Genetically versus environmentally influenced obesity and risk of mortality and non-communicable diseases: A cohort study from the UK Biobank

Genetically versus environmentally influenced obesity and risk of mortality and non-communicable diseases: A cohort study from the UK Biobank | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Genetically versus environmentally influenced obesity and risk of mortality and non-communicable diseases: A cohort study from the UK Biobank Elsa Ojalehto Lindfors, Shayan Mostafaei, Martin Nakash, Chandra A Reynolds, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8296230/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Previous research indicate that obesity is less harmful in people with genetic predisposition to high body mass index (BMI), compared to obesity driven by other factors such as environment or lifestyle. Yet differences between genetically and environmentally influenced adiposity in relation to health outcomes remain unexplored, and have not examined adiposity measures beyond BMI. Therefore, we examined differences between genetically versus environmentally influenced adiposity, measured with BMI as well as waist-hip ratio (WHR), in relation to risk of mortality and key non-communicable diseases. Methods We followed 484,858 UK Biobank participants (aged 40–69 years at baseline) over on average 13 years. Baseline BMI and WHR categories were tested in interaction with polygenic scores (PGS) for respective trait, to distinguish between high adiposity influenced by genetic predisposition (obesity or high WHR and high PGS) versus by environmental factors (obesity or high WHR but low PGS). Risk of all-cause mortality, cardiovascular disease (CVD), cancers, chronic respiratory disease (CRD), and diabetes were modelled in Cox proportional hazard regression, adjusting for age, sex, ethnicity, and socioeconomic factors.. Results The PGS for BMI interacted with obesity, such that genetic predisposition to higher BMI attenuated obesity associations with CVD, diabetes, and CRD, but strengthened that with cancer. In contrast, the PGS for WHR had the opposite effect, and genetic predisposition to higher WHR instead amplified associations between high WHR and mortality, CVD, diabetes, and CRD. Conclusions Obesity was linked to lower risk for most outcomes in people with genetic predisposition to higher BMI, compared to those with genetic predisposition to a low BMI. This indicates that genetically influenced obesity may be less detrimental than obesity influenced by environmental factors sucha as lifestyle. In contrast, the opposite was seen when adiposity was measured as WHR, where the association between high WHR and most outcomes was stronger in people with genetic predisposition to higher WHR. This highlights that BMI and WHR capture distinct adiposity profiles with opposing genetic effects, and underscores the heterogeneity in obesity. Obesity Adiposity Polygenic score Genetic predisposition Non-communicable disease Mortality Epidemiology Figures Figure 1 Figure 2 Introduction Despite ongoing efforts, obesity rates continue to rise, contributing to disability and worsening health outcomes. Non-communicable diseases (NCDs) such as cardiovascular disease (CVD), cancers, chronic respiratory diseases (CRD), and diabetes—responsible for 41 million deaths annually, or 74% of global mortality( 1 ) —are strongly associated with obesity, and obesity is projected to become the number one preventable risk factor for NCDs by 2035( 1 ). Obesity is a complex phenotype shaped by both genetic and environmental influences( 2 , 3 ), and recent research has distinguished between obesity influenced by genetic predisposition versus by other factors such as environment or lifestyle. Studies from the Swedish Twin Registry, the Health and Retirement Study, and the UK Biobank, suggest that individuals with obesity have a lower risk of cardiometabolic( 4 – 6 ) and cognitive diseases( 7 , 8 ) as well as mortality( 9 ) if they have genetic predisposition to a higher body mass index (BMI). This indicates that obesity may be less detrimental in individuals who are genetically predisposed to a higher body mass, compared to those genetically predisposed to a lean body mass, where obesity likely results from other factors such as environment or lifestyle. However, previous studies have focused exclusively on BMI, and no studies to date have examined whether similar genotype–phenotype discordance exists for other adiposity measures such as waist-hip ratio (WHR). Notably, genetic influences on BMI and WHR reflect distinct biological mechanisms: variants associated with BMI are mainly related to the central nervous system, whereas those linked to WHR, particularly WHR adjusted for BMI, are associated with adipose tissue and metabolic processes( 10 – 12 ). In summary, while findings consistently indicate lower disease risk in obesity driven by genetic factors compared to obesity influenced mainly by other factors such as environment and lifestyle, the evidence remains limited: all prior studies have focused solely on BMI rather than other measures of adiposity, and have been based primarily on the Swedish Twin Registry( 4 , 7 ) and the Health and Retirement Study( 5 , 7 – 9 ) with only a few health outcomes examined. To address this knowledge gap, we leveraged data from the UK Biobank to investigate associations between genetically versus environmentally influenced adiposity in relation to all-cause mortality and NCDs. We used the World Health Organization (WHO) definition of the four most common causes of NCD deaths; namely, CVD, cancers, CRD, and diabetes (including all types of diabetes)( 13 ). Given there is also an association between a low BMI, especially in late-life, and adverse health( 14 , 15 ), we additionally examined associations between underweight and the outcomes. In addition to BMI categories, we incorporated WHR as a complementary measure of adiposity to provide a more comprehensive assessment of genetically versus environmentally influenced adiposity in relation to associations with multiple health outcomes. Methods Study population. An overview of the study design is shown in Fig. 1 . We used data from the UK Biobank, a large prospective population-based cohort( 16 ). It includes more than 500,000 individuals aged 40–69 years at recruitment between 2006 and 2010 across 22 assessment centers throughout the UK( 16 ). At baseline, participants completed a touchscreen questionnaire and a brief computer-assisted interview. The participants also underwent a range of physical measurements, and collection of blood, urine, and saliva samples were made. Self-reported information on sociodemographic characteristics, family history, psychosocial factors, environmental factors, lifestyle, medical history, and medication use was collected systematically( 16 , 17 ). The UK Biobank combines baseline data with longitudinal follow-up by linking to multiple national health related datasets, including death registers and hospital episode statistics( 16 , 18 ). Through this linkage, it is possible to track progression, disease incidence, and mortality across many conditions with reliable accuracy for e.g. cardiovascular events( 18 ). The current study included all participants with available adiposity measures (BMI and/or WHR), genotype data, and information on mortality and disease diagnoses from register linkage. Participants with missing baseline date or key covariates were excluded. To reduce the influence of extreme outliers, participants with a BMI below 15 kg/m² or above 55 kg/m², and implausible WHR values ( 2.00) were coded as missing. Out of 502,617 UK Biobank participants, 484,858 were included for analyses. Participants with prevalent disease at baseline were excluded from respective analyses. A flow chart detailing participant inclusion and exclusions is provided in Supplementary Figure S1 . BMI and WHR measurement. Adiposity was assessed using two measures, BMI and WHR. BMI was calculated as weight in kilograms divided by the square of their height in meters (kg/m²) and categorized according to WHO: underweight (< 18.5), normal weight (18.5–24.9; reference category), overweight (25.0–29.9), and obesity (≥ 30.0). WHR was derived as waist circumference divided by hip circumference, with high WHR defined according to WHO as > 0.85 in women and > 0.90 in men. To evaluate alternative categorizations of WHR and examine non-linearity of associations, we additionally modeled sex-specific quartiles (Supplementary Table S1 ), corresponding more closely to the four BMI categories, and examined their associations with mortality (as a measure of general health in late-life, that demonstrate a J-shaped association with BMI( 14 )). Mortality risk increased across quartiles with no evidence of a J-shaped association, supporting our decision to retain the binary WHR classification for interpretability. Polygenic scores for BMI and WHR. Genotype data were available for 488,377 UK Biobank participants, and the genotyping was described in detail by Bycroft and colleagues( 17 ). Careful quality control was conducted by the UK Biobank, based on which we removed samples with high missingness or heterozygosity (n = 968) or close genetic kinship to other participants (> 10 3rd degree relatives; n = 188). Genetically predicted BMI and WHR were constructed with a polygenic score (PGS) for respective trait (PGS BMI and PGS WHR ). A PGS is created by, for each individual, summing up genetic variants across the genome, weighted by their effect size from a genome-wide association study (GWAS) of the trait. The UK Biobank data were included in the most recent GWAS of BMI and WHR( 10 , 11 ), and to avoid inflation through overlap between the discovery and target data we therefore used the earlier GWAS of the traits. The GWAS of BMI by Locke et al. Included data on 339,224 individuals and identified 97 BMI-associated loci( 19 ). The GWAS of WHR by Shungin et al.( 20 ) Included 224,459 individuals and identified 49 genetic variants associated with WHR adjusted for BMI. The GWAS also analyzed WHR, without adjustment for BMI, which was used in the current study to estimate genetically predicted WHR. Prior to calculating the PGS in the UK Biobank data, the summary statistics from each GWAS was processed with SBayesR to handle correlations between genetic variants (linkage disequilibrium)( 21 ), using ~ 1 million HapMap3 variants. Prior to analyses, each PGS was standardized to mean = 0 and standard deviation (SD) = 1, so that results represent change in disease risk per standard deviation higher PGS. To test the predictive ability, each PGS was modelled as a predictor of respective trait in a linear regression model. Non-communicable diseases ( NCDs) In line with the WHO, NCDs were defined as chronic conditions that are not transmitted from person to person and represent leading causes of global mortality and morbidity. Specifically, we focused on four major NCD groups prioritized by WHO: CVD, cancers, CRD, and diabetes( 13 ). These disease groups account for most NCD-related deaths worldwide and have well-established links with obesity and metabolic risk factors( 1 , 13 ). In addition, specific subtypes were examined: ischemic heart disease, cerebrovascular disease, chronic obstructive pulmonary disease (COPD), and type 2 diabetes (T2D). We also examined the six most common forms of cancer, according to the International Agency for Research on Cancer: breast, colorectal, stomach, lung, prostate, and skin cancer( 22 ). Outcome diagnoses were identified through linkage to hospital inpatient records, with primary and secondary diagnoses coded according to International Classification of Diseases (ICD) 10 codes (Supplementary Table S2). Individuals with a first diagnosis of the outcome prior to baseline were excluded from respective analyses. In addition, self-reported diagnoses are available, reported in diagnostic codes (Supplementary Table S3), and were used to exclude individuals with prevalent disease at baseline. Covariates All analyses were adjusted for a range of baseline covariates: age at baseline (in years), sex (female/male), educational attainment (high, intermediate, low), UK Biobank assessment center (22 centers across Scotland, England, and Wales)( 16 ), ethnicity/ancestry (White, Asian or Asian British, Black or Black British, Mixed, Other, or Unknown), and the Townsend deprivation index (score corresponding to the output area in which their postcode is located). All models including a PGS were also adjusted for the first 10 genetic principal components (PCs; provided by the UK Biobank) to account for population stratification. In sensitivity analyses, we additionally adjusted for lifestyle factors: smoking status (never, former, current), alcohol consumption, physical activity, and overall dietary quality. Alcohol consumption was coded in ordinal categories as 1 = Daily or almost daily, 2 = Three or four times a week, 3 = Once or twice a week, 4 = Once or twice a month, 5 = Never, 6 = Occasionally, 7 = Previous drinker. Physical activity was quantified as metabolic equivalent task hours per week (MET-h/week), with participants reporting on average ~ 46 MET-h/week, equivalent to approximately 6–12 hours of moderate-intensity activity per week, indicating that most participants met or exceeded WHO physical activity recommendations. A healthy diet score was derived based on ten dietary components (fruit, vegetables, whole grains, fish, dairy, vegetable oils, refined grains, red meat, processed meat, and sugar-sweetened beverages), following the scoring approach described by Zhuang et al.( 23 ). Statistical analyses All statistical analyses were performed using R version R-4.5.1( 24 ) and RStudio-2025.09.0.