidance: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance. However, approximately a quarter of the main articles included in this review were restricted to White populations. Around two thirds reported on the whole population, though given that a large number of these were conducted in Europe and North America, it is likely that the populations in those articles were predominantly White as well. In order to gain a more complete picture of the relationship between sunlight exposure and mortality, further studies investigating the impact of skin type/colour or ethnicity are warranted. This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).”Many thanks for this thorough, helpful review of our paper. We have attempted to address your points and respond to each below.Competing Interests: No competing interests were disclosed. Close Title - Seeing the variety of types of exposures, I’m wondering whether “sunlight exposure” is the most appropriate term. You could also add something about the international sample. Response: We understand your point, though as sunlight exposure was the underlying effect we were interested in, and we have defined what this term encompasses, we feel that this is an appropriate title. Abst - Based on the Results, I suggest changing the final number of the articles in the Abstract and Figure 1 to 55. Response: Thank you. We have amended both the abstract and Fig 1 to make it clear that 55 articles were included in the narrative synthesis. Lay Summary 1. Could be clarified. The first two sentences make it seem that burns rather than UV exposure cause all skin cancers. Response: We have edited this sentence to make it more about UV exposure, whilst keeping it appropriate for a PLS: “Sunlight exposure of the skin can lead to the development of skin cancer” 2. Sunscreen is not mentioned. Response: We have now included sunscreen: “Organizations providing advice on sun exposure say to wear sunscreen and covering clothing, and to stay in the shade between 11am and 3pm.” 3. I think something along the lines of “limits of available data” would be more accurate than “problems in how the studies were done.” Response: We feel that limits of available data is a slightly different issue to the one we were referring to here (essentially RoB concerns). However, we have changed the wording to include the issue of data availability and to hopefully clarify the point: “However, there were issues with the amount of data available, as well as the quality of some of the data that was available…” 4. A reminder at the end about sun protection would seem appropriate. Response: We have added to the final sentence to include this: “Currently, there is not strong enough evidence to alter sun exposure advice and so people should continue to follow the guidance.” Intro 1. Recommendations from the US National Cancer Institute might be more comparable to the other sources you note as opposed to the AAD. Response: Thank you. We have changed this reference from the AAD to the NCI. 2. I would include something about sunburns in childhood increasing risk for BCC too. Response: We have added the statistic about childhood sunburns from the same review: “the risk of developing basal cell carcinoma (BCC), the most common form of non-melanoma skin cancer (NMSC), is 1.86 and 2.12 times greater with every five sunburns experienced as a child and as an adult, respectively”

1. If no randomized trials were included, I would not mention that design. If they were, please report something about them, specifically. Response: No RCTs were included but we did search for them. We feel it is important to highlight that our search terms included RCTs, even if we failed to find any that met the inclusion criteria. Therefore, in the section of the results discussing the characteristics of the included studies, we have added the sentence “We did not identify any eligible randomized controlled trials.” 2. Summarize the bias criteria. Response: We have provided some more information regarding the domains assessed with ROBINS-E: “ROBINS-E is a tool designed to evaluate the risk of bias in results of observational epidemiologic studies, covering aspects of design, conduct and reporting of the study. It assesses risk of bias across seven domains: confounding, measurement of the exposure, selection of participants, post-exposure interventions, missing data, measurement of the outcome and selection of the reported result.” 3. Briefly describe the GRADE domains. Response: We have provided a brief description of the five GRADE domains: “The GRADE domains assess the risk of bias in the included studies (risk of bias); how consistent the results are within each outcome (inconsistency); how applicable the evidence is to the review question (indirectness); how precise the estimates are (imprecision); and the possibility of selective publication of results (publication bias).” 4. For studies that didn’t report race or ethnicity: Would it be appropriate to consider the race and ethnic composition of some of the study countries during the time(s) of data collection? E.g., studies conducted in China likely included mostly Asian participants. Or to consider findings for whites vs. non-whites, though admittedly less than ideal. Response: Unfortunately, there was a lack of reporting of individual data regarding skin type/colour or ethnicity across the population level studies. Whilst we could use separate demographic information for countries/regions at the time of the study to infer race/ethnicity for some of the ecological studies, this is unfortunately beyond the scope of the current review and may not always be appropriate to perform as it would require broad assumptions. It also wouldn’t give us individual data, and therefore would not be appropriate to apply to case-control and cohort studies. 5. Would it be appropriate to weight the findings based on sample size, level of bias, null findings, and/or something else? Response: As there is no statistical synthesis, any weighting would be challenging. However, we have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes in order to assist in interpretation. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140).

p.12 - right paragraph – There seems to be an error in the reporting of the direction of the relationships. Response: We agree, this reporting was incorrect (and confusingly worded) – thank you for pointing this out. The text has now been corrected and hopefully the wording is clearer: “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.”

1. There seem to be contradictory statements in the first paragraph. Response: We have adapted the first sentence so that it now reads “a mixed message about the association between sunlight exposure and all-cause mortality risk.” 2. The US population of Native Americans is quite small, and reporting for that group is notoriously poor. Those findings should be considered with caution. Response: We have added a note of caution to this result: “Although this result should be treated with caution as the Native American population is relatively small.” 3. Expand on the bias issues. What are the major confounders? One would be physical activity. Assuming that was considered in some studies, it might be worth including as an additional reporting variable. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, addressing the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 4. The review separated studies by measure of UV, but as mentioned, risk is determined by available UV (e.g., latitude) and exposure to UV (e.g., behavior). Can these types of measures be combined somehow for studies in a similar/same place/time? Response: Although this would be an interesting extension to the findings, unfortunately there is insufficient individual data available to do this currently. 5. Wondering if it would be useful to include the tables/figures on specific cancers (other than skin) as supplements? Response: NIHR Open requires all information that is necessary to follow the methods and results of the review to be included in the paper. Therefore, we are unable to send these to the supplement. 6. It would seem that the pancreatic cancer findings could be emphasized since they were mostly consistent. Response: We have now added a sentence in the discussion to highlight the pancreatic cancer results: “In contrast, most articles examining the five cancers with the highest UK mortality rate (breast, prostate, lung, bowel and pancreatic cancer) found that higher levels of sunlight were associated with lower risks of mortality. In particular, for pancreatic cancer, where the available evidence was somewhat consistent in suggesting a beneficial effect of sunlight.” 7. The recent guidelines by Australians regarding tailored recommendations should be discussed in the Intro and/or Discussion and why those may be valid or not as compared to the current review. Response: We have added reference to this guidance in the discussion, in relation to the paragraph addressing the lack of available evidence on skin type/race and the need for more tailored guidance: “Skin cancer mortality was the only outcome for which the evidence clearly suggested that the risks of sunlight exposure outweighed the benefits. Given that melanoma mortality is known to be lower in those with darker skin (Lopes, et al, 2021), this highlights the need for nuanced sun exposure guidance. However, approximately a quarter of the main articles included in this review were restricted to White populations. Around two thirds reported on the whole population, though given that a large number of these were conducted in Europe and North America, it is likely that the populations in those articles were predominantly White as well. In order to gain a more complete picture of the relationship between sunlight exposure and mortality, further studies investigating the impact of skin type/colour or ethnicity are warranted. This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” Views 0 How to cite this report: Brash D. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36173) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36173 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36173 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 29 Aug 2025 Douglas Brash, Yale School of Medicine, New Haven, USA Approved with Reservations VIEWS 0 Summary: The authors used defined criteria to conduct an extensive search of the literature for epidemiological studies relating sunlight exposure to all-causes mortality and to more specific diesease endpoints. The results from these studies are then translated to a common ... Continue reading I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Close Summary: The authors used defined criteria to conduct an extensive search of the literature for epidemiological studies relating sunlight exposure to all-causes mortality and to more specific diesease endpoints. The results from these studies are then translated to a common effect metric and compared for each endpoint. The authors find that many studies indicate a beneficial effect of sunlight exposure for one endpoint or another, while other studies do not. The authors conclude that there is insufficient evidence to advise a change to current government and professional society advice on cautious sun-exposure. Critique: This prodigious effort addresses an important question raised by evidence that sunlight has various beneficial effects such as producing vitamin D and nitric oxide: How to balance these benefits against the fact that sunlight causes skin cancer, and balance against the pervasive advice to stay out of the sun and use sunscreen? An indicator of this need is that I and a colleague separately tried to get the American Academy of Dermatology to provide the references on which they based their sun protection recommendations, to no avail. The present report appears carefully done (I leave technical comments on the statistical methods and statistical-bias evaluations to other reviewers), and its summary of the literature is balanced rather than partisan. In using a Cochrane-like method, it progresses beyond earlier summaries such as Hoel et al, Dermato-Endocrinology 2016. I concur that the diversity in measurements used by the subject papers makes meta-analysis a fraught approach and I'm okay with the narrative approach plus Forest plots. The paper is therefore worthy of publication and NIHR Open Research is a suitable venue. I do have several comments and corrections, generally along the lines of the data not being as null as the authors state. Science comments: 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Writing comments: p.5 "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? p.6 "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. p.9 All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". p.12 "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. Critique: This prodigious effort addresses an important question raised by evidence that sunlight has various beneficial effects such as producing vitamin D and nitric oxide: How to balance these benefits against the fact that sunlight causes skin cancer, and balance against the pervasive advice to stay out of the sun and use sunscreen? An indicator of this need is that I and a colleague separately tried to get the American Academy of Dermatology to provide the references on which they based their sun protection recommendations, to no avail. The present report appears carefully done (I leave technical comments on the statistical methods and statistical-bias evaluations to other reviewers), and its summary of the literature is balanced rather than partisan. In using a Cochrane-like method, it progresses beyond earlier summaries such as Hoel et al, Dermato-Endocrinology 2016. I concur that the diversity in measurements used by the subject papers makes meta-analysis a fraught approach and I'm okay with the narrative approach plus Forest plots. The paper is therefore worthy of publication and NIHR Open Research is a suitable venue. I do have several comments and corrections, generally along the lines of the data not being as null as the authors state. Science comments: 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Writing comments: p.5 "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? p.6 "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. p.9 All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". p.12 "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? I cannot comment. A qualified statistician is required. - Are the conclusions drawn adequately supported by the results presented in the review? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: Biophysics, molecular genetics, and cell biology of sunlight-induced skin cancer; photobiology. CITE HOW TO CITE THIS REPORT Brash D. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36173) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36173 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36173 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below. ... Continue reading Many thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below. 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. Response: Thank you for pointing out this oversight. We have now redrawn all forest plots using the exponential scale on the x-axis in order to accurately represent the beneficial effects. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. Response: We agree with you that Lin et al inappropriately refer to “biological damage (BD/m2)”. The definition they provide is unclear and the use of the term is confusing; no biological damage has actually been measured. What it appears they have done is to obtain average July UVR data via TOMS in order to estimate individual levels of UV erythemal exposure, which in turn is divided into quartiles. As such, we believe it is correctly categorised as a ‘radiation’ measure. To avoid confusion, we have decided to remove mention of the units BD/m2 and instead labelled/referred to this exposure measure as quartiles of “Average July erythemal UVR.” 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. Response: Thank you for these carefully considered and insightful suggestions for further analyses. We have now added a paragraph in the discussion addressing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Whilst we agree that the suggestions you make for further analyses would make valuable contribution to the paper, unfortunately we don’t believe we have sufficient data to fully address suggestion a) at this stage. Given the lack of reporting on skin type/colour or ethnicity in the included studies, we don’t have the individual data required. As such, we would have to incorporate further population level data to map against each study – which is possible but comes with its own challenges and complications that are beyond the scope of this project currently. This is also the case for suggestion b), which may be more feasible to implement from the available data but would require a good deal of additional investigation. However, we believe these suggestions are absolutely worth exploring and as such have incorporated them into our suggestions for future research: “Firstly, there is a clear need for more research into the effects of sunlight on people with darker skin. The majority of studies were conducted in predominantly White populations, with many specifically limiting inclusion to White people only. Moreover, skin type/colour or ethnicity data ought to be more commonly reported in population level studies. This would allow for more insightful analyses of the available data. For example, allowing for further investigating into the hypothesis that those with darker skin benefit most from sunlight in high UVR areas.” 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". Response: We have changed the term ‘inconclusive’ to ‘mixed’ which, as pointed out in the comment here, more accurately reflects the findings laid out in the following section of the abstract. We have also adapted the conclusion of the abstract to read “Findings from observational epidemiological studies of the association between sunlight exposure and mortality vary across different disease outcome and location being investigated. As such, the findings do not provide a strong rationale for changes to sun protection guidance.” 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Response: We found this review in our initial scoping. Whilst it was a useful and relevant paper, we did not prioritise investigating their reference list as it was not a systematic review. However, we have since checked and do not believe we missed any studies that ought to have been included. Most were identified in our searches, some of which were included. Those that were excluded were most commonly either not measuring mortality; conducted on people with pre-existing disease; or were vitamin D studies. Of those that were not identified in our searches, having checked we do not believe any would meet the inclusion criteria. Writing comments p.5 - "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? Response: We have amended this to ‘type(s) of exposure’ p.6 - "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. Response: Where this category is first introduced, we have amended the text to “…categorised as measures of incident radiation (herein referred to as ‘radiation’).” This is hopefully more clear and accurate. p. 9 - All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". Response: Whilst many of the results have wide confidence intervals (in the forest plot), or suggest very small associations (in the table), the overall direction of the results suggests a beneficial effect. Most of the studies report a point estimate, regression, or correlation result that favours sunlight – particularly when looking at the regression and correlation results reported in Table 1. p.12 - "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. Response: I think the wording was actually incorrect the first time – apologies and thank you for highlighting this. We have changed the wording to – “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.” Hopefully this is clearer.Many thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below.Competing Interests: No competing interests were disclosed. Close 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. Response: Thank you for pointing out this oversight. We have now redrawn all forest plots using the exponential scale on the x-axis in order to accurately represent the beneficial effects. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. Response: We agree with you that Lin et al inappropriately refer to “biological damage (BD/m2)”. The definition they provide is unclear and the use of the term is confusing; no biological damage has actually been measured. What it appears they have done is to obtain average July UVR data via TOMS in order to estimate individual levels of UV erythemal exposure, which in turn is divided into quartiles. As such, we believe it is correctly categorised as a ‘radiation’ measure. To avoid confusion, we have decided to remove mention of the units BD/m2 and instead labelled/referred to this exposure measure as quartiles of “Average July erythemal UVR.” 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. Response: Thank you for these carefully considered and insightful suggestions for further analyses. We have now added a paragraph in the discussion addressing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Whilst we agree that the suggestions you make for further analyses would make valuable contribution to the paper, unfortunately we don’t believe we have sufficient data to fully address suggestion a) at this stage. Given the lack of reporting on skin type/colour or ethnicity in the included studies, we don’t have the individual data required. As such, we would have to incorporate further population level data to map against each study – which is possible but comes with its own challenges and complications that are beyond the scope of this project currently. This is also the case for suggestion b), which may be more feasible to implement from the available data but would require a good deal of additional investigation. However, we believe these suggestions are absolutely worth exploring and as such have incorporated them into our suggestions for future research: “Firstly, there is a clear need for more research into the effects of sunlight on people with darker skin. The majority of studies were conducted in predominantly White populations, with many specifically limiting inclusion to White people only. Moreover, skin type/colour or ethnicity data ought to be more commonly reported in population level studies. This would allow for more insightful analyses of the available data. For example, allowing for further investigating into the hypothesis that those with darker skin benefit most from sunlight in high UVR areas.” 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". Response: We have changed the term ‘inconclusive’ to ‘mixed’ which, as pointed out in the comment here, more accurately reflects the findings laid out in the following section of the abstract. We have also adapted the conclusion of the abstract to read “Findings from observational epidemiological studies of the association between sunlight exposure and mortality vary across different disease outcome and location being investigated. As such, the findings do not provide a strong rationale for changes to sun protection guidance.” 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Response: We found this review in our initial scoping. Whilst it was a useful and relevant paper, we did not prioritise investigating their reference list as it was not a systematic review. However, we have since checked and do not believe we missed any studies that ought to have been included. Most were identified in our searches, some of which were included. Those that were excluded were most commonly either not measuring mortality; conducted on people with pre-existing disease; or were vitamin D studies. Of those that were not identified in our searches, having checked we do not believe any would meet the inclusion criteria. Writing comments p.5 - "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? Response: We have amended this to ‘type(s) of exposure’ p.6 - "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. Response: Where this category is first introduced, we have amended the text to “…categorised as measures of incident radiation (herein referred to as ‘radiation’).” This is hopefully more clear and accurate. p. 9 - All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". Response: Whilst many of the results have wide confidence intervals (in the forest plot), or suggest very small associations (in the table), the overall direction of the results suggests a beneficial effect. Most of the studies report a point estimate, regression, or correlation result that favours sunlight – particularly when looking at the regression and correlation results reported in Table 1. p.12 - "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. Response: I think the wording was actually incorrect the first time – apologies and thank you for highlighting this. We have changed the wording to – “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.” Hopefully this is clearer. COMMENTS ON THIS REPORT - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below. ... Continue reading Many thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below. 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. Response: Thank you for pointing out this oversight. We have now redrawn all forest plots using the exponential scale on the x-axis in order to accurately represent the beneficial effects. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. Response: We agree with you that Lin et al inappropriately refer to “biological damage (BD/m2)”. The definition they provide is unclear and the use of the term is confusing; no biological damage has actually been measured. What it appears they have done is to obtain average July UVR data via TOMS in order to estimate individual levels of UV erythemal exposure, which in turn is divided into quartiles. As such, we believe it is correctly categorised as a ‘radiation’ measure. To avoid confusion, we have decided to remove mention of the units BD/m2 and instead labelled/referred to this exposure measure as quartiles of “Average July erythemal UVR.” 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. Response: Thank you for these carefully considered and insightful suggestions for further analyses. We have now added a paragraph in the discussion addressing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Whilst we agree that the suggestions you make for further analyses would make valuable contribution to the paper, unfortunately we don’t believe we have sufficient data to fully address suggestion a) at this stage. Given the lack of reporting on skin type/colour or ethnicity in the included studies, we don’t have the individual data required. As such, we would have to incorporate further population level data to map against each study – which is possible but comes with its own challenges and complications that are beyond the scope of this project currently. This is also the case for suggestion b), which may be more feasible to implement from the available data but would require a good deal of additional investigation. However, we believe these suggestions are absolutely worth exploring and as such have incorporated them into our suggestions for future research: “Firstly, there is a clear need for more research into the effects of sunlight on people with darker skin. The majority of studies were conducted in predominantly White populations, with many specifically limiting inclusion to White people only. Moreover, skin type/colour or ethnicity data ought to be more commonly reported in population level studies. This would allow for more insightful analyses of the available data. For example, allowing for further investigating into the hypothesis that those with darker skin benefit most from sunlight in high UVR areas.” 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". Response: We have changed the term ‘inconclusive’ to ‘mixed’ which, as pointed out in the comment here, more accurately reflects the findings laid out in the following section of the abstract. We have also adapted the conclusion of the abstract to read “Findings from observational epidemiological studies of the association between sunlight exposure and mortality vary across different disease outcome and location being investigated. As such, the findings do not provide a strong rationale for changes to sun protection guidance.” 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Response: We found this review in our initial scoping. Whilst it was a useful and relevant paper, we did not prioritise investigating their reference list as it was not a systematic review. However, we have since checked and do not believe we missed any studies that ought to have been included. Most were identified in our searches, some of which were included. Those that were excluded were most commonly either not measuring mortality; conducted on people with pre-existing disease; or were vitamin D studies. Of those that were not identified in our searches, having checked we do not believe any would meet the inclusion criteria. Writing comments p.5 - "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? Response: We have amended this to ‘type(s) of exposure’ p.6 - "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. Response: Where this category is first introduced, we have amended the text to “…categorised as measures of incident radiation (herein referred to as ‘radiation’).” This is hopefully more clear and accurate. p. 9 - All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". Response: Whilst many of the results have wide confidence intervals (in the forest plot), or suggest very small associations (in the table), the overall direction of the results suggests a beneficial effect. Most of the studies report a point estimate, regression, or correlation result that favours sunlight – particularly when looking at the regression and correlation results reported in Table 1. p.12 - "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. Response: I think the wording was actually incorrect the first time – apologies and thank you for highlighting this. We have changed the wording to – “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.” Hopefully this is clearer.Many thanks for your review - your comments and suggestions were very helpful and have helped us to improve the paper. We have attempted to respond to each point below.Competing Interests: No competing interests were disclosed. Close 1. The graphical presentations in the Forest plots are misleading, as they underrepresent the beneficial effects. This is because the x-axis distance between 0.5 and 1.0 (reduction in harmful effect) is drawn as the same length as between 1.0 and 1.5 (harms), whereas 0.5 means a 2-fold reduction. The solution is to plot log2 of the effect size. Or at least plot 0.5 at the same distance from 1.0 as 2.0 is. Response: Thank you for pointing out this oversight. We have now redrawn all forest plots using the exponential scale on the x-axis in order to accurately represent the beneficial effects. 