( 25 ). Data management and analyses were carried out using the haven( 26 ), dplyr( 27 ), and survival( 28 ), and plots were created with ggplot2( 29 ) and gridExtra( 30 ). We applied Cox proportional hazard models with age as the underlying timescale, following individuals from baseline to end of follow-up (December 31, 2022), diagnosis of respective outcome (except when mortality was the outcome), or death, whichever occurred first. For each adiposity measure (BMI and WHR) and outcome (mortality, CVD, cancer, CRD, and diabetes), the following set of models were applied: 1) separate models of either the adiposity measure or the PGS; 2) a mutually adjusted model including both the adiposity measure and corresponding PGS; and 3) an interaction model, additionally including an interaction term between the adiposity measure and corresponding PGS. Model 3 thereby examines if the association between the adiposity measure and outcome differ as a function of genetic predisposition to higher adiposity (interpreted per 1 SD increase in the PGS). All models were adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including a PGS were also adjusted for the first ten PCs. The proportional hazards assumption was assessed visually by plotting survival curves by exposure and covariate levels, and statistically with the cox.zph function in R and did not indicate violations of the assumption. Corrections for multiple testing were not performed due to the correlated nature of the exposure and outcome variables, and the sample size difference between outcomes. Rather than focusing on the much debated p-values( 31 ), we recommend a holistic view of the analyses and results while being mindful of potential false positive findings. To visualize the interactions between the adiposity measures and PGS, the predict function (survival package) was used based on Model 3 to estimate the probability (and corresponding 95% CI) of each event at age 65 for individuals with obesity or high WHR, with the corresponding PGS set to the mean level and 1–2 SD below and above the mean. Covariates were set to the reference values: White males of European genetic ancestry, assessed in England and with mean levels of education, Townsend deprivation index, and each PC. The following sensitivity analyses were applied: first, additional covariate adjustment was applied by including lifestyle factors (smoking status, alcohol intake, physical activity, and diet score) in the models. Second, interactions between each PGS and the other adiposity measure were tested (BMI categories in interaction with the PGS WHR , and WHR in interaction with the PGS BMI ). Third, subtype outcome analyses were performed to explore potential heterogeneity in associations. For cancer, separate models were estimated for breast, colorectal, stomach, lung, prostate, and skin cancer. For other NCD outcomes, we distinguished ischemic heart disease and cerebrovascular disease from CVD, T2D from all types of diabetes, and COPD from CRD. Fourth, sex-stratified and age-stratified analyses were conducted, repeating all models separately for females and males, and within midlife (ages 49–59) and late life (ages 60–69) groups. Finally, genetically stratified analyses were conducted within participants of European and non-European ancestry. Results Description of the study population. Table 1 describes the total analytical sample and individuals who developed each of the respective outcomes during follow-up. The average age was 56.5 (SD 8.1) years at baseline and 70.0 (SD 8.0) at end of follow-up. Number of events and mean follow-up time for each outcome and exposure combination is shown in Table 2 and Table 3 . Of the 484,858 participants in the analytical sample, 2,472 (0.5%) were individuals with underweight, 153,058 (31.6%) with normal weight, 208,341 (43.0%) with overweight, and 120,777 (24.9%) with obesity. Based on sex-specific cut-offs for WHR, 245,876 (50.7%) participants were categorized as having normal WHR and 238,980 (49.3%) as high WHR. One SD higher PGS BMI was associated with 2.32 kg/m² higher measured BMI (95% CI 2.30–2.33), and one SD higher PGS WHR was associated with 0.044 units higher measured WHR (95% CI 0.043–0.044). Associations were slightly stronger among participants with European ancestry (BMI: β = 2.32, 95% CI 2.31–2.34; WHR: β = 0.0449, 95% CI 0.0447–0.0450), but comparable in those with non-European ancestry (BMI: β = 2.30, 95% CI 2.26–2.33; WHR: β = 0.0385, 95% CI 0.0381–0.0389). Table 1 Descriptive statistics for the total study population and individuals diagnosed with respective outcomes during the follow-up. All Mortality CVD Cancer CRD Diabetes N 484,858 42,350 85,937 68,928 53,619 22,656 Female sex, N (%) 263,006 (54.2%) 17,321 (40.9%) 43,542 (50.7%) 32,197 (46.7%) 25, 471 (47.5%) 9, 948 (43.9%) Education, N (%) - Low 82,617 (17.3%) 12,953 (31.1%) 16,787 (19.8%) 14,950 (22%) 13, 879 (26.3%) 6479 (29.2%) - Intermediate 23,9311 (50%) 18,922 (45.5%) 42,568 (50.3%) 32,982 (48.5%) 25, 576 (48.5%) 11,015 (49.7%) - High 156,791 (32.8%) 9747 (23.4%) 25,326 (29.9%) 20,081 (29.5%) 13, 255 (25.1%) 4671 (21.1%) Age at baseline, M (SD) 56.5 (8.1) 61.7 (6.4) 58.1 (7.7) 59.8 ( 7 ) 59.5 (7.3) 58.8 (7.5) Age at end of follow up or death/ diagnosis, M (SD) 70 (8.0) 71 (7.6) 65.4 (8.5) 65.8 (8.3) 67.6 (8.1) 67.1 (8.1) BMI at baseline, M (SD) 27.4 (4.7) 28.3 (5.3) 27.4 (4.6) 27.7 (4.8) 28.1 (5.0) 31.1 (5.4) WHR at baseline, M (SD) 0.87 (0.09) 0.91 (0.09) 0.88 (0.09) 0.89 (0.09) 0.90 (0.09) 0.93 (0.08) PGS BMI , M (SD) 0.00 (1.00) 0.21 (1.04) 0.01 (0.98) 0.06 (0.99) 0.13 (1.02) 0.51 (1.02) PGS WHR , M (SD) 0.00 (1.00) 0.38 (0.94) 0.11 (0.96) 0.21 (0.96) 0.24 (0.96) 0.45 (0.91) Smokers, N (%) - Never smoked 264,085 (54.7%) 16,341 (38.9%) 43,895 (51.4%) 33,259 (48.5%) 22,803 (42.8%) 10,199 (45.4%) - Previous smoker 167,497 (34.7%) 17,754 (42.3%) 30,607 (35.8%) 26,950 (39.3%) 21,128 (39.7%) 8843 (39.3%) - Current smoker 50,850 (10.5%) 7913 (18.8%) 10,930 (12.8%) 8339 (12.2%) 9320 (17.5%) 3436 (15.3%) Alcohol use, M (SD) 2.99 (1.63) 3.11 (1.76) 2.98 (1.65) 2.89 (1.62) 3.08 (1.71) 3.46 (1.78) Physical activity (MET-h/week), M (SD) 46.16 (63.62) 43.88 (65.21) 48.64 (68.01) 46.38 (64.97) 47.51 (69.61) 43.61 (69.99) Diet score, M (SD) 46.53 (9.07) 46.23 (9.03) 46.41 (9.04) 46.38 (8.8) 46.17 (9.03) 45.7 (9.11) Townsend deprivation index, M (SD) -1.31 (3.09) -0.77 (3.37) -1.31 (3.09) -1.34 (3.09) -0.57 (3.36) -0.49 (3.42) Descriptive statistics for all individuals included for analyses and stratified by mortality and non-communicable disease outcomes defined by WHO: cardiovascular disease (CVD), cancer, chronic respiratory disease, and diabetes. Statistics are presented as number (%) for categorical variables and mean (standard deviation) for continuous variables. All values reflect baseline measurements except age at end of follow-up. BMI and WHR were assessed using measured weight, height, waist, and hip circumference. PGS BMI and PGS WHR refer to standardized polygenic scores for body mass index and waist-hip ratio, respectively. Alcohol was coded as seven ordinal categories ranging from daily/almost daily to previous consumption. Physical activity was measured in metabolic equivalent task hours per week (MET-h/week), with the mean level of approximately 46 MET-h/week corresponding to about 6–12 hours of moderate-intensity activity per week. The diet score is based on a healthy diet index ranging from 0 to 100. The Townsend deprivation index reflects area-level socioeconomic deprivation, with higher values indicating greater deprivation. Abbreviations: BMI, body mass index; CRD, chronic respiratory disease; CVD, cardiovascular disease; MET, metabolic equivalent of task; N, number; PGS, polygenic score; SD, standard deviation; WHR, waist-to-hip ratio. Measured and genetically predicted adiposity in relation to risk of mortality and NCDs Obesity was associated with higher risk of all outcomes (Table 2 ), while overweight was associated with higher risk of CVD, CRD, and diabetes, but a lower risk of mortality. Underweight was associated with a higher risk of mortality, CVD, and CRD. The PGS BMI was associated with higher risk of all outcomes, except for mortality where the association was inversed. In mutually adjusted models, the associations between obesity and the outcomes were robust, while those between the PGS BMI and the outcomes were attenuated and even inverse for diabetes and CRD. Both WHR and the PGS WHR were associated with higher risk of all outcomes, with attenuated associations in mutually adjusted models (Table 3 ). Further adjustment for lifestyle factors, including smoking, alcohol consumption, physical activity, and diet, had negligible impact on the associations between BMI, WHR, their respective polygenic scores, and the outcomes (Table S4). Genetically versus environmentally influenced adiposity in relation to risk of mortality and NCDs In the interaction models, the hazard ratios (HRs) for each interaction term reflects the change in the association between the adiposity measure and the outcomes per one standard deviation higher PGS. As such, for obesity, an interaction term with HR < 1 means an attenuated association with higher PGS BMI and indicates that the association with disease risk is stronger for obesity influenced by environmental factors (obesity despite a low PGS BMI ) than obesity influenced by genetic predisposition (obesity and a high PGS BMI ; Fig. 1 ) In contrast, an interaction term with HR > 1 means a stronger association with higher PGS BMI and indicates the opposite: that the risk of disease is higher in genetically influenced obesity, compared to obesity influenced predominantly by environmental factors (Fig. 1 ). Genetic predisposition to higher BMI modified the associations between obesity and the outcomes: it attenuated the relationships between obesity and CVD, CRD, and diabetes (interaction term with HR 1). In Fig. 2 , visualizing predicted probabilities of each outcome in an individual with obesity, this is seen as a higher risk for CVD, CRD, and diabetes at lower levels of the PGS BMI , but higher risk of cancer at higher levels of the PGS BMI . For overweight, genetic predisposition to higher BMI similarly increased the association with cancer, while no evidence of interaction was observed for underweight across outcomes. Additional adjustment for lifestyle factors had little influence on the associations (Table S4). For WHR, genetic predisposition to higher WHR strengthened the associations with all outcomes except cancer, where no interaction was observed. In all cases, interaction terms were above 1, indicating that the association between high WHR and the outcomes were more pronounced among individuals genetically predisposed to higher WHR (Table 3 ). In Fig. 2 , this is seen as a higher probability of the outcomes in an individual with high WHR if they have higher levels of the PGS WHR . As for BMI, adjusting for lifestyle factors did not affect the associations (Table S4). To further test genetic influences on associations between adiposity and the outcomes, we examined cross-trait influences of polygenic scores: BMI category in interaction with the PGS WHR and high WHR in interaction with the PGS BMI (Table S5). Both PGS influenced associations between the other adiposity measure and the outcomes in a similar manner as with their corresponding traits. While no evidence of interaction was observed between obesity and PGS WHR for any of the outcomes, the association of overweight with mortality, CVD and diabetes was stronger among individuals with genetically predicted higher WHR (interaction terms with HR > 1; Table S5). The interaction with the PGS BMI attenuated the associations between high WHR and CVD and diabetes but strengthened the association with cancer, while no statistically significant interactions were observed for mortality or CRD. Table 2 Risk of mortality and NCDs in relation to BMI and a polygenic score for BMI. Mortality (N: 484,648) CVD (N: 306,904) Cancer (N: 442,063) CRD (N: 403,203) Diabetes (N: 459,688) N of events (%) 42,294 (8.7%) 85,899 (28%) 68, 896 (15.6%) 53,580 (13.3%) 22,624 (4.9%) Mean follow up (SD) 13.51 (1.89) 11.94 (3.75) 12.76 (3.03) 12.92 (2.78) 13.3 (2.25) Separate models Underweight 2.45 (2.20–2.72) 1.16 (1.06–1.28) 1.04 (0.92–1.17) 1.71 (1.53–1.90) 1.07 (0.77–1.51) Normal weight Ref Ref Ref Ref Ref Overweight 0.95 (0.92–0.97) 1.23 (1.21–1.25) 1.02 (1.00-1.03) 1.05 (1.02–1.07) 2.48 (2.36–2.60) Obesity 1.27 (1.24–1.31) 1.73 (1.70–1.76) 1.07 (1.05–1.09) 1.36 (1.33–1.39) 7.05 (6.72–7.38) PGS BMI 0.99 (0.98-1.00) 1.11 (1.10–1.12) 1.02 (1.01–1.03) 1.02 (1.01–1.03) 1.31 (1.29–1.33) Mutually adjusted models Underweight 2.46 (2.21–2.73) 1.17 (1.06–1.28) 1.04 (0.92–1.17) 1.71 (1.53–1.91) 1.07 (0.76–1.50) Normal weight Ref Ref Ref Ref Ref Overweight 0.98 (0.95–1.01) 1.21 (1.19–1.23) 1.01 (0.99–1.03) 1.07 (1.05–1.10) 2.56 (2.44–2.70) Obesity 1.33 (1.29–1.37) 1.70 (1.66–1.73) 1.06 (1.04–1.09) 1.40 (1.36–1.44) 7.36 (7.00-7.74) PGS BMI 0.96 (0.95–0.97) 1.02 (1.01–1.03) 1.01 (1.00-1.02) 0.97 (0.96–0.98) 0.96 (0.94–0.98) Interaction models Underweight 2.45 (2.18–2.75) 1.22 (1.10–1.36) 1.02 (0.89–1.18) 1.75 (1.55–1.98) 0.95 (0.64–1.43) Normal weight Ref Ref Ref Ref Ref Overweight 0.97 (0.95-1.00) 1.19 (1.17–1.21) 1.02 (1.00-1.04) 1.06 (1.04–1.09) 2.33 (2.21–2.46) Obesity 1.34 (1.30–1.39) 1.71 (1.68–1.75) 1.06 (1.04–1.09) 1.40 (1.37–1.44) 7.48 (7.10–7.88) PGS BMI 0.95 (0.93–0.97) 1.04 (1.03–1.06) 0.98 (0.97-1.00) 0.98 (0.96–1.01) 1.07 (1.02–1.13) Underweight * PGS BMI 0.99 (0.87–1.13) 1.10 (0.99–1.24) 0.97 (0.84–1.12) 1.05 (0.92–1.19) 0.78 (0.52–1.15) Overweight * PGS BMI 1.03 (1.00-1.06) 0.99 (0.97-1.00) 1.03 (1.01–1.05) 0.98 (0.96–1.01) 1.00 (0.94–1.05) Obesity * PGS BMI 1.00 (0.96–1.03) 0.93 (0.91–0.95) 1.04 (1.02–1.07) 0.96 (0.94–0.99) 0.80 (0.76–0.85) Risk of mortality and NCDs in relation to measured BMI and PGS BMI, presented as hazard ratios and 95% confidence intervals. BMI was categorized into underweight, normal weight (reference), overweight, and obesity. The PGS BMI was standardized to reflect associations per one standard deviation higher PGS. Separate models include either BMI categories or the PGS BMI ; mutually adjusted models include BMI categories and the PGS BMI in the same model; interaction models additionally include interaction terms between each BMI category and the PGS BMI . All models are adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including the PGS were additionally adjusted for the first ten genetic principal components (PC1–PC10). Statistically significant associations (at α = 0.05) are presented in bold. Abbreviations: BMI, body mass index; CRD, chronic respiratory disease; CVD, cardiovascular disease; HR, Hazard ratio; NCD, non-communicable disease, PGS, polygenic score. Table 3 Risk of mortality and NCDs in relation to WHR and a polygenic score for WHR. Mortality (N: 484,856) CVD (N: 306,977) Cancer (N: 442,247) CRD (N: 403,334) Diabetes (N: 459,848) N of events (%) 42,350 (8.7%) 85,937 (28%) 68, 928 (15.6%) 53,619 (13.3%) 22,656 (4.9%) Mean follow up (SD) 13.51 (1.89) 11.94 (3.75) 12.76 (3.03) 12.92 (2.78) 13.3 (2.25) Separate models Normal WHR Ref Ref Ref Ref Ref High WHR 1.32 (1.29–1.35) 1.36 (1.34–1.38) 1.11 (1.09–1.13) 1.30 (1.28–1.33) 3.46 (3.35–3.59) PGS WHR 1.14 (1.12–1.15) 1.13 (1.12–1.14) 1.03 (1.02–1.03) 1.11 (1.10–1.12) 1.61 (1.59–1.64) Mutually adjusted models Normal WHR Ref Ref Ref Ref Ref High WHR 1.20 (1.17–1.23) 1.26 (1.24–1.29) 1.11 (1.08–1.13) 1.22 (1.19–1.25) 2.49 (2.40–2.59) PGS WHR 1.09 (1.08–1.10) 1.07 (1.06–1.08) 1.00 (0.99–1.01) 1.06 (1.05–1.07) 1.35 (1.33–1.37) Interaction models Normal WHR Ref Ref Ref Ref Ref High WHR 1.21 (1.18–1.25) 1.26 (1.24–1.29) 1.11 (1.08–1.13) 1.22 (1.19–1.25) 2.50 (2.40–2.59) PGS WHR 1.02 (1.00-1.05) 1.05 (1.04–1.07) 1.00 (0.98–1.01) 1.03 (1.02–1.05) 1.31 (1.27–1.35) High WHR * PGS WHR 1.10 (1.07–1.12) 1.03 (1.01–1.05) 1.01 (0.99–1.03) 1.05 (1.02–1.07) 1.04 (1.00-1.08) Risk of mortality and NCDs in relation to measured WHR and PGS WHR , presented as hazard ratios and 95% confidence intervals. WHR was categorized as normal (reference) or high using sex-specific thresholds (≤ 0.90 for men and ≤ 0.85 for women). The PGS WHR was standardized to reflect associations per one standard deviation higher PGS. Separate models include WHR categories or the PGS WHR ; mutually adjusted models include WHR categories and the PGS WHR in the same model; interaction models additionally include interaction terms between WHR category and the PGS WHR . All models were adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including the PGS were additionally adjusted for the first ten genetic principal components (PC1–PC10). Statistically significant associations (at α = 0.05) are presented in bold. Abbreviations: CRD, chronic respiratory disease; CVD, cardiovascular disease; HR, Hazard ratio; NCD, non-communicable disease; PGS, polygenic score; WHR, waist-to-hip ratio. Differences in associations between adiposity and disease subtypes Analyses of cancer subtypes, though limited by lower power, showed that overweight and obesity were associated with higher risk of breast, colorectal, and stomach cancer, but lower risk of lung, prostate, and skin cancer (Table S6). High WHR was associated with higher risk of breast, lung, colorectal, and stomach cancer, while no clear associations were observed for prostate or skin cancer (Table S6). Across other NCD subtypes, associations were broadly consistent with the main analyses but varied in magnitude (Table S7). Obesity, as well as the obesity–PGS BMI interaction, was stronger for ischemic heart disease than for cerebrovascular disease, and weaker for chronic obstructive pulmonary disease than for CRD. Results for T2D were comparable to those observed for all diabetes in the main analyses. Associations between high WHR and the subtypes were similar to the main analyses. Interactions between high WHR and PGS WHR were overall attenuated and statistically significant only for COPD. Predicted probability and 95% confidence intervals of the outcomes