2. The Lin[76] paper's measurement is correctly described in the text as a sunlight exposure measurement, but the Forest plot figures label it "biological damage". That is indeed reference 76's terminology, but it is inaccurate and, importantly, makes it harder to see the categorization of measurement metrics into Radiation exposure, Proxy for radiation exposure, and Behavioural. These entries should be relabeled "exposure" or something of the sort. Response: We agree with you that Lin et al inappropriately refer to “biological damage (BD/m2)”. The definition they provide is unclear and the use of the term is confusing; no biological damage has actually been measured. What it appears they have done is to obtain average July UVR data via TOMS in order to estimate individual levels of UV erythemal exposure, which in turn is divided into quartiles. As such, we believe it is correctly categorised as a ‘radiation’ measure. To avoid confusion, we have decided to remove mention of the units BD/m2 and instead labelled/referred to this exposure measure as quartiles of “Average July erythemal UVR.” 3. For all-cause and all-CVD mortality (and less dramatically for others), the Lindqvist, Yang, and Goggins papers showed a beneficial effect sunlight, in striking contrast to other studies. The Discussion section notes that the first two were conducted in Sweden and so might reflect a sunlight benefit limited to low sun-exposure regions. It then worries that Goggins was conducted in Hong Kong, near the equator, so doesn't fit the concept. But if one factors in the darker skin in Hong Kong, a revised hypothesis would be that individuals with low sunlight penetration into skin might be the ones benefitting the most from sunlight. How to test that? Two approaches occur to me, not mutually exclusive. a) Make a 4-place table of # of studies finding a sunlight benefit vs harm X # conducted on populations with low vs ordinary sunlight presence in the skin. Then do a Fisher's exact test. b) It is of course conventional to plot effect sizes relative to the local control group, as done here. But that control group's absolute rate will vary between Sweden, UK, Hong Kong, and US, especially if there is indeed a beneficial or harmful effect of sunlight, and this change of the baseline might obscure the effect. How about plotting the absolute mortality rate in controls and +sun comparison groups (so two points per analysis instead of one)? In the case of a beneficial effect of sun, does sun exposure reduce the mortality rate in a low-sun region to that of a high-sun region? In which case one might not expect more sun in the high-sun region to have much beneficial effect. Another way to look at this is as a dose response that has an optimum (e.g. as vitamin-related effects often do). An increase in dose to the left of the minimal mortality peak (valley) will have one effect and the same amount of increase to the right of the valley will have the opposite effect. Item #3 is a suggestion, not a requirement for a revised ms. Response: Thank you for these carefully considered and insightful suggestions for further analyses. We have now added a paragraph in the discussion addressing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UV R. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” Whilst we agree that the suggestions you make for further analyses would make valuable contribution to the paper, unfortunately we don’t believe we have sufficient data to fully address suggestion a) at this stage. Given the lack of reporting on skin type/colour or ethnicity in the included studies, we don’t have the individual data required. As such, we would have to incorporate further population level data to map against each study – which is possible but comes with its own challenges and complications that are beyond the scope of this project currently. This is also the case for suggestion b), which may be more feasible to implement from the available data but would require a good deal of additional investigation. However, we believe these suggestions are absolutely worth exploring and as such have incorporated them into our suggestions for future research: “Firstly, there is a clear need for more research into the effects of sunlight on people with darker skin. The majority of studies were conducted in predominantly White populations, with many specifically limiting inclusion to White people only. Moreover, skin type/colour or ethnicity data ought to be more commonly reported in population level studies. This would allow for more insightful analyses of the available data. For example, allowing for further investigating into the hypothesis that those with darker skin benefit most from sunlight in high UVR areas.” 4. Abstract's Results summary. Using the word "inconclusive" will just tempt the reader to skip this paper and go on to the next one. The results are only inconclusive with regard to providing a single advisory on how much sunlight a person should seek. This situation is reasonable, as each biological effect of sunlight will have a different dose response (e.g. basal and squamous cell carcinomas have different UV dose responses), although it would have been nice if Nature had established a clear net benefit at some exposure. Unless the suggestions in #3 turn up something, I suggest saying that the literature's studies give mixed results that vary with the disease endpoint and with the geographical location of the population being studied, and it is therefore not possible to identify an ideal amount of sunlight exposure. That's different from inconclusive: I tell my students that an experiment that didn't support a beautiful hypothesis wasn't a failure, it was a success and the answer was "no". Response: We have changed the term ‘inconclusive’ to ‘mixed’ which, as pointed out in the comment here, more accurately reflects the findings laid out in the following section of the abstract. We have also adapted the conclusion of the abstract to read “Findings from observational epidemiological studies of the association between sunlight exposure and mortality vary across different disease outcome and location being investigated. As such, the findings do not provide a strong rationale for changes to sun protection guidance.” 5. Some of the studies and meta-analyses cited in Hoel et al, Dermato-Endocrinology 2016 were not included here. Is it worth pointing out the basis for these omissions? Response: We found this review in our initial scoping. Whilst it was a useful and relevant paper, we did not prioritise investigating their reference list as it was not a systematic review. However, we have since checked and do not believe we missed any studies that ought to have been included. Most were identified in our searches, some of which were included. Those that were excluded were most commonly either not measuring mortality; conducted on people with pre-existing disease; or were vitamin D studies. Of those that were not identified in our searches, having checked we do not believe any would meet the inclusion criteria. Writing comments p.5 - "measured a single exposure", "measured two exposures" What does this mean? Two actual exposures, eg two sunburns? Two types of exposure? Two measurement metrics for exposure? Response: We have amended this to ‘type(s) of exposure’ p.6 - "Radiation" in the heading here and elsewhere. Sunlight is radiation, a thing. What is meant here is not the thing but amount of it to which the human is exposed, the "radiation exposure", "radiation dose", or more technically "radiation fluence". Maybe clearest to the average reader would be "Incident radiation" for these headings. Response: Where this category is first introduced, we have amended the text to “…categorised as measures of incident radiation (herein referred to as ‘radiation’).” This is hopefully more clear and accurate. p. 9 - All-cancer mortality overview. These results don't strike me as "the majority being in the direction of a beneficial association". Response: Whilst many of the results have wide confidence intervals (in the forest plot), or suggest very small associations (in the table), the overall direction of the results suggests a beneficial effect. Most of the studies report a point estimate, regression, or correlation result that favours sunlight – particularly when looking at the regression and correlation results reported in Table 1. p.12 - "the results of six were in the direction of a beneficial effect of sunlight: two found that higher latitude (i.e. lower sunlight) was associated with lower melanoma" has to be read twice to find out it is not a contradiction. Perhaps re-word this. Response: I think the wording was actually incorrect the first time – apologies and thank you for highlighting this. We have changed the wording to – “two found a positive relationship between degree of latitude and melanoma mortality, suggesting that those living at more northerly latitudes, where there is less sunlight, experienced higher levels of melanoma mortality.” Hopefully this is clearer. Views 0 How to cite this report: Assis LVMd. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36172) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36172 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36172 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 19 Aug 2025 Approved with Reservations VIEWS 0 The manuscript from Parkhouse and colleagues is a systematic literature review that evaluated the effects of sunlight on human health. The authors used a solid literature review method, providing a clear methodological view on papers that were considered and ... Continue reading I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Close The manuscript from Parkhouse and colleagues is a systematic literature review that evaluated the effects of sunlight on human health. The authors used a solid literature review method, providing a clear methodological view on papers that were considered and discarded. The overall design of the study is well performed. Due to the mixed evidence in all parameters analyzed, the authors decided to write a narrative description of the data, which is pleasant and easy to follow. Overall, the authors’ conclusion is mixed in almost all parameters analyzed, except for the risk of skin cancer. On the other hand, the authors show evidence, albeit mixed, that sunlight has beneficial effects in many other parameters, including other forms of cancer. In the next lines, I have more specific suggestions to improve the manuscript: - Although the authors used the term sunlight, they rightly mentioned that other wavelengths are also included. However, the authors overfocused on UV, mostly UVB effects, but they did not mention the potential effects of visible light as a potential player in the negative and potentially beneficial effects of sunlight. In addition, the authors should discuss that, although UVB has been considered mostly a negative player, it only represents 2-5% of the solar spectrum. Conversely, visible light represents ca. 45% and infrared the remaining portion that reaches Earth. Therefore, I considered that this in itself is a biased reporting in the literature, where the effects of solar radiation have been attributed to UV (mostly UVB), leaving the impression that the other wavelengths (visible light and infrared) are innocuous. Several experimental studies show the potential negative effects of visible light on the skin, but when provided at a low level, visible light and its different wavelengths can have beneficial effects. The authors should address this issue, as it is critical to provide a well-balanced view on the sunlight effect. - Broadly defining the effect of sunlight on humans as either beneficial or negative is an oversimplification of the real situation. Sunlight effects are influenced by many factors, which, in essence, make the assessment of their effects a mighty task, which the authors elegantly tried to address. However, in my view, this could be better emphasized in the text, especially in the lay text. In addition, so many variables and covariates prevent the assessment of the sunlight effect per se. One suggestion to include is that sunlight has direct and indirect effects. Regarding the direct effects, for instance, one can argue about the consequences of UVB, UVA, and visible light on skin cells, contributing to skin cancer. However, the effects of sunlight on other parameters, such as CVD, breast, prostate cancer, etc, are likely due to indirect effects, such as increased physical activity/exercise, better life quality actions (e.g., diet) in this population. Therefore, while assessing the direct effects can be straightforward, evaluating the indirect effects of sunlight coexist with many variables that one cannot fully disentangle. Due to the complex nature of these effects, I suggest that the authors consider my suggestion and include a paragraph in the discussion. - Since the overall conclusions are already mixed, the authors should further emphasize that subgroup analysis for skin type is underpowered. I would further emphasize that such analyses and their conclusions are limited. - One should also consider that sunlight effects, either direct or indirect, can be influenced by the region/country, as the other variables, described in 2, are more diverse. This could be further discussed in the revised manuscript. - An interesting aspect of the study is that the authors classify the bias levels of the study. I find this very interesting. Considering that the evidence is mixed, could the authors, at least, suggest a few actions that could make future studies less biased and more standard? I understand that this is not the focus of this study, but it could serve as a building block for a future unifying methodological paper. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? Yes - Are the conclusions drawn adequately supported by the results presented in the review? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: I'm experienced in the fields of skin biology, circadian biology, photobiology, and metabolism. CITE HOW TO CITE THIS REPORT Assis LVMd. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36172) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36172 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36172 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseThank you for providing this thorough review. We have attempted to address your points and respond to each below: 1. Although the authors used the term sunlight, they rightly ... Continue reading Thank you for providing this thorough review. We have attempted to address your points and respond to each below: 1. Although the authors used the term sunlight, they rightly mentioned that other wavelengths are also included. However, the authors overfocused on UV, mostly UVB effects, but they did not mention the potential effects of visible light as a potential player in the negative and potentially beneficial effects of sunlight. In addition, the authors should discuss that, although UVB has been considered mostly a negative player, it only represents 2-5% of the solar spectrum. Conversely, visible light represents ca. 45% and infrared the remaining portion that reaches Earth. Therefore, I considered that this in itself is a biased reporting in the literature, where the effects of solar radiation have been attributed to UV (mostly UVB), leaving the impression that the other wavelengths (visible light and infrared) are innocuous. Several experimental studies show the potential negative effects of visible light on the skin, but when provided at a low level, visible light and its different wavelengths can have beneficial effects. The authors should address this issue, as it is critical to provide a well-balanced view on the sunlight effect. Response: Thank you. We have highlighted your point regarding consideration to VL and IRR effects with respect to mortality, and have incorporated this within our paragraph that discusses the solar radiation measurements in the studies included in this review. We have also referenced the helpful paper you have cited. We have not specifically mentioned % of ambient radiation as effects are also dependent on factors including energy of the specific radiation (e.g. as would apply to a UVA vs UVB comparison), and this aspect would be rather technical to explain fully. “Therefore, it is warranted to consider a broader range of solar radiation and its effects with respect to mortality, than UVB alone. Apart from UVR, this includes the VL and IRR which are also emitted by the sun, and reach and penetrate the skin where they may have biological effects. For example, experimental research has indicated that VL may contribute to skin cancer development as well as having potential beneficial effects (de Assis, et al 2021).” https://doi.org/10.1016/j.jphotochemrev.2021.100403 2. Broadly defining the effect of sunlight on humans as either beneficial or negative is an oversimplification of the real situation. Sunlight effects are influenced by many factors, which, in essence, make the assessment of their effects a mighty task, which the authors elegantly tried to address. However, in my view, this could be better emphasized in the text, especially in the lay text. In addition, so many variables and covariates prevent the assessment of the sunlight effect per se. One suggestion to include is that sunlight has direct and indirect effects. Regarding the direct effects, for instance, one can argue about the consequences of UVB, UVA, and visible light on skin cells, contributing to skin cancer. However, the effects of sunlight on other parameters, such as CVD, breast, prostate cancer, etc, are likely due to indirect effects, such as increased physical activity/exercise, better life quality actions (e.g., diet) in this population. Therefore, while assessing the direct effects can be straightforward, evaluating the indirect effects of sunlight coexist with many variables that one cannot fully disentangle. Due to the complex nature of these effects, I suggest that the authors consider my suggestion and include a paragraph in the discussion. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, elaborating on the point you make here to discuss the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 3. Since the overall conclusions are already mixed, the authors should further emphasize that subgroup analysis for skin type is underpowered. I would further emphasize that such analyses and their conclusions are limited. Response: We have added a sentence at the end of the paragraph on skin type subgrouping in the discussion to emphasise the limited conclusions due to lack of data: “Though, given the lack of available data on skin type/colour or ethnicity, any conclusions drawn from these studies are limited.” 4. One should also consider that sunlight effects, either direct or indirect, can be influenced by the region/country, as the other variables, described in 2, are more diverse. This could be further discussed in the revised manuscript. Response: We hope the final sentence at the end of the new paragraph (quoted in point 2 above) adequately addresses the potential variations in confounders, such as diet and quality of life, across regions. 5. An interesting aspect of the study is that the authors classify the bias levels of the study. I find this very interesting. Considering that the evidence is mixed, could the authors, at least, suggest a few actions that could make future studies less biased and more standard? I understand that this is not the focus of this study, but it could serve as a building block for a future unifying methodological paper. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it.Thank you for providing this thorough review. We have attempted to address your points and respond to each below:Competing Interests: No competing interests were disclosed. Close 1. Although the authors used the term sunlight, they rightly mentioned that other wavelengths are also included. However, the authors overfocused on UV, mostly UVB effects, but they did not mention the potential effects of visible light as a potential player in the negative and potentially beneficial effects of sunlight. In addition, the authors should discuss that, although UVB has been considered mostly a negative player, it only represents 2-5% of the solar spectrum. Conversely, visible light represents ca. 45% and infrared the remaining portion that reaches Earth. Therefore, I considered that this in itself is a biased reporting in the literature, where the effects of solar radiation have been attributed to UV (mostly UVB), leaving the impression that the other wavelengths (visible light and infrared) are innocuous. Several experimental studies show the potential negative effects of visible light on the skin, but when provided at a low level, visible light and its different wavelengths can have beneficial effects. The authors should address this issue, as it is critical to provide a well-balanced view on the sunlight effect. Response: Thank you. We have highlighted your point regarding consideration to VL and IRR effects with respect to mortality, and have incorporated this within our paragraph that discusses the solar radiation measurements in the studies included in this review. We have also referenced the helpful paper you have cited. We have not specifically mentioned % of ambient radiation as effects are also dependent on factors including energy of the specific radiation (e.g. as would apply to a UVA vs UVB comparison), and this aspect would be rather technical to explain fully. “Therefore, it is warranted to consider a broader range of solar radiation and its effects with respect to mortality, than UVB alone. Apart from UVR, this includes the VL and IRR which are also emitted by the sun, and reach and penetrate the skin where they may have biological effects. For example, experimental research has indicated that VL may contribute to skin cancer development as well as having potential beneficial effects (de Assis, et al 2021).” https://doi.org/10.1016/j.jphotochemrev.2021.100403 2. Broadly defining the effect of sunlight on humans as either beneficial or negative is an oversimplification of the real situation. Sunlight effects are influenced by many factors, which, in essence, make the assessment of their effects a mighty task, which the authors elegantly tried to address. However, in my view, this could be better emphasized in the text, especially in the lay text. In addition, so many variables and covariates prevent the assessment of the sunlight effect per se. One suggestion to include is that sunlight has direct and indirect effects. Regarding the direct effects, for instance, one can argue about the consequences of UVB, UVA, and visible light on skin cells, contributing to skin cancer. However, the effects of sunlight on other parameters, such as CVD, breast, prostate cancer, etc, are likely due to indirect effects, such as increased physical activity/exercise, better life quality actions (e.g., diet) in this population. Therefore, while assessing the direct effects can be straightforward, evaluating the indirect effects of sunlight coexist with many variables that one cannot fully disentangle. Due to the complex nature of these effects, I suggest that the authors consider my suggestion and include a paragraph in the discussion. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, elaborating on the point you make here to discuss the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 3. Since the overall conclusions are already mixed, the authors should further emphasize that subgroup analysis for skin type is underpowered. I would further emphasize that such analyses and their conclusions are limited. Response: We have added a sentence at the end of the paragraph on skin type subgrouping in the discussion to emphasise the limited conclusions due to lack of data: “Though, given the lack of available data on skin type/colour or ethnicity, any conclusions drawn from these studies are limited.” 4. One should also consider that sunlight effects, either direct or indirect, can be influenced by the region/country, as the other variables, described in 2, are more diverse. This could be further discussed in the revised manuscript. Response: We hope the final sentence at the end of the new paragraph (quoted in point 2 above) adequately addresses the potential variations in confounders, such as diet and quality of life, across regions. 5. An interesting aspect of the study is that the authors classify the bias levels of the study. I find this very interesting. Considering that the evidence is mixed, could the authors, at least, suggest a few actions that could make future studies less biased and more standard? I understand that this is not the focus of this study, but it could serve as a building block for a future unifying methodological paper. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it. COMMENTS ON THIS REPORT - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseThank you for providing this thorough review. We have attempted to address your points and respond to each below: 1. Although the authors used the term sunlight, they rightly ... Continue reading Thank you for providing this thorough review. We have attempted to address your points and respond to each below: 1. Although the authors used the term sunlight, they rightly mentioned that other wavelengths are also included. However, the authors overfocused on UV, mostly UVB effects, but they did not mention the potential effects of visible light as a potential player in the negative and potentially beneficial effects of sunlight. In addition, the authors should discuss that, although UVB has been considered mostly a negative player, it only represents 2-5% of the solar spectrum. Conversely, visible light represents ca. 45% and infrared the remaining portion that reaches Earth. Therefore, I considered that this in itself is a biased reporting in the literature, where the effects of solar radiation have been attributed to UV (mostly UVB), leaving the impression that the other wavelengths (visible light and infrared) are innocuous. Several experimental studies show the potential negative effects of visible light on the skin, but when provided at a low level, visible light and its different wavelengths can have beneficial effects. The authors should address this issue, as it is critical to provide a well-balanced view on the sunlight effect. Response: Thank you. We have highlighted your point regarding consideration to VL and IRR effects with respect to mortality, and have incorporated this within our paragraph that discusses the solar radiation measurements in the studies included in this review. We have also referenced the helpful paper you have cited. We have not specifically mentioned % of ambient radiation as effects are also dependent on factors including energy of the specific radiation (e.g. as would apply to a UVA vs UVB comparison), and this aspect would be rather technical to explain fully. “Therefore, it is warranted to consider a broader range of solar radiation and its effects with respect to mortality, than UVB alone. Apart from UVR, this includes the VL and IRR which are also emitted by the sun, and reach and penetrate the skin where they may have biological effects. For example, experimental research has indicated that VL may contribute to skin cancer development as well as having potential beneficial effects (de Assis, et al 2021).” https://doi.org/10.1016/j.jphotochemrev.2021.100403 2. Broadly defining the effect of sunlight on humans as either beneficial or negative is an oversimplification of the real situation. Sunlight effects are influenced by many factors, which, in essence, make the assessment of their effects a mighty task, which the authors elegantly tried to address. However, in my view, this could be better emphasized in the text, especially in the lay text. In addition, so many variables and covariates prevent the assessment of the sunlight effect per se. One suggestion to include is that sunlight has direct and indirect effects. Regarding the direct effects, for instance, one can argue about the consequences of UVB, UVA, and visible light on skin cells, contributing to skin cancer. However, the effects of sunlight on other parameters, such as CVD, breast, prostate cancer, etc, are likely due to indirect effects, such as increased physical activity/exercise, better life quality actions (e.g., diet) in this population. Therefore, while assessing the direct effects can be straightforward, evaluating the indirect effects of sunlight coexist with many variables that one cannot fully disentangle. Due to the complex nature of these effects, I suggest that the authors consider my suggestion and include a paragraph in the discussion. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, elaborating on the point you make here to discuss the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 3. Since the overall conclusions are already mixed, the authors should further emphasize that subgroup analysis for skin type is underpowered. I would further emphasize that such analyses and their conclusions are limited. Response: We have added a sentence at the end of the paragraph on skin type subgrouping in the discussion to emphasise the limited conclusions due to lack of data: “Though, given the lack of available data on skin type/colour or ethnicity, any conclusions drawn from these studies are limited.” 4. One should also consider that sunlight effects, either direct or indirect, can be influenced by the region/country, as the other variables, described in 2, are more diverse. This could be further discussed in the revised manuscript. Response: We hope the final sentence at the end of the new paragraph (quoted in point 2 above) adequately addresses the potential variations in confounders, such as diet and quality of life, across regions. 