at age 70, for individuals with obesity by level of genetic predisposition to higher BMI (PGS BMI ) in the left panels, and individuals with high WHR by level of genetic predisposition to higher WHR (PGS WHR ) in the right panels. Predictions are based on Cox proportional hazard models adjusted for age, sex, ethnicity, education, assessment center, Townsend deprivation index, and genetic principal components. Predictions represent those for a White male of European genetic ancestry, assessed in England and with mean levels of education, deprivation index, and genetic principal components. Abbreviations: BMI body mass index; WHR waist–hip ratio; PGS polygenic score. Differences in associations between adiposity and the outcomes across sex, age, and genetic ancestry Associations between obesity and the outcomes were generally comparable in sex-stratified models (Tables S8–S9), but slightly stronger in females than in males. Associations between high WHR or the PGS WHR and the outcomes were similar in females and males, but the interaction between WHR and the PGS WHR (interaction term with HR > 1) was only present among males. Among females, all interaction terms were close to 1, except for diabetes where a higher PGS WHR instead reduced the association between high WHR and disease risk (interaction HR: 0.92, 95% CI 0.88–0.96). In age-stratified analyses, associations between midlife measures of adiposity (ages 49–59 at baseline) and the outcomes were largely consistent with the main results (Table S10). For adiposity measures taken in late life (ages 60–69), associations remained similar but were overall weaker (Table S11). Across genetic ancestry groups, (Tables S12–S13), associations between adiposity measures and disease outcomes were consistent in both European and non-European ancestry groups, although estimates in the non-European group were less precise due to smaller sample size. Discussion We investigated how phenotypic and genetically predicted adiposity interact in relation to risk of mortality and the four most common causes of death by NCDs: CVD, cancers, CRD, and diabetes. This study extends prior work on genetically versus environmentally influenced obesity by, for the first time, additionally examining WHR and genetic predisposition to WHR in relation to multiple NCDs. Genetic predisposition to higher BMI attenuated the associations of obesity with CVD, CRD, and diabetes, but strengthened the association between obesity and cancer, a phenotype not previously examined in this context. In contrast, the association of a high WHR with risk of mortality, CVD, CRD, and diabetes was stronger in those with a higher genetically predicted WHR. These findings indicate a complex interaction between adiposity and genetic predisposition. For most outcomes, obesity is less detrimental in individuals with genetic predisposition to a higher BMI, while genetic predisposition to higher WHR instead amplifies associations between adiposity and risk of mortality and NCDs. That genetic predisposition to higher BMI attenuated the associations of obesity with several of the outcomes is consistent with the concept of genotype–phenotype discordance, where exceeding one’s genetically predicted BMI confers greater health risks. Prior studies have reported similar patterns: data from the Swedish Twin Registry showed stronger associations of midlife obesity with dementia and CVD in those genetically predisposed to a lower BMI( 4 , 7 ), and findings from the Health and Retirement Study indicated similar discordance effects for mortality, cardiovascular outcomes, and cognitive decline( 5 , 8 , 9 ). More recently, findings from the UK Biobank data demonstrated that individuals whose measured BMI exceeded their genetically predicted BMI had increased risk of diabetes( 6 ). Together with our results, these studies suggest that obesity driven, at least partly, by genetic factors, may be less detrimental than obesity influenced by other factors such as environment and lifestyle. In contrast, genetic predisposition to a higher WHR strengthened the association between high WHR and the outcomes. This may reflect that BMI and WHR not only capture different aspects of adiposity but are driven by distinct genetic influences. Previous work show that genetic variants associated with BMI act mainly through the central nervous system through regulation of appetite and energy metabolism, while those associated with WHR act through adipose tissue and the digestive system( 12 ). Moreover, some genetic variants associated with BMI are associated with a rather favorable metabolic profile( 32 , 33 ). Winkler and colleagues could separate 159 genetic variants associated with BMI, WHR, or WHR adjusted for BMI into four groups: Genetic variants associated with higher BMI and higher WHR (n = 82), only with higher BMI (n = 25), with higher BMI but lower WHR (n = 24), or only with lower WHR (n = 28)( 12 ). Thus, while many genetic variants simultaneously increase both BMI and WHR, the PGS BMI and PGS WHR will also capture unique genetic influences as well as variants with opposing effects on the two adiposity measures. Highlighting that the PGS BMI and PGS WHR are distinct, sensitivity analyses of cross-trait interactions between each adiposity measures and a PGS for the other measure were largely consistent with the effect of the PGS in the main analyses: a high BMI, at least in the overweight range, was more detrimental in individuals with genetic predisposition to a higher WHR, and a high WHR less detrimental in individuals with genetic predisposition to higher BMI. A theoretical model of body weight postulates an inherent set point around which body weight is biologically controlled from the brain, through feedback mechanisms from the periphery( 34 ). However, environmental influences can override biological control, resulting in body weight outside the set point range with greater metabolic dysregulation and increased disease risk( 34 ). It is plausible that genetically influenced obesity is less detrimental because it is driven, at least partly, by a high body weight set point, while environmentally influenced obesity is a result of overridden biological control of body weight, resulting in more adverse health effects. As mentioned, genetic influences on WHR are less related to the central nervous system, and instead act predominantly through adipose tissue and the digestive system( 12 ). Central adiposity indicates visceral adipose tissue with fat deposit stored in and around organs, which is linked to the adverse metabolic consequences of obesity( 35 ). Thus, while genetic predisposition to a higher BMI may be related to a higher inherent set point, acting through the central nervous system, genetic predisposition to higher WHR may act more directly on adipose tissue and body fat distribution, leading to metabolic dysfunction and exacerbation of the adverse effects of a high WHR. This may explain why genetic predisposition to higher BMI and WHR have opposing effects on the associations between adiposity and adverse health. Pronounced sex differences were observed in the interactions between WHR and genetic predisposition in relation to the outcomes. While genetic predisposition to higher WHR amplified the association between WHR and all outcomes in males, the interaction between high WHR and PGS WHR was significant only for diabetes in females, where it weakened the association between high WHR and diabetes. These findings align with previous evidence showing extensive sexual dimorphism in the genetic architecture of WHR, with most WHR-associated loci exhibiting stronger effects in women but distinct metabolic consequences across fat depots( 35 ). Recent work modeling the genomic architecture of adiposity across the lifespan also reported sex-specific loadings for latent factors representing abdominal size (where WHR loads) and general adiposity (where BMI loads)( 36 ), further supporting sex differences in adiposity-related genetic influences. The interplay between sex hormones and adipose tissue regulation may contribute to these patterns, underscoring the importance of accounting for sex-specific genetic and physiological factors when assessing the metabolic effects of body fat distribution. It is also possible that menopausal status contributed to these sex-specific patterns, as fat distribution typically shifts toward central adiposity after menopause( 37 ). In contrast to the lower risk of CVD, CRD, and diabetes in genetically influenced obesity, our findings indicate a higher cancer risk among individuals with obesity if they also have genetic predisposition to higher BMI. The divergent patterns between cancer and cardiometabolic outcomes may reflect disease-specific mechanisms: metabolic and hemodynamic pathways explain the robust positive relationship with cardiometabolic outcomes, whereas hormonal regulation, reverse causality (e.g., pre-diagnostic weight loss), and lifestyle interactions such as smoking complicate associations with certain cancers( 38 , 39 ). Prior research has shown that adiposity does not affect all cancer types uniformly, and the International Agency for Research on Cancer, part of WHO, defines 13 obesity-related cancers, which include postmenopausal breast cancer, colorectal cancer, and stomach cancer, but not lung, prostate, or skin cancer( 40 , 41 ). In line with these differences, our sensitivity analyses of the six most common cancers demonstrated associations between obesity and higher risk of the obesity-related cancers, but lower risk of lung, prostate, and skin cancer. It should be noted that high WHR was associated with higher risk of all cancer types except for skin cancer, where no association was seen. In line with this, several meta-analyses and Mendelian randomization studies report inverse associations between BMI and lung cancer( 42 – 44 ), whereas measures of abdominal adiposity, such as WHR, show null or positive associations( 43 ). While results should be interpreted with caution due to differences in statistical power, there were no clear differences in the effect of interaction between the adiposity measures and PGS between subtypes. Underweight was also associated with higher risk of mortality, CVD, and CRD in the current study. In late life or among individuals with pre-existing chronic or degenerative conditions, a higher BMI is often linked to a reduced risk of adverse health outcomes, possibly due to reverse causation rather than a protective effect of higher BMI( 14 , 15 ). Previous work indicates that associations between low BMI and adverse health also differ by genetic predisposition. We demonstrated that in late life, a higher BMI was associated with a lower risk of dementia, but only among those genetically predicted to a high BMI( 7 ). Similarly, in a recent study of individuals with chronic obstructive pulmonary disease (COPD), those with a measured BMI lower than their genetically predicted BMI had the highest risk of mortality( 45 ). In contrast, there was no evidence of an interaction between underweight and the PGS BMI in the current study. It is plausible that this lack of a difference is due to the relatively younger sample, where reverse causation due to unintentional weight loss is less evident than in older samples. However, associations between underweight and the outcomes were comparable in sensitivity analyses stratifying the sample into measures taken in midlife versus late-life, with no indications of interactions in either age group. Associations between higher BMI and disease outcomes as well as interactions with the PGS BMI were generally weaker in the late-life sample however, possibly reflecting the influence of pre-existing illness, loss of muscle mass, or selective survival among older individuals with obesity. In contrast, WHR showed more stable associations across age groups, which is expected given that fat distribution tends to shift with age, with relatively greater central adiposity( 46 ). These findings underscore the limitations of BMI as a measure of adiposity in older populations, where it may not fully capture body composition or metabolic risk. The main strengths of this study lie in its large, population-based design with detailed phenotypic and genotypic characterization, permitting robust evaluation of interactions between measured and genetically predicted adiposity across mortality and four prioritized NCDs defined by the WHO. The inclusion of both BMI and WHR, together with their respective PGS, allowed a comprehensive comparison between overall and central adiposity, providing novel insights into their distinct genetic and metabolic underpinnings. The availability of harmonized covariate data further strengthened the analyses, enabling multiple sensitivity and stratified models, including testing lifestyle influences. However, the observational design limits causal interpretation, and BMI and WHR were measured only at baseline and do not capture weight or fat distribution changes over time. While follow-up of disease diagnoses through healthcare registers is a strength, it should be noted that inpatient data may best capture more severe cases, while missing milder forms of disease diagnosed in primary care( 47 ). Additionally, information about menopause and other relevant information was not available during follow-up, limiting investigations of the causes of observed sex-differences. In addition, despite the large sample size, statistical power was more limited for sensitivity analyses, particularly for certain subtypes, such as cancer outcomes, and comparisons should be made with caution. The PGS, while informative, explain only a fraction of the heritable component of adiposity, and residual confounding from unmeasured lifestyle or socioeconomic factors cannot be excluded. Nevertheless, results indicated clear differences between the PGS BMI and PGS WHR , which were robust in cross-trait analyses. It should also be noted that the PGS are most powerful to predict respective trait in samples of European ancestry. While ancestry-stratified analyses were overall robust, the UK Biobank primarily includes individuals of European genetic ancestry, and the generalizability of the findings to more diverse populations remain to be established. In summary, while it is important to note that obesity was associated with increased risk of all outcomes regardless of genetic predisposition, the association with CVD, CRD, and diabetes weakened with genetic predisposition to higher BMI. This indicates that obesity influenced by genetic predisposition may be less detrimental than obesity influenced by environmental factors. For cancer, the direction of the interaction differed from the other outcomes and indicated a higher risk in those with obesity and genetic predisposition to a higher BMI. When adiposity was measured as high WHR, the associations with mortality, CVD, CRD, and diabetes were instead stronger with genetic predisposition to higher WHR. These findings emphasize that BMI and measure of body fat distribution represent distinct phenotypes, both on the phenotypic and genetic level. Future research should further disentangle the shared and unique genetic influences on BMI and WHR to clarify how genetic susceptibility interacts with environmental exposures in shaping the heterogeneity of adiposity and its consequences for health. Declarations Ethics approval and consent to participate: All individuals in the sample gave written informed consent to participate in UKB and for data to be used in future research. The UKB study has ethical approval from the Northwest Multi-Centre Research Ethics Committee. The current study was approved by the Swedish Ethical Review Authority, DNR 2024-03706-01. Consent for publication: Not applicable. Competing interests: The authors declare no competing interests. Funding: The study was funded by the Swedish Research Council for Health, Working Life and Welfare (Forte; 2022 − 00672); the Strategic Research Program in Epidemiology (SFOepi) at Karolinska Institutet; Karolinska Institutet's Research Foundation (2022 − 01718, 2024–02898); Loo and Hans Osterman's Foundation (2022 − 01222, 2023 − 01855, 2024–02197); and the Foundation for Geriatric Diseases at Karolinska Institutet (2022 − 01296, 2023 − 01854, 2024–02116). Author Contribution EOL, CAR, YZ, ADA, JJ, and IK contributed to conception or design of the work. SH contributed to data acquisition. EOL, SM, and IK contributed to the statistical analyses. All authors contributed to interpretation of the results. EOL, MN, and IK drafted the first version of the manuscript, and all authors contributed to critical revision and approved the final version for submission. Acknowledgement We are grateful to UK Biobank participants and research team, without whom the research would not have been possible. Data Availability This research was conducted using the UK Biobank resource under the Application Number 22224. UK Biobank data are available upon application to the UK Biobank (https:/www.ukbiobank.ac.uk ). The summary statistics used for the polygenic score construction are publicly available from the GWAS publications(19, 20).The analytical plan for the current study was preregistered on the Open Science Framework (OSF; https:/osf.io/k3qxy ) in January 2025. The analysis plan includes descriptions of the study aims, hypotheses, exposure and outcome definitions, and planned statistical analyses. Final codes used for data processing and analysis are also available at the Open Science Framework, along with the preregistration. References Hildebrand S, Pfeifer A. The obesity pandemic and its impact on non-communicable disease burden. Pflugers Arch. 2025;477(5):657–68. Silventoinen K, Jelenkovic A, Sund R, Yokoyama Y, Hur Y-M, Cozen W, et al. Differences in genetic and environmental variation in adult BMI by sex, age, time period, and region: an individual-based pooled analysis of 40 twin cohorts. Am J Clin Nutr. 2017;106(2):457–66. Min J, Chiu DT, Wang Y. Variation in the heritability of body mass index based on diverse twin studies: a systematic review. Obes Rev. 2013;14(11):871–82. Ojalehto E, Zhan Y, Jylhävä J, Reynolds CA, Dahl Aslan AK, Karlsson IK. Genetically and environmentally predicted obesity in relation to cardiovascular disease: a nationwide cohort study. EClinicalMedicine. 2023;58:101943. Davidson T, Vinneau-Palarino J, Goode JA, Boardman JD. Utilizing genome wide data to highlight the social behavioral pathways to health: The case of obesity and cardiovascular health among older adults. Soc Sci Med. 2021;273:113766. Rhee T-M, Choi J, Lee H, Merino J, Park J-B, Kwak SH. Discrepancy Between Genetically Predicted and Observed BMI Predicts Incident Type 2 Diabetes. Diabetes Care. 2024;47(10):1826–33. Karlsson IK, Lehto K, Gatz M, Reynolds CA, Dahl Aslan AK. Age-dependent effects of body mass index across the adult life span on the risk of dementia: a cohort study with a genetic approach. BMC Med. 2020;18(1):131. Karlsson IK, Gatz M, Arpawong TE, Dahl Aslan AK, Reynolds CA. The dynamic association between body mass index and cognition from midlife through late-life, and the effect of sex and genetic influences. Sci Rep. 2021;11(1):7206. Vinneau JM, Huibregtse BM, Laidley TM, Goode JA, Boardman JD. Mortality and Obesity Among U.S. Older Adults: The Role of Polygenic Risk. J Gerontol B Psychol Sci Soc Sci. 2021;76(2):343–7. Yengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, et al. Meta-analysis of genome-wide association studies for height and body mass index in ∼700000 individuals of European ancestry. Hum Mol Genet. 2018;27(20):3641–9. Pulit SL, Stoneman C, Morris AP, Wood AR, Glastonbury CA, Tyrrell J, et al. Meta-analysis of genome-wide association studies for body fat distribution in 694 649 individuals of European ancestry. Hum Mol Genet. 2019;28(1):166–74. Winkler TW, Günther F, Höllerer S, Zimmermann M, Loos RJ, Kutalik Z, et al. A joint view on genetic variants for adiposity differentiates subtypes with distinct metabolic implications. Nat Commun. 2018;9(1):1946. World Health Organization. Global status report on noncommunicable diseases 2014. 2014. Carslake D, Davey Smith G, Gunnell D, Davies N, Nilsen TIL, Romundstad P. Confounding by ill health in the observed association between BMI and mortality: evidence from the HUNT Study using offspring BMI as an instrument. Int J Epidemiol. 2018;47(3):760–70. Dye L, Boyle NB, Champ C, Lawton C. The relationship between obesity and cognitive health and decline. Proc Nutr Soc. 2017;76(4):443–54. Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12(3):e1001779. Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. Kivimäki M, Batty GD, Singh-Manoux A, Britton A, Brunner EJ, Shipley MJ. Validity of Cardiovascular Disease Event Ascertainment Using Linkage to UK Hospital Records. Epidemiology. 2017;28(5):735–9. Locke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR, et al. Genetic studies of body mass index yield new insights for obesity biology. Nature. 2015;518(7538):197–206. Shungin D, Winkler TW, Croteau-Chonka DC, Ferreira T, Locke AE, Mägi R, et al. New genetic loci link adipose and insulin biology to body fat distribution. Nature. 2015;518(7538):187–96. Lloyd-Jones LR, Zeng J, Sidorenko J, Yengo L, Moser G, Kemper KE, et al. Improved polygenic prediction by Bayesian multiple regression on summary statistics. Nat Commun. 2019;10(1):5086. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49. Zhuang P, Liu X, Li Y, Wan X, Wu Y, Wu F, et al. Effect of Diet Quality and Genetic Predisposition on Hemoglobin A1c and Type 2 Diabetes Risk: Gene-Diet Interaction Analysis of 357,419 Individuals. Diabetes Care. 2021;44(11):2470–9. R Core Team. R: A Language and Environment for Statistical Computing [Internet]. R Foundation for Statistical Computing. 2025. Available from: https://www.r-project.org Posit team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: Posit Software, PBC. 2025. Available from: https://posit.co Wickham H, Miller E, Smith D. haven: Import and Export 'SPSS', 'Stata' and 'SAS' Files. R package version 2.5.5 ed2025. Wickham H, François R, Henry L, Müller K, Vaughan D. dplyr: A Grammar of Data Manipulation. R package version 1.1.4 ed2025. Therneau TM, until TL, Elizabeth A, survival Cynthia C. Survival Analysis. 2024. Wickham H. ggplot2: Elegant Graphics for Data Analysis. 2016. Auguie B, Antonov A, gridExtra. Miscellaneous Functions for Grid Graphics. 2017. Boscardin CK, Sewell JL, Tolsgaard MG, Pusic MV. How to Use and Report on p-values. Perspect Med Educ. 2024;13(1):250–4. Kim D, Highland HM, Smit RAJ, Hysong MR, Buchanan VL, Young KL, et al. Genetic underpinnings of the heterogeneous impact of obesity on lipid levels and cardiovascular disease. Genome Med. 2025;17(1):113. Chami N, Wang Z, Svenstrup V, Obrero VD, Hemerich D, Huang Y, et al. Genetic subtyping of obesity reveals biological insights into the uncoupling of adiposity from its cardiometabolic comorbidities. Nat Med. 2025;31(11):3801–12. Müller MJ, Geisler C, Heymsfield SB, Bosy-Westphal A. Recent advances in understanding body weight homeostasis in humans. F1000Res. 2018;7:F1000-Faculty Rev-25. Sulc J, Winkler TW, Heid IM, Kutalik Z. Heterogeneity in Obesity: Genetic Basis and Metabolic Consequences. Curr Diab Rep. 2020;20(1):1. Arehart CH, Lin M, Gibson RA, Colorado Center for Personalized M, Raghavan S, Gignoux CR, et al. Modeling the genomic architecture of adiposity and anthropometrics across the lifespan. Nat Commun. 2025;16(1):7494. Toth MJ, Tchernof A, Sites CK, Poehlman ET. Menopause-related changes in body fat distribution. Ann N Y Acad Sci. 2000;904:502–6. Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 2008;371(9612):569–78. Bai T, Wu C. Association of cardiovascular disease on cancer: observational and mendelian randomization analyses. Sci Rep. 2024;14(1):28465. Larsson SC, Spyrou N, Mantzoros CS. Body fatness associations with cancer: evidence from recent epidemiological studies and future directions. Metabolism. 2022;137:155326. Lauby-Secretan B, Scoccianti C, Loomis D, Grosse Y, Bianchini F, Straif K, et al. Body Fatness and Cancer–Viewpoint of the IARC Working Group. N Engl J Med. 2016;375(8):794–8. Yang Y, Dong J, Sun K, Zhao L, Zhao F, Wang L, et al. Obesity and incidence of lung cancer: a meta-analysis. Int J Cancer. 2013;132(5):1162–9. Hidayat K, Du X, Chen G, Shi M, Shi B. Abdominal Obesity and Lung Cancer Risk: Systematic Review and Meta-Analysis of Prospective Studies. Nutrients. 2016;8(12):810. Zhou W, Liu G, Hung RJ, Haycock PC, Aldrich MC, Andrew AS, et al. Causal relationships between body mass index, smoking and lung cancer: Univariable and multivariable Mendelian randomization. Int J Cancer. 2021;148(5):1077–86. Zhang J, Moll M, Hobbs BD, Bakke P, Regan EA, Xu H, et al. Genetically Predicted Body Mass Index and Mortality in Chronic Obstructive Pulmonary Disease. Am J Respir Crit Care Med. 2024;210(7):890–9. Kuk JL, Saunders TJ, Davidson LE, Ross R. Age-related changes in total and regional fat distribution. Ageing Res Rev. 2009;8(4):339–48. Wilkinson T, Schnier C, Bush K, Rannikmäe K, Henshall DE, Lerpiniere C, et al. Identifying dementia outcomes in UK Biobank: a validation study of primary care, hospital admissions and mortality data. Eur J Epidemiol. 2019;34(6):557–65. Additional Declarations No competing interests reported. Supplementary Files UKBPRSSupplementaryMaterial20251205.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8296230","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":569744609,"identity":"38234916-7e51-4619-a940-088df9648317","order_by":0,"name":"Elsa Ojalehto Lindfors","email":"","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":false,"prefix":"","firstName":"Elsa","middleName":"Ojalehto","lastName":"Lindfors","suffix":""},{"id":569744611,"identity":"aca4a38c-eb8d-453c-b68a-80a6e1986cb3","order_by":1,"name":"Shayan Mostafaei","email":"","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":false,"prefix":"","firstName":"Shayan","middleName":"","lastName":"Mostafaei","suffix":""},{"id":569744613,"identity":"07225062-2e4f-4df4-97c2-63894675a7e9","order_by":2,"name":"Martin Nakash","email":"","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Nakash","suffix":""},{"id":569744614,"identity":"3ca3e53f-3b22-4d37-8f41-ccde387edfe0","order_by":3,"name":"Chandra A Reynolds","email":"","orcid":"","institution":"University of Colorado Boulder","correspondingAuthor":false,"prefix":"","firstName":"Chandra","middleName":"A","lastName":"Reynolds","suffix":""},{"id":569744616,"identity":"4251b104-587d-4198-aa5d-bf8f692254b6","order_by":4,"name":"Yiqiang Zhan","email":"","orcid":"","institution":"Sun Yat- Sen University","correspondingAuthor":false,"prefix":"","firstName":"Yiqiang","middleName":"","lastName":"Zhan","suffix":""},{"id":569744618,"identity":"99068f6c-658c-4e09-9294-be358402c401","order_by":5,"name":"Anna Dahl Aslan","email":"","orcid":"","institution":"University of Skövde","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"Dahl","lastName":"Aslan","suffix":""},{"id":569744619,"identity":"e96e6865-5a5c-4410-8194-b61c6f62ba0a","order_by":6,"name":"Juulia Jylhävä","email":"","orcid":"","institution":"University of Tampere","correspondingAuthor":false,"prefix":"","firstName":"Juulia","middleName":"","lastName":"Jylhävä","suffix":""},{"id":569744620,"identity":"1cbb41a6-e9c6-4e7f-983c-339ddee0d3b8","order_by":7,"name":"Sara Hägg","email":"","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Hägg","suffix":""},{"id":569744624,"identity":"c28a5b43-d22f-4742-a0a3-cb1545b4ad13","order_by":8,"name":"Ida K Karlsson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYBACNoYEBsYGBgYZMO/DAQnitfCAOIwziNHCgKyFmecAERr42JMfMM6ouccj33/42GObMxYM8u0NBBzG88yAccOxYh6DG2npxjk3JBgMzhCwik0iwYDxAVsCj4EEj5l0zgegFokEQlrSPzA++JcAdNgZM2kLoBb5+Q8IackxYNzYlsDDcCDHTJoB6DCGG/h1AP3ypuDgzL4EkF/SJHvOSPAYnCHgMPn29I0Pe74lyIFCTOLHsTo5+fYDBKwBAhQlPITVj4JRMApGwSggCAAamT7QoI1tqgAAAABJRU5ErkJggg==","orcid":"","institution":"Karolinska Institutet","correspondingAuthor":true,"prefix":"","firstName":"Ida","middleName":"K","lastName":"Karlsson","suffix":""}],"badges":[],"createdAt":"2025-12-06 17:38:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8296230/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8296230/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99692174,"identity":"5e1b8abb-649a-4962-b903-138f6de66ad0","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":616732,"visible":true,"origin":"","legend":"","description":"","filename":"MSUKBPRS20251205.docx","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/d60e6b1d11a965bf2452c008.docx"},{"id":99692169,"identity":"e42de767-b2d4-477d-b876-e243b8358940","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11208,"visible":true,"origin":"","legend":"","description":"","filename":"3d3fba435b3948c49534bea728aa7232.json","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/a2fd34c5ea08e0dba662087a.json"},{"id":99692175,"identity":"05e66f93-24df-40b9-9ab8-11ce42faafeb","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":135678,"visible":true,"origin":"","legend":"","description":"","filename":"UKBPRSSupplementaryMaterial20251205.docx","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/a6816b89f3c103826a654ef5.docx"},{"id":99795776,"identity":"cd4cc636-48e2-4281-9362-83e8da91d55f","added_by":"auto","created_at":"2026-01-08 13:39:41","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156375,"visible":true,"origin":"","legend":"","description":"","filename":"3d3fba435b3948c49534bea728aa72321enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/2c8d0fcecac3917ac7b359ef.xml"},{"id":99797073,"identity":"f05548ea-b2ad-49c5-bf1a-21418584f9a4","added_by":"auto","created_at":"2026-01-08 13:44:32","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86287,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/b0f5a6abbd6318323ae96339.png"},{"id":99796800,"identity":"f6f53d32-2534-4b5e-bb65-a8f38e016a7e","added_by":"auto","created_at":"2026-01-08 13:43:42","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37793,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/6ee171ae855cb0079127cb79.png"},{"id":99692178,"identity":"4b2fd509-d244-49cb-b471-dd5f082c735b","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153240,"visible":true,"origin":"","legend":"","description":"","filename":"3d3fba435b3948c49534bea728aa72321structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/34507c7b0665214b2a71586a.xml"},{"id":99692177,"identity":"106da260-e917-476f-8970-d2b8ce7e152f","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"html","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":165732,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/c5f4078bc6cd057d11ee556a.html"},{"id":99692168,"identity":"ac47c5a2-45ba-4377-9015-47493db27f90","added_by":"auto","created_at":"2026-01-07 10:30:07","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":250670,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy overview. \u003c/strong\u003eWe conducted a prospective cohort study based on the UK Biobank, with participants aged 40–69 years at baseline. BMI and WHR were measured during the initial assessment and used together with PGS for respective trait to differentiate between high adiposity influenced by genetic predisposition (obesity or high WHR and high PGS) versus by other factors such as environment or lifestyle (obesity or high WHR but low PGS). Mortality and four prioritized non-communicable diseases were ascertained through linkage with national health registers through 2022. Obesity and high WHR was modelled in interaction with the corresponding PGS in relation to risk of each outcome, to examine differences in associations between genetically and environmentally influenced adiposity. Abbreviations: BMI, Body mass index; PGS, Polygenic score; WHR, Waist–hip ratio. Created with BioRender.com.