5. An interesting aspect of the study is that the authors classify the bias levels of the study. I find this very interesting. Considering that the evidence is mixed, could the authors, at least, suggest a few actions that could make future studies less biased and more standard? I understand that this is not the focus of this study, but it could serve as a building block for a future unifying methodological paper. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it.Thank you for providing this thorough review. We have attempted to address your points and respond to each below:Competing Interests: No competing interests were disclosed. Close 1. Although the authors used the term sunlight, they rightly mentioned that other wavelengths are also included. However, the authors overfocused on UV, mostly UVB effects, but they did not mention the potential effects of visible light as a potential player in the negative and potentially beneficial effects of sunlight. In addition, the authors should discuss that, although UVB has been considered mostly a negative player, it only represents 2-5% of the solar spectrum. Conversely, visible light represents ca. 45% and infrared the remaining portion that reaches Earth. Therefore, I considered that this in itself is a biased reporting in the literature, where the effects of solar radiation have been attributed to UV (mostly UVB), leaving the impression that the other wavelengths (visible light and infrared) are innocuous. Several experimental studies show the potential negative effects of visible light on the skin, but when provided at a low level, visible light and its different wavelengths can have beneficial effects. The authors should address this issue, as it is critical to provide a well-balanced view on the sunlight effect. Response: Thank you. We have highlighted your point regarding consideration to VL and IRR effects with respect to mortality, and have incorporated this within our paragraph that discusses the solar radiation measurements in the studies included in this review. We have also referenced the helpful paper you have cited. We have not specifically mentioned % of ambient radiation as effects are also dependent on factors including energy of the specific radiation (e.g. as would apply to a UVA vs UVB comparison), and this aspect would be rather technical to explain fully. “Therefore, it is warranted to consider a broader range of solar radiation and its effects with respect to mortality, than UVB alone. Apart from UVR, this includes the VL and IRR which are also emitted by the sun, and reach and penetrate the skin where they may have biological effects. For example, experimental research has indicated that VL may contribute to skin cancer development as well as having potential beneficial effects (de Assis, et al 2021).” https://doi.org/10.1016/j.jphotochemrev.2021.100403 2. Broadly defining the effect of sunlight on humans as either beneficial or negative is an oversimplification of the real situation. Sunlight effects are influenced by many factors, which, in essence, make the assessment of their effects a mighty task, which the authors elegantly tried to address. However, in my view, this could be better emphasized in the text, especially in the lay text. In addition, so many variables and covariates prevent the assessment of the sunlight effect per se. One suggestion to include is that sunlight has direct and indirect effects. Regarding the direct effects, for instance, one can argue about the consequences of UVB, UVA, and visible light on skin cells, contributing to skin cancer. However, the effects of sunlight on other parameters, such as CVD, breast, prostate cancer, etc, are likely due to indirect effects, such as increased physical activity/exercise, better life quality actions (e.g., diet) in this population. Therefore, while assessing the direct effects can be straightforward, evaluating the indirect effects of sunlight coexist with many variables that one cannot fully disentangle. Due to the complex nature of these effects, I suggest that the authors consider my suggestion and include a paragraph in the discussion. Response: We have added a paragraph in the results regarding important confounders, with a brief overview of the most/least common factors controlled for: “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should have ideally been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” We have also added a paragraph in the discussion, elaborating on the point you make here to discuss the difficulty in disentangling the effects of sunlight on the skin from confounders and lifestyle factors: “Furthermore, the fact that no study was conducted using a direct measurement of individual sunlight exposure makes it difficult to disentangle the specific effects of sunlight on the skin from various potential confounders. For example, it is feasible that people living in areas with higher levels of sunlight experience health benefits such as increased outdoor physical activity, better diet, and improved quality of life. Moreover, the effect of such confounders are likely to vary across region and/or country, further complicating the relationship.” 3. Since the overall conclusions are already mixed, the authors should further emphasize that subgroup analysis for skin type is underpowered. I would further emphasize that such analyses and their conclusions are limited. Response: We have added a sentence at the end of the paragraph on skin type subgrouping in the discussion to emphasise the limited conclusions due to lack of data: “Though, given the lack of available data on skin type/colour or ethnicity, any conclusions drawn from these studies are limited.” 4. One should also consider that sunlight effects, either direct or indirect, can be influenced by the region/country, as the other variables, described in 2, are more diverse. This could be further discussed in the revised manuscript. Response: We hope the final sentence at the end of the new paragraph (quoted in point 2 above) adequately addresses the potential variations in confounders, such as diet and quality of life, across regions. 5. An interesting aspect of the study is that the authors classify the bias levels of the study. I find this very interesting. Considering that the evidence is mixed, could the authors, at least, suggest a few actions that could make future studies less biased and more standard? I understand that this is not the focus of this study, but it could serve as a building block for a future unifying methodological paper. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it. Views 0 How to cite this report: Shraim R. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36178) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36178 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36178 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 19 Aug 2025 Approved with Reservations VIEWS 0 The authors of this study present an interesting and comprehensive systematic review on the effects of sun exposure on mortality. This is timely and useful research as this topic has been gaining interest among epidemiologists and other health researchers. The ... Continue reading I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Close The authors of this study present an interesting and comprehensive systematic review on the effects of sun exposure on mortality. This is timely and useful research as this topic has been gaining interest among epidemiologists and other health researchers. The authors received input from a public advisory group and this is reflected positively in the clear and informative language of the summary. It is unfortunate that a meta-analysis could not be performed but the overall methods and approach are sound. My comments are below, mainly to improve the clarity of the reported results. - Most of the studies identified are reported as at ‘high’ or even ‘very high’ risk of bias. This merits the inclusion of some text in the results section describing the reasons why this is the case. The details are available in table S4 but considering that this is a concern for the majority of included studies, it’s worth including in the main text and, for example, mentioning the shared concern around confounding. - Related to the above, the authors do not report much on the covariates/other factors considered in these studies. It would be useful to know what these studies are generally adjusting for in their analyses or even not adjusting for. In some cases, the original studies stratified by sex or ethnicity so that is clear but in general what factors were considered? Again, some of this information is in table S4 but worth including in the main text, for example, the authors include an important note on smoking for one of the studies in all-cause mortality. - For cohort and case-control studies, some discussion on sample size is needed. The size range is wide (from the hundreds to the hundred thousands) and this should be considered in the interpretation of the results as it has important implications for statistical power. - “…different skin types (three with Afro-Caribbean ancestry and one with very pale skin).” Please use a consistent descriptor, either ancestry or skin colour. - Data extraction methods indicate ‘funding source’ was collected but no further comments are available, were there any observations drawn from this data? - In the table footnotes, RoB is included as an abbreviation but not used in some tables and ‘b’ should be replaced with ‘β’ for consistency. - What was the specific continuous outcome measured for the linear regressions? - In the discussion, “…whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index.” The possibility of interaction with population-level differences should also be mentioned (and this ties in with the next paragraph on differences between ethnicities). - In the discussion, “This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types.” A recent Australian study did this and would be an important reference to include here: [ Ref-1 : (Neale R, et al.,(2024) https://doi.org/10.1016/j.anzjph.2023.100117.] - Prior to their concluding remarks, it would be useful if the authors can include recommendations for future studies, especially considering the present high risk of bias and inconclusive results. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? Yes - Are the conclusions drawn adequately supported by the results presented in the review? Partly

1. Neale R, Beedle V, Ebeling P, Elliott T, et al.: Balancing the risks and benefits of sun exposure: A revised position statement for Australian adults. Australian and New Zealand Journal of Public Health. 2024; 48 (1). Publisher Full TextCompeting Interests: No competing interests were disclosed. Reviewer Expertise: Epidemiology, genetic epidemiology, sun exposure and vitamin D CITE HOW TO CITE THIS REPORT Shraim R. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36178) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36178 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36178 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below. 1. Most of ... Continue reading Many thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below. 1. Most of the studies identified are reported as at ‘high’ or even ‘very high’ risk of bias. This merits the inclusion of some text in the results section describing the reasons why this is the case. The details are available in table S4 but considering that this is a concern for the majority of included studies, it’s worth including in the main text and, for example, mentioning the shared concern around confounding. Response: We have added a paragraph to the results that discusses the overall risk of bias and some of the key reasons for the volume of high risk assessments (confounding and measurement of exposure): “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should ideally have been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” 2. Related to the above, the authors do not report much on the covariates/other factors considered in these studies. It would be useful to know what these studies are generally adjusting for in their analyses or even not adjusting for. In some cases, the original studies stratified by sex or ethnicity so that is clear but in general what factors were considered? Again, some of this information is in table S4 but worth including in the main text, for example, the authors include an important note on smoking for one of the studies in all-cause mortality. Response: We have added a brief overview of the types of adjustment observed within the paragraph mentioned in the point above. 3. For cohort and case-control studies, some discussion on sample size is needed. The size range is wide (from the hundreds to the hundred thousands) and this should be considered in the interpretation of the results as it has important implications for statistical power. Response: We have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140). Minor points 1. “…different skin types (three with Afro-Caribbean ancestry and one with very pale skin).” Please use a consistent descriptor, either ancestry or skin colour. Response: We have changed the wording to “three with dark skin and one with light skin” 2. Data extraction methods indicate ‘funding source’ was collected but no further comments are available, were there any observations drawn from this data? Response: Funding was poorly reported in general. Where information was available we did not note anything that raised concerns. We have added a line in the discussion that points out the potential for conflict of interest in terms of publication bias: “Source of funding was poorly reported in general and we cannot exclude that conflict of interest was also associated with publication bias.” 3. In the table footnotes, RoB is included as an abbreviation but not used in some tables and ‘b’ should be replaced with ‘β’ for consistency. Response: Thank you for pointing out this error. The final column in all tables is now abbreviated to RoB and b has been replaced by β in all table footnotes. 4. What was the specific continuous outcome measured for the linear regressions? Response: The tables have been edited in order to add the specific outcome for all results, where possible, e.g., “all-cancer mortality rate per 100,000/year.” 5. In the discussion, “…whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index.” The possibility of interaction with population-level differences should also be mentioned (and this ties in with the next paragraph on differences between ethnicities). Response: We have added a paragraph in the discussion discussing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UVR. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” 6. In the discussion, “This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types.” A recent Australian study did this and would be an important reference to include here: [ Ref-1 : (Neale R, et al.,(2024) https://doi.org/10.1016/j.anzjph.2023.100117.] Response: Thank you for highlighting this guidance. The paper has now been mentioned and referenced: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. Prior to their concluding remarks, it would be useful if the authors can include recommendations for future studies, especially considering the present high risk of bias and inconclusive results. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it.Many thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below.Competing Interests: No competing interests were disclosed. Close 1. Most of the studies identified are reported as at ‘high’ or even ‘very high’ risk of bias. This merits the inclusion of some text in the results section describing the reasons why this is the case. The details are available in table S4 but considering that this is a concern for the majority of included studies, it’s worth including in the main text and, for example, mentioning the shared concern around confounding. Response: We have added a paragraph to the results that discusses the overall risk of bias and some of the key reasons for the volume of high risk assessments (confounding and measurement of exposure): “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should ideally have been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” 2. Related to the above, the authors do not report much on the covariates/other factors considered in these studies. It would be useful to know what these studies are generally adjusting for in their analyses or even not adjusting for. In some cases, the original studies stratified by sex or ethnicity so that is clear but in general what factors were considered? Again, some of this information is in table S4 but worth including in the main text, for example, the authors include an important note on smoking for one of the studies in all-cause mortality. Response: We have added a brief overview of the types of adjustment observed within the paragraph mentioned in the point above. 3. For cohort and case-control studies, some discussion on sample size is needed. The size range is wide (from the hundreds to the hundred thousands) and this should be considered in the interpretation of the results as it has important implications for statistical power. Response: We have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140). Minor points 1. “…different skin types (three with Afro-Caribbean ancestry and one with very pale skin).” Please use a consistent descriptor, either ancestry or skin colour. Response: We have changed the wording to “three with dark skin and one with light skin” 2. Data extraction methods indicate ‘funding source’ was collected but no further comments are available, were there any observations drawn from this data? Response: Funding was poorly reported in general. Where information was available we did not note anything that raised concerns. We have added a line in the discussion that points out the potential for conflict of interest in terms of publication bias: “Source of funding was poorly reported in general and we cannot exclude that conflict of interest was also associated with publication bias.” 3. In the table footnotes, RoB is included as an abbreviation but not used in some tables and ‘b’ should be replaced with ‘β’ for consistency. Response: Thank you for pointing out this error. The final column in all tables is now abbreviated to RoB and b has been replaced by β in all table footnotes. 4. What was the specific continuous outcome measured for the linear regressions? Response: The tables have been edited in order to add the specific outcome for all results, where possible, e.g., “all-cancer mortality rate per 100,000/year.” 5. In the discussion, “…whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index.” The possibility of interaction with population-level differences should also be mentioned (and this ties in with the next paragraph on differences between ethnicities). Response: We have added a paragraph in the discussion discussing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UVR. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” 6. In the discussion, “This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types.” A recent Australian study did this and would be an important reference to include here: [ Ref-1 : (Neale R, et al.,(2024) https://doi.org/10.1016/j.anzjph.2023.100117.] Response: Thank you for highlighting this guidance. The paper has now been mentioned and referenced: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. Prior to their concluding remarks, it would be useful if the authors can include recommendations for future studies, especially considering the present high risk of bias and inconclusive results. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it. COMMENTS ON THIS REPORT - Author Response 12 Dec 2025Tom Parkhouse, NIHR Bristol Evidence Synthesis Group, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, BS8 2PS, UK12 Dec 2025Author ResponseMany thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below. 1. Most of ... Continue reading Many thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below. 1. Most of the studies identified are reported as at ‘high’ or even ‘very high’ risk of bias. This merits the inclusion of some text in the results section describing the reasons why this is the case. The details are available in table S4 but considering that this is a concern for the majority of included studies, it’s worth including in the main text and, for example, mentioning the shared concern around confounding. Response: We have added a paragraph to the results that discusses the overall risk of bias and some of the key reasons for the volume of high risk assessments (confounding and measurement of exposure): “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should ideally have been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” 2. Related to the above, the authors do not report much on the covariates/other factors considered in these studies. It would be useful to know what these studies are generally adjusting for in their analyses or even not adjusting for. In some cases, the original studies stratified by sex or ethnicity so that is clear but in general what factors were considered? Again, some of this information is in table S4 but worth including in the main text, for example, the authors include an important note on smoking for one of the studies in all-cause mortality. Response: We have added a brief overview of the types of adjustment observed within the paragraph mentioned in the point above. 3. For cohort and case-control studies, some discussion on sample size is needed. The size range is wide (from the hundreds to the hundred thousands) and this should be considered in the interpretation of the results as it has important implications for statistical power. Response: We have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140). Minor points 1. “…different skin types (three with Afro-Caribbean ancestry and one with very pale skin).” Please use a consistent descriptor, either ancestry or skin colour. Response: We have changed the wording to “three with dark skin and one with light skin” 2. Data extraction methods indicate ‘funding source’ was collected but no further comments are available, were there any observations drawn from this data? Response: Funding was poorly reported in general. Where information was available we did not note anything that raised concerns. We have added a line in the discussion that points out the potential for conflict of interest in terms of publication bias: “Source of funding was poorly reported in general and we cannot exclude that conflict of interest was also associated with publication bias.” 3. In the table footnotes, RoB is included as an abbreviation but not used in some tables and ‘b’ should be replaced with ‘β’ for consistency. Response: Thank you for pointing out this error. The final column in all tables is now abbreviated to RoB and b has been replaced by β in all table footnotes. 4. What was the specific continuous outcome measured for the linear regressions? Response: The tables have been edited in order to add the specific outcome for all results, where possible, e.g., “all-cancer mortality rate per 100,000/year.” 5. In the discussion, “…whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index.” The possibility of interaction with population-level differences should also be mentioned (and this ties in with the next paragraph on differences between ethnicities). Response: We have added a paragraph in the discussion discussing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UVR. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” 6. In the discussion, “This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types.” A recent Australian study did this and would be an important reference to include here: [ Ref-1 : (Neale R, et al.,(2024) https://doi.org/10.1016/j.anzjph.2023.100117.] Response: Thank you for highlighting this guidance. The paper has now been mentioned and referenced: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. Prior to their concluding remarks, it would be useful if the authors can include recommendations for future studies, especially considering the present high risk of bias and inconclusive results. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it.Many thanks for taking the time to provide this helpful and insightful review. We attempted to address all of your points and provide our responses below.Competing Interests: No competing interests were disclosed. Close 1. Most of the studies identified are reported as at ‘high’ or even ‘very high’ risk of bias. This merits the inclusion of some text in the results section describing the reasons why this is the case. The details are available in table S4 but considering that this is a concern for the majority of included studies, it’s worth including in the main text and, for example, mentioning the shared concern around confounding. Response: We have added a paragraph to the results that discusses the overall risk of bias and some of the key reasons for the volume of high risk assessments (confounding and measurement of exposure): “The majority of results that were assessed for risk of bias were judged to be of some concern or high risk of bias. There were no results we judged to be low risk of bias across all domains, whilst some were assessed to be at very high risk of bias. The most common reasons for potential bias were uncontrolled confounding and concerns over the measurement of exposure. Concerns in the latter of these focused on the lack of direct measurement of personal sunlight exposure, as is largely inevitable in large population-based studies, as well as the use of aggregated measures that may have misclassified individual exposure. In order to judge the risk of bias due to confounding, we prespecified a number of important confounding factors which should ideally have been controlled for, including age, sex/gender, ethnicity, smoking status, socioeconomic status, physical activity and altitude. No study included in the review controlled for all of these factors. The most commonly controlled for factors were age and sex/gender, followed by smoking. Looking at the primary outcomes, around half of the results controlled for race/ethnicity, though this was predominately achieved through restricting analysis to White participants. Physical activity was less well controlled for, whilst almost no studies controlled for altitude.” 2. Related to the above, the authors do not report much on the covariates/other factors considered in these studies. It would be useful to know what these studies are generally adjusting for in their analyses or even not adjusting for. In some cases, the original studies stratified by sex or ethnicity so that is clear but in general what factors were considered? Again, some of this information is in table S4 but worth including in the main text, for example, the authors include an important note on smoking for one of the studies in all-cause mortality. Response: We have added a brief overview of the types of adjustment observed within the paragraph mentioned in the point above. 3. For cohort and case-control studies, some discussion on sample size is needed. The size range is wide (from the hundreds to the hundred thousands) and this should be considered in the interpretation of the results as it has important implications for statistical power. Response: We have added the number of participants included in the full cohort for all cohort studies with results for the primary outcomes. Where reported, we have also added the number of deaths among the cohort (relevant to each outcome). Sample sizes and considerations of power are less relevant to interpretation than are confidence intervals around estimated effect sizes, and we report these wherever they were available to us. We do not elaborate substantially on sample size differences, since in observational studies, larger is not necessarily better (as discussed, for example, by Egger et al (“Spurious precision? Meta-analysis of observational studies”. BMJ 1998; 316; https://doi.org/10.1136/bmj.316.7125.140). Minor points 1. “…different skin types (three with Afro-Caribbean ancestry and one with very pale skin).” Please use a consistent descriptor, either ancestry or skin colour. Response: We have changed the wording to “three with dark skin and one with light skin” 2. Data extraction methods indicate ‘funding source’ was collected but no further comments are available, were there any observations drawn from this data? Response: Funding was poorly reported in general. Where information was available we did not note anything that raised concerns. We have added a line in the discussion that points out the potential for conflict of interest in terms of publication bias: “Source of funding was poorly reported in general and we cannot exclude that conflict of interest was also associated with publication bias.” 3. In the table footnotes, RoB is included as an abbreviation but not used in some tables and ‘b’ should be replaced with ‘β’ for consistency. Response: Thank you for pointing out this error. The final column in all tables is now abbreviated to RoB and b has been replaced by β in all table footnotes. 4. What was the specific continuous outcome measured for the linear regressions? Response: The tables have been edited in order to add the specific outcome for all results, where possible, e.g., “all-cancer mortality rate per 100,000/year.” 5. In the discussion, “…whilst there are long established risks associated with sunlight exposure in high UVR locations, the benefits of sunlight may possibly outweigh the harms in regions with a generally low UV index.” The possibility of interaction with population-level differences should also be mentioned (and this ties in with the next paragraph on differences between ethnicities). Response: We have added a paragraph in the discussion discussing the possibility that the relationship between location and effect of sunlight exposure is mediated by skin type. We highlight results in the review that support this theory: “It is plausible that the relationship between geographical location and effect of sunlight exposure is moderated by skin type. Those with lighter skin may experience greater benefits of sunlight in higher latitude areas with typically low UVR. For example, in this review, the beneficial effect observed on all-cause and all-CVD mortality in northern Europe (Lindqvist et al, 2014; Stevenson et al, 2023). However, for those with darker skin, the benefits might be more pronounced in lower latitude, high UVR areas, as suggested by the beneficial effect observed for all-cancer mortality in Turkey (Altug & Kilçiksiz, 2020) and Hong Kong (Goggins et al, 2013). However, of these four studies, only Stevenson et al (2023) was specific in reporting the skin type/colour or ethnicity of their sample (restricting inclusion to only White participants). Therefore, such theories can only be made based on broad assertions about the skin type of large populations, and so ought to be made with caution.” 