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/d0f65429134db976596ae379.png"},{"id":99796476,"identity":"f4784109-f325-410f-b8d9-8f41a06ec280","added_by":"auto","created_at":"2026-01-08 13:42:06","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":122423,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProbability of mortality and non-communicable diseases in people with obesity or high waist-hip ratio, by genetic predisposition to high BMI and WHR.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePredicted probability and 95% confidence intervals of the outcomes at age 70, for individuals with obesity by level of genetic predisposition to higher BMI (PGS\u003csub\u003eBMI\u003c/sub\u003e) in the left panels, and individuals with high WHR by level of genetic predisposition to higher WHR (PGS\u003csub\u003eWHR\u003c/sub\u003e) in the right panels. Predictions are based on Cox proportional hazard models adjusted for age, sex, ethnicity, education, assessment center, Townsend deprivation index, and genetic principal components. Predictions represent those for a White male of European genetic ancestry, assessed in England and with mean levels of education, deprivation index, and genetic principal components. Abbreviations: BMI body mass index; WHR waist–hip ratio; PGS polygenic score.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/7fef540bcf6e530a8537478e.png"},{"id":105752666,"identity":"87bfb393-5a80-431f-93b9-66a3cab053d5","added_by":"auto","created_at":"2026-03-30 16:04:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2016692,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/df591785-05f2-4c08-8319-0dc1cb8a5bdd.pdf"},{"id":99795224,"identity":"cdeb5730-373d-41b2-af7d-3499e42c3af5","added_by":"auto","created_at":"2026-01-08 13:37:29","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":135678,"visible":true,"origin":"","legend":"","description":"","filename":"UKBPRSSupplementaryMaterial20251205.docx","url":"https://assets-eu.researchsquare.com/files/rs-8296230/v1/800b4654ab2771207bfb3625.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Genetically versus environmentally influenced obesity and risk of mortality and non-communicable diseases: A cohort study from the UK Biobank","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDespite ongoing efforts, obesity rates continue to rise, contributing to disability and worsening health outcomes. Non-communicable diseases (NCDs) such as cardiovascular disease (CVD), cancers, chronic respiratory diseases (CRD), and diabetes\u0026mdash;responsible for 41\u0026nbsp;million deaths annually, or 74% of global mortality(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) \u0026mdash;are strongly associated with obesity, and obesity is projected to become the number one preventable risk factor for NCDs by 2035(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eObesity is a complex phenotype shaped by both genetic and environmental influences(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e), and recent research has distinguished between obesity influenced by genetic predisposition versus by other factors such as environment or lifestyle. Studies from the Swedish Twin Registry, the Health and Retirement Study, and the UK Biobank, suggest that individuals with obesity have a lower risk of cardiometabolic(\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) and cognitive diseases(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e) as well as mortality(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e) if they have genetic predisposition to a higher body mass index (BMI). This indicates that obesity may be less detrimental in individuals who are genetically predisposed to a higher body mass, compared to those genetically predisposed to a lean body mass, where obesity likely results from other factors such as environment or lifestyle. However, previous studies have focused exclusively on BMI, and no studies to date have examined whether similar genotype\u0026ndash;phenotype discordance exists for other adiposity measures such as waist-hip ratio (WHR). Notably, genetic influences on BMI and WHR reflect distinct biological mechanisms: variants associated with BMI are mainly related to the central nervous system, whereas those linked to WHR, particularly WHR adjusted for BMI, are associated with adipose tissue and metabolic processes(\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn summary, while findings consistently indicate lower disease risk in obesity driven by genetic factors compared to obesity influenced mainly by other factors such as environment and lifestyle, the evidence remains limited: all prior studies have focused solely on BMI rather than other measures of adiposity, and have been based primarily on the Swedish Twin Registry(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) and the Health and Retirement Study(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e) with only a few health outcomes examined. To address this knowledge gap, we leveraged data from the UK Biobank to investigate associations between genetically versus environmentally influenced adiposity in relation to all-cause mortality and NCDs. We used the World Health Organization (WHO) definition of the four most common causes of NCD deaths; namely, CVD, cancers, CRD, and diabetes (including all types of diabetes)(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Given there is also an association between a low BMI, especially in late-life, and adverse health(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e), we additionally examined associations between underweight and the outcomes. In addition to BMI categories, we incorporated WHR as a complementary measure of adiposity to provide a more comprehensive assessment of genetically versus environmentally influenced adiposity in relation to associations with multiple health outcomes.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e \u003cb\u003eStudy population.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAn overview of the study design is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. We used data from the UK Biobank, a large prospective population-based cohort(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). It includes more than\u003c/p\u003e \u003cp\u003e500,000 individuals aged 40\u0026ndash;69 years at recruitment between 2006 and 2010 across 22 assessment centers throughout the UK(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). At baseline, participants completed a touchscreen questionnaire and a brief computer-assisted interview. The participants also underwent a range of physical measurements, and collection of blood, urine, and saliva samples were made. Self-reported information on sociodemographic characteristics, family history, psychosocial factors, environmental factors, lifestyle, medical history, and medication use was collected systematically(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). The UK Biobank combines baseline data with longitudinal follow-up by linking to multiple national health related datasets, including death registers and hospital episode statistics(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Through this linkage, it is possible to track progression, disease incidence, and mortality across many conditions with reliable accuracy for e.g. cardiovascular events(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe current study included all participants with available adiposity measures (BMI and/or WHR), genotype data, and information on mortality and disease diagnoses from register linkage. Participants with missing baseline date or key covariates were excluded. To reduce the influence of extreme outliers, participants with a BMI below 15 kg/m\u0026sup2; or above 55 kg/m\u0026sup2;, and implausible WHR values (\u0026lt;\u0026thinsp;0.40 or \u0026gt;\u0026thinsp;2.00) were coded as missing. Out of 502,617 UK Biobank participants, 484,858 were included for analyses. Participants with prevalent disease at baseline were excluded from respective analyses. A flow chart detailing participant inclusion and exclusions is provided in Supplementary Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eBMI and WHR measurement.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAdiposity was assessed using two measures, BMI and WHR. BMI was calculated as weight in kilograms divided by the square of their height in meters (kg/m\u0026sup2;) and categorized according to WHO: underweight (\u0026lt;\u0026thinsp;18.5), normal weight (18.5\u0026ndash;24.9; reference category), overweight (25.0\u0026ndash;29.9), and obesity (\u0026ge;\u0026thinsp;30.0). WHR was derived as waist circumference divided by hip circumference, with high WHR defined according to WHO as \u0026gt;\u0026thinsp;0.85 in women and \u0026gt;\u0026thinsp;0.90 in men.\u003c/p\u003e \u003cp\u003eTo evaluate alternative categorizations of WHR and examine non-linearity of associations, we additionally modeled sex-specific quartiles (Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), corresponding more closely to the four BMI categories, and examined their associations with mortality (as a measure of general health in late-life, that demonstrate a J-shaped association with BMI(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e)). Mortality risk increased across quartiles with no evidence of a J-shaped association, supporting our decision to retain the binary WHR classification for interpretability.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePolygenic scores for BMI and WHR.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eGenotype data were available for 488,377 UK Biobank participants, and the genotyping was described in detail by Bycroft and colleagues(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Careful quality control was conducted by the UK Biobank, based on which we removed samples with high missingness or heterozygosity (n\u0026thinsp;=\u0026thinsp;968) or close genetic kinship to other participants (\u0026gt;\u0026thinsp;10 3rd degree relatives; n\u0026thinsp;=\u0026thinsp;188).\u003c/p\u003e \u003cp\u003eGenetically predicted BMI and WHR were constructed with a polygenic score (PGS) for respective trait (PGS\u003csub\u003eBMI\u003c/sub\u003e and PGS\u003csub\u003eWHR\u003c/sub\u003e). A PGS is created by, for each individual, summing up genetic variants across the genome, weighted by their effect size from a genome-wide association study (GWAS) of the trait. The UK Biobank data were included in the most recent GWAS of BMI and WHR(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e), and to avoid inflation through overlap between the discovery and target data we therefore used the earlier GWAS of the traits. The GWAS of BMI by Locke et al. Included data on 339,224 individuals and identified 97 BMI-associated loci(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). The GWAS of WHR by Shungin et al.(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) Included 224,459 individuals and identified 49 genetic variants associated with WHR adjusted for BMI. The GWAS also analyzed WHR, without adjustment for BMI, which was used in the current study to estimate genetically predicted WHR. Prior to calculating the PGS in the UK Biobank data, the summary statistics from each GWAS was processed with SBayesR to handle correlations between genetic variants (linkage disequilibrium)(\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e), using\u0026thinsp;~\u0026thinsp;1\u0026nbsp;million HapMap3 variants. Prior to analyses, each PGS was standardized to mean\u0026thinsp;=\u0026thinsp;0 and standard deviation (SD)\u0026thinsp;=\u0026thinsp;1, so that results represent change in disease risk per standard deviation higher PGS. To test the predictive ability, each PGS was modelled as a predictor of respective trait in a linear regression model.\u003c/p\u003e \u003cp\u003e \u003cb\u003eNon-communicable diseases\u003c/b\u003e (\u003cb\u003eNCDs)\u003c/b\u003e\u003c/p\u003e \u003cp\u003eIn line with the WHO, NCDs were defined as chronic conditions that are not transmitted from person to person and represent leading causes of global mortality and morbidity. Specifically, we focused on four major NCD groups prioritized by WHO: CVD, cancers, CRD, and diabetes(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). These disease groups account for most NCD-related deaths worldwide and have well-established links with obesity and metabolic risk factors(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). In addition, specific subtypes were examined: ischemic heart disease, cerebrovascular disease, chronic obstructive pulmonary disease (COPD), and type 2 diabetes (T2D). We also examined the six most common forms of cancer, according to the International Agency for Research on Cancer: breast, colorectal, stomach, lung, prostate, and skin cancer(\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Outcome diagnoses were identified through linkage to hospital inpatient records, with primary and secondary diagnoses coded according to International Classification of Diseases (ICD) 10 codes (Supplementary Table S2). Individuals with a first diagnosis of the outcome prior to baseline were excluded from respective analyses. In addition, self-reported diagnoses are available, reported in diagnostic codes (Supplementary Table S3), and were used to exclude individuals with prevalent disease at baseline.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCovariates\u003c/h2\u003e \u003cp\u003eAll analyses were adjusted for a range of baseline covariates: age at baseline (in years), sex (female/male), educational attainment (high, intermediate, low), UK Biobank assessment center (22 centers across Scotland, England, and Wales)(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e), ethnicity/ancestry (White, Asian or Asian British, Black or Black British, Mixed, Other, or Unknown), and the Townsend deprivation index (score corresponding to the output area in which their postcode is located). All models including a PGS were also adjusted for the first 10 genetic principal components (PCs; provided by the UK Biobank) to account for population stratification.\u003c/p\u003e \u003cp\u003eIn sensitivity analyses, we additionally adjusted for lifestyle factors: smoking status (never, former, current), alcohol consumption, physical activity, and overall dietary quality. Alcohol consumption was coded in ordinal categories as 1\u0026thinsp;=\u0026thinsp;Daily or almost daily, 2\u0026thinsp;=\u0026thinsp;Three or four times a week, 3\u0026thinsp;=\u0026thinsp;Once or twice a week, 4\u0026thinsp;=\u0026thinsp;Once or twice a month, 5\u0026thinsp;=\u0026thinsp;Never, 6\u0026thinsp;=\u0026thinsp;Occasionally, 7\u0026thinsp;=\u0026thinsp;Previous drinker. Physical activity was quantified as metabolic equivalent task hours per week (MET-h/week), with participants reporting on average\u0026thinsp;~\u0026thinsp;46 MET-h/week, equivalent to approximately 6\u0026ndash;12 hours of moderate-intensity activity per week, indicating that most participants met or exceeded WHO physical activity recommendations. A healthy diet score was derived based on ten dietary components (fruit, vegetables, whole grains, fish, dairy, vegetable oils, refined grains, red meat, processed meat, and sugar-sweetened beverages), following the scoring approach described by Zhuang et al.(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStatistical analyses\u003c/h3\u003e\n\u003cp\u003eAll statistical analyses were performed using R version R-4.5.1(\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) and RStudio-2025.09.0.(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). Data management and analyses were carried out using the haven(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e), dplyr(\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e), and survival(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), and plots were created with ggplot2(\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e) and gridExtra(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). We applied Cox proportional hazard models with age as the underlying timescale, following individuals from baseline to end of follow-up (December 31, 2022), diagnosis of respective outcome (except when mortality was the outcome), or death, whichever occurred first. For each adiposity measure (BMI and WHR) and outcome (mortality, CVD, cancer, CRD, and diabetes), the following set of models were applied: 1) separate models of either the adiposity measure or the PGS; 2) a mutually adjusted model including both the adiposity measure and corresponding PGS; and 3) an interaction model, additionally including an interaction term between the adiposity measure and corresponding PGS. Model 3 thereby examines if the association between the adiposity measure and outcome differ as a function of genetic predisposition to higher adiposity (interpreted per 1 SD increase in the PGS). All models were adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including a PGS were also adjusted for the first ten PCs. The proportional hazards assumption was assessed visually by plotting survival curves by exposure and covariate levels, and statistically with the cox.zph function in R and did not indicate violations of the assumption. Corrections for multiple testing were not performed due to the correlated nature of the exposure and outcome variables, and the sample size difference between outcomes. Rather than focusing on the much debated p-values(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e), we recommend a holistic view of the analyses and results while being mindful of potential false positive findings. To visualize the interactions between the adiposity measures and PGS, the predict function (survival package) was used based on Model 3 to estimate the probability (and corresponding 95% CI) of each event at age 65 for individuals with obesity or high WHR, with the corresponding PGS set to the mean level and 1\u0026ndash;2 SD below and above the mean. Covariates were set to the reference values: White males of European genetic ancestry, assessed in England and with mean levels of education, Townsend deprivation index, and each PC.\u003c/p\u003e \u003cp\u003eThe following sensitivity analyses were applied: first, additional covariate adjustment was applied by including lifestyle factors (smoking status, alcohol intake, physical activity, and diet score) in the models. Second, interactions between each PGS and the other adiposity measure were tested (BMI categories in interaction with the PGS\u003csub\u003eWHR\u003c/sub\u003e, and WHR in interaction with the PGS\u003csub\u003eBMI\u003c/sub\u003e). Third, subtype outcome analyses were performed to explore potential heterogeneity in associations. For cancer, separate models were estimated for breast, colorectal, stomach, lung, prostate, and skin cancer. For other NCD outcomes, we distinguished ischemic heart disease and cerebrovascular disease from CVD, T2D from all types of diabetes, and COPD from CRD. Fourth, sex-stratified and age-stratified analyses were conducted, repeating all models separately for females and males, and within midlife (ages 49\u0026ndash;59) and late life (ages 60\u0026ndash;69) groups. Finally, genetically stratified analyses were conducted within participants of European and non-European ancestry.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eDescription of the study population.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e describes the total analytical sample and individuals who developed each of the respective outcomes during follow-up. The average age was 56.5 (SD 8.1) years at baseline and 70.0 (SD 8.0) at end of follow-up. Number of events and mean follow-up time for each outcome and exposure combination is shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eOf the 484,858 participants in the analytical sample, 2,472 (0.5%) were individuals with underweight, 153,058 (31.6%) with normal weight, 208,341 (43.0%) with overweight, and 120,777 (24.9%) with obesity. Based on sex-specific cut-offs for WHR, 245,876 (50.7%) participants were categorized as having normal WHR and 238,980 (49.3%) as high WHR. One SD higher PGS\u003csub\u003eBMI\u003c/sub\u003e was associated with 2.32 kg/m\u0026sup2; higher measured BMI (95% CI 2.30\u0026ndash;2.33), and one SD higher PGS\u003csub\u003eWHR\u003c/sub\u003e was associated with 0.044 units higher measured WHR (95% CI 0.043\u0026ndash;0.044). Associations were slightly stronger among participants with European ancestry (BMI: β\u0026thinsp;=\u0026thinsp;2.32, 95% CI 2.31\u0026ndash;2.34; WHR: β\u0026thinsp;=\u0026thinsp;0.0449, 95% CI 0.0447\u0026ndash;0.0450), but comparable in those with non-European ancestry (BMI: β\u0026thinsp;=\u0026thinsp;2.30, 95% CI 2.26\u0026ndash;2.33; WHR: β\u0026thinsp;=\u0026thinsp;0.0385, 95% CI 0.0381\u0026ndash;0.0389).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDescriptive statistics for the total study population and individuals diagnosed with respective outcomes during the follow-up.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAll\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMortality\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCVD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCancer\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eCRD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e484,858\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42,350\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e85,937\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e68,928\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e53,619\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e22,656\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale sex, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e263,006 (54.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17,321 (40.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43,542 (50.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32,197 (46.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25, 471 (47.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9, 948 (43.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- Low\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82,617 (17.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12,953 (31.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16,787 (19.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14,950 (22%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13, 879 (26.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e6479 (29.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- Intermediate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23,9311 (50%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18,922 (45.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42,568 (50.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32,982 (48.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25, 576 (48.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e11,015 (49.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e156,791 (32.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9747 (23.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e25,326 (29.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20,081 (29.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13, 255 (25.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e4671 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge at baseline, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e56.5 (8.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61.7 (6.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e58.1 (7.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59.8 (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e59.5 (7.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e58.8 (7.5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge at end of follow up or death/ diagnosis, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e70 (8.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71 (7.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65.4 (8.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e65.8 (8.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e67.6 (8.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e67.1 (8.1)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI at baseline, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27.4 (4.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28.3 (5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e27.4 (4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e27.7 (4.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e28.1 (5.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e31.1 (5.4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWHR at baseline, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.87 (0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.91 (0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.88 (0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.89 (0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.90 (0.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.93 (0.08)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eBMI\u003c/sub\u003e, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.00 (1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.21 (1.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.01 (0.98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.06 (0.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.13 (1.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.51 (1.02)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eWHR\u003c/sub\u003e, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.00 (1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.38 (0.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.11 (0.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.21 (0.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.24 (0.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.45 (0.91)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmokers, N (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- Never smoked\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e264,085 (54.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16,341 (38.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e43,895 (51.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33,259 (48.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22,803 (42.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e10,199 (45.4%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- Previous smoker\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e167,497 (34.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e17,754 (42.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30,607 (35.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e26,950 (39.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e21,128 (39.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8843 (39.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e- Current smoker\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e50,850 (10.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7913 (18.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10,930 (12.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8339 (12.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9320 (17.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3436 (15.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlcohol use, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.99 (1.63)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.11 (1.76)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.98 (1.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.89 (1.62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3.08 (1.71)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3.46 (1.78)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhysical activity (MET-h/week), M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46.16 (63.62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e43.88 (65.21)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48.64 (68.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e46.38 (64.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e47.51 (69.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e43.61 (69.99)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiet score, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e46.53 (9.07)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e46.23 (9.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e46.41 (9.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e46.38 (8.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e46.17 (9.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e45.7 (9.11)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTownsend deprivation index, M (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-1.31 (3.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-0.77 (3.37)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-1.31 (3.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-1.34 (3.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.57 (3.36)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-0.49 (3.42)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eDescriptive statistics for all individuals included for analyses and stratified by mortality and non-communicable disease outcomes defined by WHO: cardiovascular disease (CVD), cancer, chronic respiratory disease, and diabetes. Statistics are presented as number (%) for categorical variables and mean (standard deviation) for continuous variables. All values reflect baseline measurements except age at end of follow-up. BMI and WHR were assessed using measured weight, height, waist, and hip circumference. PGS\u003csub\u003eBMI\u003c/sub\u003e and PGS\u003csub\u003eWHR\u003c/sub\u003e refer to standardized polygenic scores for body mass index and waist-hip ratio, respectively. Alcohol was coded as seven ordinal categories ranging from daily/almost daily to previous consumption. Physical activity was measured in metabolic equivalent task hours per week (MET-h/week), with the mean level of approximately 46 MET-h/week corresponding to about 6\u0026ndash;12 hours of moderate-intensity activity per week. The diet score is based on a healthy diet index ranging from 0 to 100. The Townsend deprivation index reflects area-level socioeconomic deprivation, with higher values indicating greater deprivation. Abbreviations: BMI, body mass index; CRD, chronic respiratory disease; CVD, cardiovascular disease; MET, metabolic equivalent of task; N, number; PGS, polygenic score; SD, standard deviation; WHR, waist-to-hip ratio.\u003c/p\u003e\n\u003ch3\u003eMeasured and genetically predicted adiposity in relation to risk of mortality and NCDs\u003c/h3\u003e\n\u003cp\u003eObesity was associated with higher risk of all outcomes (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), while overweight was associated with higher risk of CVD, CRD, and diabetes, but a lower risk of mortality. Underweight was associated with a higher risk of mortality, CVD, and CRD. The PGS\u003csub\u003eBMI\u003c/sub\u003e was associated with higher risk of all outcomes, except for mortality where the association was inversed. In mutually adjusted models, the associations between obesity and the outcomes were robust, while those between the PGS\u003csub\u003eBMI\u003c/sub\u003e and the outcomes were attenuated and even inverse for diabetes and CRD. Both WHR and the PGS\u003csub\u003eWHR\u003c/sub\u003e were associated with higher risk of all outcomes, with attenuated associations in mutually adjusted models (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFurther adjustment for lifestyle factors, including smoking, alcohol consumption, physical activity, and diet, had negligible impact on the associations between BMI, WHR, their respective polygenic scores, and the outcomes (Table S4).