6. In the discussion, “This, in turn, would allow organizations responsible for sun safety messaging to provide more nuanced guidance for those with different skin types.” A recent Australian study did this and would be an important reference to include here: [ Ref-1 : (Neale R, et al.,(2024) https://doi.org/10.1016/j.anzjph.2023.100117.] Response: Thank you for highlighting this guidance. The paper has now been mentioned and referenced: “This, in turn, would help organizations responsible for sun safety messaging to provide more appropriate guidance for those with different skin types. For example, a recent summit of sun exposure experts in Australia promoted the use of sun safety advice that clearly distinguishes recommendations for people of different skin types (Neale et al, 2004).” 7. Prior to their concluding remarks, it would be useful if the authors can include recommendations for future studies, especially considering the present high risk of bias and inconclusive results. Response: We have now included a section for future research suggestions. These cover the need for more accurate reporting of skin type/colour or ethnicity, as well as the broader need for more studies specifically looking at populations with darker skin types; the need for more studies utilising individual-level, personal sun exposure data; and the suggestion to focus on standardising the measurement of sunlight exposure, and the methods used to measure it. Views 0 How to cite this report: Brito MT, Stevenson A, Gu J and Weller R. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36177) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36177 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36177 NOTE: it is important to ensure the information in square brackets after the title is included in this citation. Reviewer Report 19 Aug 2025 Maria Teresa Brito, The University of Edinburgh College of Medicine and Veterinary Medicine, Edinburgh, Scotland, UK Andrew Stevenson, School of GeoSciences, University of Edinburgh, Edinburgh, UK Jiayue Gu, The University of Edinburgh College of Medicine and Veterinary Medicine, Edinburgh, Scotland, UK Not Approved VIEWS 0 This is an important subject and we are delighted to see it reviewed and grateful to the authors for their work on this. Our research group is actively involved in research on the effects of UV/sunshine on systemic health and ... Continue reading 2. Harding R, Healy E, Ray A, Ellis N, et al.: Evidence for Variable Selective Pressures at MC1R. The American Journal of Human Genetics. 2000; 66 (4): 1351-1361 Publisher Full Text We confirm that we have read this submission and believe that we have an appropriate level of expertise to state that we do not consider it to be of an acceptable scientific standard, for reasons outlined above. Close This is an important subject and we are delighted to see it reviewed and grateful to the authors for their work on this. Our research group is actively involved in research on the effects of UV/sunshine on systemic health and we do however have significant concerns over the data and wish to make some comments. 1. America and Europe are very different UV environments and cannot be combined. We are concerned that not enough context has been given for where studies looking at the relationship between sunlight exposure were performed as this significantly affects response measures. This factor alone accounts for most of the variability in the relationship between UV and major outcome measures. For example, reduced all-cause and cardiovascular mortality with increased sunlight exposure are found by Lindqvist, Stevenson, Yang and Goggins. Lindqvist, Stevenson and Yang studied relative sun exposure in white skinned populations living in northern Europe (Sweden and the UK 50 to 68° N). Lin and He found increased all-cause and cardiovascular mortality associated with sun exposure, but these studies were of white skinned people living in the USA which is much closer to the equator (25 to 45° N) with significantly higher UV. Goggins finds reduced all-cause and cardiovascular mortality with increased sunlight exposure in an ethnically Chinese population studied in Taipei and Hong Kong (22 and 25°N). Skin colour determines biological response to sunlight/UV and white European skin is an evolutionary adaptation to the low light environment in Europe with strong selective pressure being seen for white skinned gene variants (Wilde S, et al., (2014) [Ref: 1] 10.1073/pnas.1316513111). Similarly, strong functional constraints against MC1R variants coding for pale skin are seen in all African populations (Harding R, et al., (2000) [Ref:2] https://doi.org/10.1086/302863). South of North Carolina, the United States are in African latitudes and thus UV environments, and any adverse effects of excess UV here bear little relevance to white skinned north European populations. This fundamental biological and evolutionary understanding is not given enough weight in this review, where studies from all latitudes are analysed together rather than considering whether a population is matched to ambient sunlight exposure by skin type. The relationship between cancer mortality and sunlight is similar to that of all-cause/CVS mortality. The great majority of papers shown an inverse relationship between sunlight exposure and cancer mortality. Again, Lin and He studying white skinned populations in the USA find increased cancer mortality. Studies of other populations, when adequately powered all find reduced mortality from cancers overall with increased UV exposure, despite increased skin cancer in white skinned subjects. These studies are all of populations living in a UV environment to which they are evolutionarily adapted. Altug (Turkey), Chen (China), Fukuda (Japan), Goggins (Hong Kong) and Stevenson (White British) all find an inverse relationship between sunlight exposure and cancer mortality. Lindqvist (Sweden) and Yang (Sweden) were underpowered. 2. Biological plausibility and dose response relationships. Not enough weight has been given to biological plausibility. Lin’s paper looking at American populations identifies higher mortality with UV, yet mortality from Cancer, Respiratory disease, Stroke, Injury, and Diabetes all peaked in Quartile 3 and decreased in Quartile 4 (the highest exposure), which does not support a straightforward dose–response relationship between UV exposure and disease-specific mortality. Instead, it suggests a potential threshold or non-linear association, or possibly reflects methodological limitations, such as exposure misclassification, arbitrary cutoff selection, or residual confounding. The only neat dose-response relationship Lin showed is that UV decreased infectious disease. The most prevalent causes of cancer found by Lin were lung, liver and prostate cancer, even though such a link with UV has never been shown before (in fact, the opposite). Melanoma mortality was raised (as expected) but only in the highest quartile of UV exposure (equating to UV exposure in Algeria). No mention is made of non-melanoma skin cancer mortality being raised with UV exposure, even though cutaneous SCC is undisputedly induced by UV exposure in a dose dependent fashion and can act as a positive control for UV exposure. By contrast, studies identifying all-cause mortality reductions with sunlight exposure in high latitude countries demonstrate a dose-response relationship (Lindqvist). This is consistent with a sunlight-driven beneficial mechanism. 3. The authors include an old pre-print version of Stevenson et al. The analysis has been improved since publication and peer review and could represent among the higher quality studies in the review. For example: • Domain 2 (was rated as high): risk of bias arising from measurement of the exposure. The exposure is now a direct measure of solar radiation rather than latitude (proxy). • Domain 5 (was rated as high): risk of bias due to missing date. The new analysis was multiple imputation to impute missing values. Should be low now. • Triangulation and use of negative control help with internal validity 4. We are concerned about the inclusion of studies that report only unadjusted correlations, as these are highly susceptible to bias. Including such studies alongside adjusted analyses may give the impression that they carry equal evidentiary weight. We suggest either excluding these studies based on a revised criterion (e.g., excluding studies that do not adjust for potential confounders) or clearly separating them into a distinct section or supplementary file. 5. Studies using different levels of evidence for sun exposure are treated equally. For example, He’s paper relies on dermatologist recorded actinic skin damage, but this proxy exposure can be influenced by a myriad of other environmental factors, and as the authors admit was heavily confounded, being strongly correlated with lower education levels, deprivation, outdoor working and living in the American south. By contrast, Stevenson and Lindqvist record direct measures of sunlight exposure and these are then validated (Stevenson) against serum vitamin D levels to confirm their accuracy. The authors should give more emphasis to the validation of sun exposure measures in the studies utilised, as this will influence what conclusions can be drawn from reliable data given the variability in measures within the field. 6. It would be helpful for the reader to better understand the evidence base if the review structured the evidence by study quality (perhaps by rating (lowest, middle, highest), or by ecological versus individual studies, if all studies in a section are rated the same quality). Also, by low and high sunlight environments – the authors could maybe colour code results according to low/high sun in graphs/tables. Ecological studies, which analyse data aggregated at the group level (e.g. countries, regions), have different strengths and weaknesses compared to studies that examine individual-level data. Separating them allows for a more accurate assessment of the evidence and avoids the potential for ecological fallacy. The authors note results that are compatible with higher or lower risk (harmful or beneficial) when they are just not statistically significant. If an association is not significant – but near statistical significance – it makes sense to consider it in a harmful/beneficial direction. But when the point estimate is higher or lower than 1 and the CI is wide (high p-value), it should not be considered beneficial or harmful. In the Discussion section, the authors start by stating that half of the articles addressing their primary outcome reported a positive direction effect, and the other half a negative direction effect. However, they do not refer to the fact that several of them have confidence intervals that show the results to not be significant. Of those that do not span 1 in their confidence intervals, the majority of results reports benefits from sunlight exposure. Little emphasis was put on multivariate vs univariate analyses in the Results section. In an area as subject to confounding as this is, it would be helpful to refer to the number of factors that were controlled for in each paper. This will also influence the weight that can be attributed to studies’ results – a statistically significant result with inadequate adjustment for confounding is still not reliable. 7. It would be helpful to describe the relative contribution of mortality from CVD, cancer and specific cancers (maybe top 5 vs melanoma/NMSC) in the UK in the introduction since the paper is examining the benefits/harms of sun exposure; this provides context to the scope of the issue. 8. The sun protection guidance focuses on skin cancer; however, in this review the evidence for sunlight exposure and skin cancer mortality is similar or lower to other cancers with a higher burden. This suggests that there is currently an overemphasis on skin cancer in the guidance and potential benefits of safe sun exposure should at least be acknowledged. Example, using the benefit/harm considered in text: - Melanoma mortality risk: 34/46 74% of results harmful - Pancreatic: 9/11 = 81% beneficial - Bowel 22/31 = 71% beneficial - Lung cancer 7/12 = beneficial 58% - Breast 13/17 = beneficial 76% - All cancer 20/22 = beneficial 91% 9. The authors rated most studies as high risk of bias. There should be a greater discussion about why the studies are high risk – e.g., what are the important confounders that are missing; direction of bias; consider reverse causation and misclassification. What are the authors recommendations for future cohort studies to reduce the risk of bias? This would help future researchers to design their studies to improve the evidence base. 10. While ROBINS-E was appropriately used to assess for risk of bias, would a tool like the Newcastle-Ottawa scale be useful to also assess the individual quality of the studies? This is especially relevant since certain articles were excluded from the ROBINS-E analysis due to the format of their findings; as such these would still undergo some form of quality assessment. This would help more substantial decisions to be made about what we can draw from the body of evidence. 11. The conclusion of this review is that ‘findings for overall mortality are too variable to provide a rationale for changes to sun protection guidance’. As the authors feel that the evidence is in a state of equipoise, they advise no change to guidance; however, existing guidance is to actively limit sun exposure, and thus ‘no change’ in effect means continue avoiding sunlight. This would be appropriate if the evidence pointed firmly to overall harm induced by sunlight, but no such evidence is found. Our own interpretation of the data presented is that all studies from north Europe show benefits exceeding risks for sunlight exposure. Only studies of white skinned populations in high UV environments show harm. Of note, the Australian Skin and Skin Cancer Research Centre sun safety guidelines were revised in 2024 to actively acknowledge and consider sunlight’s health benefits and to tailor advise based on skin colour. As Australia is a markedly higher UV environment than the UK it is remarkable that they should take this approach, while we continue advising sunlight avoidance. 1. America and Europe are very different UV environments and cannot be combined. We are concerned that not enough context has been given for where studies looking at the relationship between sunlight exposure were performed as this significantly affects response measures. This factor alone accounts for most of the variability in the relationship between UV and major outcome measures. For example, reduced all-cause and cardiovascular mortality with increased sunlight exposure are found by Lindqvist, Stevenson, Yang and Goggins. Lindqvist, Stevenson and Yang studied relative sun exposure in white skinned populations living in northern Europe (Sweden and the UK 50 to 68° N). Lin and He found increased all-cause and cardiovascular mortality associated with sun exposure, but these studies were of white skinned people living in the USA which is much closer to the equator (25 to 45° N) with significantly higher UV. Goggins finds reduced all-cause and cardiovascular mortality with increased sunlight exposure in an ethnically Chinese population studied in Taipei and Hong Kong (22 and 25°N). Skin colour determines biological response to sunlight/UV and white European skin is an evolutionary adaptation to the low light environment in Europe with strong selective pressure being seen for white skinned gene variants (Wilde S, et al., (2014) [Ref: 1] 10.1073/pnas.1316513111). Similarly, strong functional constraints against MC1R variants coding for pale skin are seen in all African populations (Harding R, et al., (2000) [Ref:2] https://doi.org/10.1086/302863). South of North Carolina, the United States are in African latitudes and thus UV environments, and any adverse effects of excess UV here bear little relevance to white skinned north European populations. This fundamental biological and evolutionary understanding is not given enough weight in this review, where studies from all latitudes are analysed together rather than considering whether a population is matched to ambient sunlight exposure by skin type. The relationship between cancer mortality and sunlight is similar to that of all-cause/CVS mortality. The great majority of papers shown an inverse relationship between sunlight exposure and cancer mortality. Again, Lin and He studying white skinned populations in the USA find increased cancer mortality. Studies of other populations, when adequately powered all find reduced mortality from cancers overall with increased UV exposure, despite increased skin cancer in white skinned subjects. These studies are all of populations living in a UV environment to which they are evolutionarily adapted. Altug (Turkey), Chen (China), Fukuda (Japan), Goggins (Hong Kong) and Stevenson (White British) all find an inverse relationship between sunlight exposure and cancer mortality. Lindqvist (Sweden) and Yang (Sweden) were underpowered. 2. Biological plausibility and dose response relationships. Not enough weight has been given to biological plausibility. Lin’s paper looking at American populations identifies higher mortality with UV, yet mortality from Cancer, Respiratory disease, Stroke, Injury, and Diabetes all peaked in Quartile 3 and decreased in Quartile 4 (the highest exposure), which does not support a straightforward dose–response relationship between UV exposure and disease-specific mortality. Instead, it suggests a potential threshold or non-linear association, or possibly reflects methodological limitations, such as exposure misclassification, arbitrary cutoff selection, or residual confounding. The only neat dose-response relationship Lin showed is that UV decreased infectious disease. The most prevalent causes of cancer found by Lin were lung, liver and prostate cancer, even though such a link with UV has never been shown before (in fact, the opposite). Melanoma mortality was raised (as expected) but only in the highest quartile of UV exposure (equating to UV exposure in Algeria). No mention is made of non-melanoma skin cancer mortality being raised with UV exposure, even though cutaneous SCC is undisputedly induced by UV exposure in a dose dependent fashion and can act as a positive control for UV exposure. By contrast, studies identifying all-cause mortality reductions with sunlight exposure in high latitude countries demonstrate a dose-response relationship (Lindqvist). This is consistent with a sunlight-driven beneficial mechanism. 3. The authors include an old pre-print version of Stevenson et al. The analysis has been improved since publication and peer review and could represent among the higher quality studies in the review. For example: • Domain 2 (was rated as high): risk of bias arising from measurement of the exposure. The exposure is now a direct measure of solar radiation rather than latitude (proxy). • Domain 5 (was rated as high): risk of bias due to missing date. The new analysis was multiple imputation to impute missing values. Should be low now. • Triangulation and use of negative control help with internal validity 4. We are concerned about the inclusion of studies that report only unadjusted correlations, as these are highly susceptible to bias. Including such studies alongside adjusted analyses may give the impression that they carry equal evidentiary weight. We suggest either excluding these studies based on a revised criterion (e.g., excluding studies that do not adjust for potential confounders) or clearly separating them into a distinct section or supplementary file. 5. Studies using different levels of evidence for sun exposure are treated equally. For example, He’s paper relies on dermatologist recorded actinic skin damage, but this proxy exposure can be influenced by a myriad of other environmental factors, and as the authors admit was heavily confounded, being strongly correlated with lower education levels, deprivation, outdoor working and living in the American south. By contrast, Stevenson and Lindqvist record direct measures of sunlight exposure and these are then validated (Stevenson) against serum vitamin D levels to confirm their accuracy. The authors should give more emphasis to the validation of sun exposure measures in the studies utilised, as this will influence what conclusions can be drawn from reliable data given the variability in measures within the field. 6. It would be helpful for the reader to better understand the evidence base if the review structured the evidence by study quality (perhaps by rating (lowest, middle, highest), or by ecological versus individual studies, if all studies in a section are rated the same quality). Also, by low and high sunlight environments – the authors could maybe colour code results according to low/high sun in graphs/tables. Ecological studies, which analyse data aggregated at the group level (e.g. countries, regions), have different strengths and weaknesses compared to studies that examine individual-level data. Separating them allows for a more accurate assessment of the evidence and avoids the potential for ecological fallacy. The authors note results that are compatible with higher or lower risk (harmful or beneficial) when they are just not statistically significant. If an association is not significant – but near statistical significance – it makes sense to consider it in a harmful/beneficial direction. But when the point estimate is higher or lower than 1 and the CI is wide (high p-value), it should not be considered beneficial or harmful. In the Discussion section, the authors start by stating that half of the articles addressing their primary outcome reported a positive direction effect, and the other half a negative direction effect. However, they do not refer to the fact that several of them have confidence intervals that show the results to not be significant. Of those that do not span 1 in their confidence intervals, the majority of results reports benefits from sunlight exposure. Little emphasis was put on multivariate vs univariate analyses in the Results section. In an area as subject to confounding as this is, it would be helpful to refer to the number of factors that were controlled for in each paper. This will also influence the weight that can be attributed to studies’ results – a statistically significant result with inadequate adjustment for confounding is still not reliable. 7. It would be helpful to describe the relative contribution of mortality from CVD, cancer and specific cancers (maybe top 5 vs melanoma/NMSC) in the UK in the introduction since the paper is examining the benefits/harms of sun exposure; this provides context to the scope of the issue. 8. The sun protection guidance focuses on skin cancer; however, in this review the evidence for sunlight exposure and skin cancer mortality is similar or lower to other cancers with a higher burden. This suggests that there is currently an overemphasis on skin cancer in the guidance and potential benefits of safe sun exposure should at least be acknowledged. Example, using the benefit/harm considered in text: - Melanoma mortality risk: 34/46 74% of results harmful - Pancreatic: 9/11 = 81% beneficial - Bowel 22/31 = 71% beneficial - Lung cancer 7/12 = beneficial 58% - Breast 13/17 = beneficial 76% - All cancer 20/22 = beneficial 91% 9. The authors rated most studies as high risk of bias. There should be a greater discussion about why the studies are high risk – e.g., what are the important confounders that are missing; direction of bias; consider reverse causation and misclassification. What are the authors recommendations for future cohort studies to reduce the risk of bias? This would help future researchers to design their studies to improve the evidence base. 10. While ROBINS-E was appropriately used to assess for risk of bias, would a tool like the Newcastle-Ottawa scale be useful to also assess the individual quality of the studies? This is especially relevant since certain articles were excluded from the ROBINS-E analysis due to the format of their findings; as such these would still undergo some form of quality assessment. This would help more substantial decisions to be made about what we can draw from the body of evidence. 11. The conclusion of this review is that ‘findings for overall mortality are too variable to provide a rationale for changes to sun protection guidance’. As the authors feel that the evidence is in a state of equipoise, they advise no change to guidance; however, existing guidance is to actively limit sun exposure, and thus ‘no change’ in effect means continue avoiding sunlight. This would be appropriate if the evidence pointed firmly to overall harm induced by sunlight, but no such evidence is found. Our own interpretation of the data presented is that all studies from north Europe show benefits exceeding risks for sunlight exposure. Only studies of white skinned populations in high UV environments show harm. Of note, the Australian Skin and Skin Cancer Research Centre sun safety guidelines were revised in 2024 to actively acknowledge and consider sunlight’s health benefits and to tailor advise based on skin colour. As Australia is a markedly higher UV environment than the UK it is remarkable that they should take this approach, while we continue advising sunlight avoidance. - Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes - Are sufficient details of the methods and analysis provided to allow replication by others? Yes - Is the statistical analysis and its interpretation appropriate? Partly - Are the conclusions drawn adequately supported by the results presented in the review? No

1. Wilde S, Timpson A, Kirsanow K, Kaiser E, et al.: Direct evidence for positive selection of skin, hair, and eye pigmentation in Europeans during the last 5,000 y. Proceedings of the National Academy of Sciences. 2014; 111 (13): 4832-4837 Publisher Full Text2. Harding R, Healy E, Ray A, Ellis N, et al.: Evidence for Variable Selective Pressures at MC1R. The American Journal of Human Genetics. 2000; 66 (4): 1351-1361 Publisher Full Text Competing Interests: No competing interests were disclosed. Reviewer Expertise: Dermatology, photobiology, epidemiology, translational medicine CITE HOW TO CITE THIS REPORT Brito MT, Stevenson A, Gu J and Weller R. Reviewer Report For: The effects of sunlight exposure on mortality: a systematic review of epidemiological studies [version 1; peer review: 4 approved with reservations, 2 not approved]. NIHR Open Res 2025, 5:51 (https://doi.org/10.3310/nihropenres.15197.r36177) The direct URL for this report is: https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36177 https://openresearch.nihr.ac.uk/articles/5-51/v1#referee-response-36177 NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article. Alongside their report, reviewers assign a status to the article: - Approved - Approved with reservations - Not approved | Invited Reviewers | |||||| |---|---|---|---|---|---|---| | 1 | 2 | 3 | 4 | 5 | 6 | | | Version 2 (revision) 28 Nov 25 | read | read | read | read | read | | | Version 1 18 Jun 25 | read | read | read | read | read | read | - Maria Teresa Brito, The University of Edinburgh College of Medicine and Veterinary Medicine, Edinburgh, UKAndrew Stevenson, University of Edinburgh, Edinburgh, UKJiayue Gu, The University of Edinburgh College of Medicine and Veterinary Medicine, Edinburgh, UK - Douglas Brash, Yale School of Medicine, New Haven, USA - Richard McKenzie, National Institute of Water and Atmospheric Research Lauder Atmospheric Research Station (Ringgold ID: 563277), Lauder, New Zealand; Earth Sciences, New Zealand, Lauder, New ZealandBen Liley, Earth Sciences New Zealand, Lauder, New ZealandJames Liley, University of Durham, Durham, UK Sign up for content alerts You are now signed up to receive this alert Alongside their report, reviewers assign a status to the article: Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit. Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list: Examples of 'Non-Financial Competing Interests' - Within the past 4 years, you have held joint grants, published or collaborated with any of the authors of the selected paper. - You have a close personal relationship (e.g. parent, spouse, sibling, or domestic partner) with any of the authors. - You are a close professional associate of any of the authors (e.g. scientific mentor, recent student). - You work at the same institute as any of the authors. - You hope/expect to benefit (e.g. favour or employment) as a result of your submission. - You are an Editor for the journal in which the article is published. Examples of 'Financial Competing Interests' - You expect to receive, or in the past 4 years have received, any of the following from any commercial organisation that may gain financially from your submission: a salary, fees, funding, reimbursements. - You expect to receive, or in the past 4 years have received, shared grant support or other funding with any of the authors. - You hold, or are currently applying for, any patents or significant stocks/shares relating to the subject matter of the paper you are commenting on. Sign up for content alerts and receive a weekly or monthly email with all newly published articles Register with NIHR Open Research Already registered? Sign in close Error If you are a previous or current NIHR award holder, sign up for information about developments, publishing and publications from NIHR Open Research. We'll keep you updated on any major new updates to NIHR Open Research Sign In If you've forgotten your password, please enter your email address below and we'll send you instructions on how to reset your password. Email us for further assistance. The email address should be the one you originally registered with F1000. Email address not valid, please try again You registered with F1000 via Google, so we cannot reset your password. To sign in, please click here. If you still need help with your Google account password, please click here. You registered with F1000 via Facebook, so we cannot reset your password. To sign in, please click here. If you still need help with your Facebook account password, please click here. Code not correct, please try again Server error, please try again. If your email address is registered with us, we will email you instructions to reset your password. If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance. Please wait...