\u003c/p\u003e\n\u003ch3\u003eGenetically versus environmentally influenced adiposity in relation to risk of mortality and NCDs\u003c/h3\u003e\n\u003cp\u003eIn the interaction models, the hazard ratios (HRs) for each interaction term reflects the change in the association between the adiposity measure and the outcomes per one standard deviation higher PGS. As such, for obesity, an interaction term with HR\u0026thinsp;\u0026lt;\u0026thinsp;1 means an attenuated association with higher PGS\u003csub\u003eBMI\u003c/sub\u003e and indicates that the association with disease risk is stronger for obesity influenced by environmental factors (obesity despite a low PGS\u003csub\u003eBMI\u003c/sub\u003e) than obesity influenced by genetic predisposition (obesity and a high PGS\u003csub\u003eBMI\u003c/sub\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) In contrast, an interaction term with HR\u0026thinsp;\u0026gt;\u0026thinsp;1 means a stronger association with higher PGS\u003csub\u003eBMI\u003c/sub\u003e and indicates the opposite: that the risk of disease is higher in genetically influenced obesity, compared to obesity influenced predominantly by environmental factors (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGenetic predisposition to higher BMI modified the associations between obesity and the outcomes: it attenuated the relationships between obesity and CVD, CRD, and diabetes (interaction term with HR\u0026thinsp;\u0026lt;\u0026thinsp;1), but strengthened the association with cancer (interaction term with HR\u0026thinsp;\u0026gt;\u0026thinsp;1). In Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, visualizing predicted probabilities of each outcome in an individual with obesity, this is seen as a higher risk for CVD, CRD, and diabetes at lower levels of the PGS\u003csub\u003eBMI\u003c/sub\u003e, but higher risk of cancer at higher levels of the PGS\u003csub\u003eBMI\u003c/sub\u003e. For overweight, genetic predisposition to higher BMI similarly increased the association with cancer, while no evidence of interaction was observed for underweight across outcomes. Additional adjustment for lifestyle factors had little influence on the associations (Table S4).\u003c/p\u003e \u003cp\u003eFor WHR, genetic predisposition to higher WHR strengthened the associations with all outcomes except cancer, where no interaction was observed. In all cases, interaction terms were above 1, indicating that the association between high WHR and the outcomes were more pronounced among individuals genetically predisposed to higher WHR (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, this is seen as a higher probability of the outcomes in an individual with high WHR if they have higher levels of the PGS\u003csub\u003eWHR\u003c/sub\u003e. As for BMI, adjusting for lifestyle factors did not affect the associations (Table S4).\u003c/p\u003e \u003cp\u003eTo further test genetic influences on associations between adiposity and the outcomes, we examined cross-trait influences of polygenic scores: BMI category in interaction with the PGS\u003csub\u003eWHR\u003c/sub\u003e and high WHR in interaction with the PGS\u003csub\u003eBMI\u003c/sub\u003e (Table S5). Both PGS influenced associations between the other adiposity measure and the outcomes in a similar manner as with their corresponding traits. While no evidence of interaction was observed between obesity and PGS\u003csub\u003eWHR\u003c/sub\u003e for any of the outcomes, the association of overweight with mortality, CVD and diabetes was stronger among individuals with genetically predicted higher WHR (interaction terms with HR\u0026thinsp;\u0026gt;\u0026thinsp;1; Table S5). The interaction with the PGS\u003csub\u003eBMI\u003c/sub\u003e attenuated the associations between high WHR and CVD and diabetes but strengthened the association with cancer, while no statistically significant interactions were observed for mortality or CRD.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRisk of mortality and NCDs in relation to BMI and a polygenic score for BMI.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMortality\u003c/p\u003e \u003cp\u003e(N: 484,648)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCVD\u003c/p\u003e \u003cp\u003e(N: 306,904)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCancer\u003c/p\u003e \u003cp\u003e(N: 442,063)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCRD\u003c/p\u003e \u003cp\u003e(N: 403,203)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003cp\u003e(N: 459,688)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN of events (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42,294 (8.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85,899 (28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68, 896 (15.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e53,580 (13.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22,624 (4.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean follow up (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.51 (1.89)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.94 (3.75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12.76 (3.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12.92 (2.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13.3 (2.25)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSeparate models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnderweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e2.45 (2.20\u0026ndash;2.72)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.16 (1.06\u0026ndash;1.28)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.04 (0.92\u0026ndash;1.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.71 (1.53\u0026ndash;1.90)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.07 (0.77\u0026ndash;1.51)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal weight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.95 (0.92\u0026ndash;0.97)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.23 (1.21\u0026ndash;1.25)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.02 (1.00-1.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.05 (1.02\u0026ndash;1.07)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e2.48 (2.36\u0026ndash;2.60)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.27 (1.24\u0026ndash;1.31)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.73 (1.70\u0026ndash;1.76)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.07 (1.05\u0026ndash;1.09)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.36 (1.33\u0026ndash;1.39)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e7.05 (6.72\u0026ndash;7.38)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.99 (0.98-1.00)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.11 (1.10\u0026ndash;1.12)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.02 (1.01\u0026ndash;1.03)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.02 (1.01\u0026ndash;1.03)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.31 (1.29\u0026ndash;1.33)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMutually adjusted models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnderweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e2.46 (2.21\u0026ndash;2.73)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.17 (1.06\u0026ndash;1.28)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.04 (0.92\u0026ndash;1.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.71 (1.53\u0026ndash;1.91)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.07 (0.76\u0026ndash;1.50)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal weight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.98 (0.95\u0026ndash;1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.21 (1.19\u0026ndash;1.23)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.01 (0.99\u0026ndash;1.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.07 (1.05\u0026ndash;1.10)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e2.56 (2.44\u0026ndash;2.70)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.33 (1.29\u0026ndash;1.37)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.70 (1.66\u0026ndash;1.73)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.06 (1.04\u0026ndash;1.09)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.40 (1.36\u0026ndash;1.44)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e7.36 (7.00-7.74)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.96 (0.95\u0026ndash;0.97)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.02 (1.01\u0026ndash;1.03)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.01 (1.00-1.02)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.97 (0.96\u0026ndash;0.98)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e0.96 (0.94\u0026ndash;0.98)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eInteraction models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnderweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e2.45 (2.18\u0026ndash;2.75)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.22 (1.10\u0026ndash;1.36)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.02 (0.89\u0026ndash;1.18)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.75 (1.55\u0026ndash;1.98)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.95 (0.64\u0026ndash;1.43)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal weight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverweight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.97 (0.95-1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.19 (1.17\u0026ndash;1.21)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.02 (1.00-1.04)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.06 (1.04\u0026ndash;1.09)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e2.33 (2.21\u0026ndash;2.46)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.34 (1.30\u0026ndash;1.39)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.71 (1.68\u0026ndash;1.75)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.06 (1.04\u0026ndash;1.09)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.40 (1.37\u0026ndash;1.44)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e7.48 (7.10\u0026ndash;7.88)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.95 (0.93\u0026ndash;0.97)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.04 (1.03\u0026ndash;1.06)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.98 (0.97-1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98 (0.96\u0026ndash;1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.07 (1.02\u0026ndash;1.13)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUnderweight * PGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.99 (0.87\u0026ndash;1.13)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.10 (0.99\u0026ndash;1.24)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.97 (0.84\u0026ndash;1.12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.05 (0.92\u0026ndash;1.19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.78 (0.52\u0026ndash;1.15)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOverweight * PGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.03 (1.00-1.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.99 (0.97-1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.03 (1.01\u0026ndash;1.05)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.98 (0.96\u0026ndash;1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.00 (0.94\u0026ndash;1.05)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObesity * PGS\u003csub\u003eBMI\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.00 (0.96\u0026ndash;1.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.93 (0.91\u0026ndash;0.95)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.04 (1.02\u0026ndash;1.07)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.96 (0.94\u0026ndash;0.99)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e0.80 (0.76\u0026ndash;0.85)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eRisk of mortality and NCDs in relation to measured BMI and PGS\u003csub\u003eBMI,\u003c/sub\u003e presented as hazard ratios and 95% confidence intervals. BMI was categorized into underweight, normal weight (reference), overweight, and obesity. The PGS\u003csub\u003eBMI\u003c/sub\u003e was standardized to reflect associations per one standard deviation higher PGS. Separate models include either BMI categories or the PGS\u003csub\u003eBMI\u003c/sub\u003e; mutually adjusted models include BMI categories and the PGS\u003csub\u003eBMI\u003c/sub\u003e in the same model; interaction models additionally include interaction terms between each BMI category and the PGS\u003csub\u003eBMI\u003c/sub\u003e. All models are adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including the PGS were additionally adjusted for the first ten genetic principal components (PC1\u0026ndash;PC10). Statistically significant associations (at α\u0026thinsp;=\u0026thinsp;0.05) are presented in bold. Abbreviations: BMI, body mass index; CRD, chronic respiratory disease; CVD, cardiovascular disease; HR, Hazard ratio; NCD, non-communicable disease, PGS, polygenic score.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRisk of mortality and NCDs in relation to WHR and a polygenic score for WHR.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMortality\u003c/p\u003e \u003cp\u003e(N: 484,856)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCVD\u003c/p\u003e \u003cp\u003e(N: 306,977)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCancer\u003c/p\u003e \u003cp\u003e(N: 442,247)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCRD\u003c/p\u003e \u003cp\u003e(N: 403,334)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003cp\u003e(N: 459,848)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN of events (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e42,350 (8.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85,937 (28%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e68, 928 (15.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e53,619 (13.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22,656 (4.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean follow up (SD)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.51 (1.89)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.94 (3.75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12.76 (3.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12.92 (2.78)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13.3 (2.25)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSeparate models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.32 (1.29\u0026ndash;1.35)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.36 (1.34\u0026ndash;1.38)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.11 (1.09\u0026ndash;1.13)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.30 (1.28\u0026ndash;1.33)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e3.46 (3.35\u0026ndash;3.59)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eWHR\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.14 (1.12\u0026ndash;1.15)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.13 (1.12\u0026ndash;1.14)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.03 (1.02\u0026ndash;1.03)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.11 (1.10\u0026ndash;1.12)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.61 (1.59\u0026ndash;1.64)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMutually adjusted models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.20 (1.17\u0026ndash;1.23)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.26 (1.24\u0026ndash;1.29)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.11 (1.08\u0026ndash;1.13)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.22 (1.19\u0026ndash;1.25)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e2.49 (2.40\u0026ndash;2.59)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eWHR\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.09 (1.08\u0026ndash;1.10)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.07 (1.06\u0026ndash;1.08)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00 (0.99\u0026ndash;1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.06 (1.05\u0026ndash;1.07)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.35 (1.33\u0026ndash;1.37)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eInteraction models\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNormal WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh WHR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.21 (1.18\u0026ndash;1.25)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.26 (1.24\u0026ndash;1.29)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e1.11 (1.08\u0026ndash;1.13)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.22 (1.19\u0026ndash;1.25)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e2.50 (2.40\u0026ndash;2.59)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePGS\u003csub\u003eWHR\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.02 (1.00-1.05)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.05 (1.04\u0026ndash;1.07)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.00 (0.98\u0026ndash;1.01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.03 (1.02\u0026ndash;1.05)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.31 (1.27\u0026ndash;1.35)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh WHR * PGS\u003csub\u003eWHR\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e1.10 (1.07\u0026ndash;1.12)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.03 (1.01\u0026ndash;1.05)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.01 (0.99\u0026ndash;1.03)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.05 (1.02\u0026ndash;1.07)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003e1.04 (1.00-1.08)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eRisk of mortality and NCDs in relation to measured WHR and PGS\u003csub\u003eWHR\u003c/sub\u003e, presented as hazard ratios and 95% confidence intervals. WHR was categorized as normal (reference) or high using sex-specific thresholds (\u0026le;\u0026thinsp;0.90 for men and \u0026le;\u0026thinsp;0.85 for women). The PGS\u003csub\u003eWHR\u003c/sub\u003e was standardized to reflect associations per one standard deviation higher PGS. Separate models include WHR categories or the PGS\u003csub\u003eWHR\u003c/sub\u003e; mutually adjusted models include WHR categories and the PGS\u003csub\u003eWHR\u003c/sub\u003e in the same model; interaction models additionally include interaction terms between WHR category and the PGS\u003csub\u003eWHR\u003c/sub\u003e. All models were adjusted for age, sex, ethnicity, education, assessment center, and Townsend deprivation index. Models including the PGS were additionally adjusted for the first ten genetic principal components (PC1\u0026ndash;PC10). Statistically significant associations (at α\u0026thinsp;=\u0026thinsp;0.05) are presented in bold. Abbreviations: CRD, chronic respiratory disease; CVD, cardiovascular disease; HR, Hazard ratio; NCD, non-communicable disease; PGS, polygenic score; WHR, waist-to-hip ratio.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eDifferences in associations between adiposity and disease subtypes\u003c/h2\u003e \u003cp\u003eAnalyses of cancer subtypes, though limited by lower power, showed that overweight and obesity were associated with higher risk of breast, colorectal, and stomach cancer, but lower risk of lung, prostate, and skin cancer (Table S6). High WHR was associated with higher risk of breast, lung, colorectal, and stomach cancer, while no clear associations were observed for prostate or skin cancer (Table S6).\u003c/p\u003e \u003cp\u003eAcross other NCD subtypes, associations were broadly consistent with the main analyses but varied in magnitude (Table S7). Obesity, as well as the obesity\u0026ndash;PGS\u003csub\u003eBMI\u003c/sub\u003e interaction, was stronger for ischemic heart disease than for cerebrovascular disease, and weaker for chronic obstructive pulmonary disease than for CRD. Results for T2D were comparable to those observed for all diabetes in the main analyses. Associations between high WHR and the subtypes were similar to the main analyses. Interactions between high WHR and PGS\u003csub\u003eWHR\u003c/sub\u003e were overall attenuated and statistically significant only for COPD.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePredicted probability and 95% confidence intervals of the outcomes at age 70, for individuals with obesity by level of genetic predisposition to higher BMI (PGS\u003csub\u003eBMI\u003c/sub\u003e) in the left panels, and individuals with high WHR by level of genetic predisposition to higher WHR (PGS\u003csub\u003eWHR\u003c/sub\u003e) in the right panels. Predictions are based on Cox proportional hazard models adjusted for age, sex, ethnicity, education, assessment center, Townsend deprivation index, and genetic principal components. Predictions represent those for a White male of European genetic ancestry, assessed in England and with mean levels of education, deprivation index, and genetic principal components. Abbreviations: BMI body mass index; WHR waist\u0026ndash;hip ratio; PGS polygenic score.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDifferences in associations between adiposity and the outcomes across sex, age, and genetic ancestry\u003c/h3\u003e\n\u003cp\u003eAssociations between obesity and the outcomes were generally comparable in sex-stratified models (Tables S8\u0026ndash;S9), but slightly stronger in females than in males. Associations between high WHR or the PGS\u003csub\u003eWHR\u003c/sub\u003e and the outcomes were similar in females and males, but the interaction between WHR and the PGS\u003csub\u003eWHR\u003c/sub\u003e (interaction term with HR\u0026thinsp;\u0026gt;\u0026thinsp;1) was only present among males. Among females, all interaction terms were close to 1, except for diabetes where a higher PGS\u003csub\u003eWHR\u003c/sub\u003e instead reduced the association between high WHR and disease risk (interaction HR: 0.92, 95% CI 0.88\u0026ndash;0.96).\u003c/p\u003e \u003cp\u003eIn age-stratified analyses, associations between midlife measures of adiposity (ages 49\u0026ndash;59 at baseline) and the outcomes were largely consistent with the main results (Table S10). For adiposity measures taken in late life (ages 60\u0026ndash;69), associations remained similar but were overall weaker (Table S11). Across genetic ancestry groups, (Tables S12\u0026ndash;S13), associations between adiposity measures and disease outcomes were consistent in both European and non-European ancestry groups, although estimates in the non-European group were less precise due to smaller sample size.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe investigated how phenotypic and genetically predicted adiposity interact in relation to risk of mortality and the four most common causes of death by NCDs: CVD, cancers, CRD, and diabetes. This study extends prior work on genetically versus environmentally influenced obesity by, for the first time, additionally examining WHR and genetic predisposition to WHR in relation to multiple NCDs. Genetic predisposition to higher BMI attenuated the associations of obesity with CVD, CRD, and diabetes, but strengthened the association between obesity and cancer, a phenotype not previously examined in this context. In contrast, the association of a high WHR with risk of mortality, CVD, CRD, and diabetes was stronger in those with a higher genetically predicted WHR. These findings indicate a complex interaction between adiposity and genetic predisposition. For most outcomes, obesity is less detrimental in individuals with genetic predisposition to a higher BMI, while genetic predisposition to higher WHR instead amplifies associations between adiposity and risk of mortality and NCDs.\u003c/p\u003e \u003cp\u003eThat genetic predisposition to higher BMI attenuated the associations of obesity with several of the outcomes is consistent with the concept of genotype\u0026ndash;phenotype discordance, where exceeding one\u0026rsquo;s genetically predicted BMI confers greater health risks. Prior studies have reported similar patterns: data from the Swedish Twin Registry showed stronger associations of midlife obesity with dementia and CVD in those genetically predisposed to a lower BMI(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e), and findings from the Health and Retirement Study indicated similar discordance effects for mortality, cardiovascular outcomes, and cognitive decline(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). More recently, findings from the UK Biobank data demonstrated that individuals whose measured BMI exceeded their genetically predicted BMI had increased risk of diabetes(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Together with our results, these studies suggest that obesity driven, at least partly, by genetic factors, may be less detrimental than obesity influenced by other factors such as environment and lifestyle.\u003c/p\u003e \u003cp\u003eIn contrast, genetic predisposition to a higher WHR strengthened the association between high WHR and the outcomes. This may reflect that BMI and WHR not only capture different aspects of adiposity but are driven by distinct genetic influences. Previous work show that genetic variants associated with BMI act mainly through the central nervous system through regulation of appetite and energy metabolism, while those associated with WHR act through adipose tissue and the digestive system(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Moreover, some genetic variants associated with BMI are associated with a rather favorable metabolic profile(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). Winkler and colleagues could separate 159 genetic variants associated with BMI, WHR, or WHR adjusted for BMI into four groups: Genetic variants associated with higher BMI and higher WHR (n\u0026thinsp;=\u0026thinsp;82), only with higher BMI (n\u0026thinsp;=\u0026thinsp;25), with higher BMI but lower WHR (n\u0026thinsp;=\u0026thinsp;24), or only with lower WHR (n\u0026thinsp;=\u0026thinsp;28)(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Thus, while many genetic variants simultaneously increase both BMI and WHR, the PGS\u003csub\u003eBMI\u003c/sub\u003e and PGS\u003csub\u003eWHR\u003c/sub\u003e will also capture unique genetic influences as well as variants with opposing effects on the two adiposity measures. Highlighting that the PGS\u003csub\u003eBMI\u003c/sub\u003e and PGS\u003csub\u003eWHR\u003c/sub\u003e are distinct, sensitivity analyses of cross-trait interactions between each adiposity measures and a PGS for the other measure were largely consistent with the effect of the PGS in the main analyses: a high BMI, at least in the overweight range, was more detrimental in individuals with genetic predisposition to a higher WHR, and a high WHR less detrimental in individuals with genetic predisposition to higher BMI.\u003c/p\u003e \u003cp\u003eA theoretical model of body weight postulates an inherent set point around which body weight is biologically controlled from the brain, through feedback mechanisms from the periphery(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). However, environmental influences can override biological control, resulting in body weight outside the set point range with greater metabolic dysregulation and increased disease risk(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). It is plausible that genetically influenced obesity is less detrimental because it is driven, at least partly, by a high body weight set point, while environmentally influenced obesity is a result of overridden biological control of body weight, resulting in more adverse health effects. As mentioned, genetic influences on WHR are less related to the central nervous system, and instead act predominantly through adipose tissue and the digestive system(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Central adiposity indicates visceral adipose tissue with fat deposit stored in and around organs, which is linked to the adverse metabolic consequences of obesity(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). Thus, while genetic predisposition to a higher BMI may be related to a higher inherent set point, acting through the central nervous system, genetic predisposition to higher WHR may act more directly on adipose tissue and body fat distribution, leading to metabolic dysfunction and exacerbation of the adverse effects of a high WHR. This may explain why genetic predisposition to higher BMI and WHR have opposing effects on the associations between adiposity and adverse health.\u003c/p\u003e \u003cp\u003ePronounced sex differences were observed in the interactions between WHR and genetic predisposition in relation to the outcomes. While genetic predisposition to higher WHR amplified the association between WHR and all outcomes in males, the interaction between high WHR and PGS\u003csub\u003eWHR\u003c/sub\u003e was significant only for diabetes in females, where it weakened the association between high WHR and diabetes. These findings align with previous evidence showing extensive sexual dimorphism in the genetic architecture of WHR, with most WHR-associated loci exhibiting stronger effects in women but distinct metabolic consequences across fat depots(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). Recent work modeling the genomic architecture of adiposity across the lifespan also reported sex-specific loadings for latent factors representing abdominal size (where WHR loads) and general adiposity (where BMI loads)(\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e), further supporting sex differences in adiposity-related genetic influences. The interplay between sex hormones and adipose tissue regulation may contribute to these patterns, underscoring the importance of accounting for sex-specific genetic and physiological factors when assessing the metabolic effects of body fat distribution. It is also possible that menopausal status contributed to these sex-specific patterns, as fat distribution typically shifts toward central adiposity after menopause(\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn contrast to the lower risk of CVD, CRD, and diabetes in genetically influenced obesity, our findings indicate a higher cancer risk among individuals with obesity if they also have genetic predisposition to higher BMI. The divergent patterns between cancer and cardiometabolic outcomes may reflect disease-specific mechanisms: metabolic and hemodynamic pathways explain the robust positive relationship with cardiometabolic outcomes, whereas hormonal regulation, reverse causality (e.g., pre-diagnostic weight loss), and lifestyle interactions such as smoking complicate associations with certain cancers(\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). Prior research has shown that adiposity does not affect all cancer types uniformly, and the International Agency for Research on Cancer, part of WHO, defines 13 obesity-related cancers, which include postmenopausal breast cancer, colorectal cancer, and stomach cancer, but not lung, prostate, or skin cancer(\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). In line with these differences, our sensitivity analyses of the six most common cancers demonstrated associations between obesity and higher risk of the obesity-related cancers, but lower risk of lung, prostate, and skin cancer. It should be noted that high WHR was associated with higher risk of all cancer types except for skin cancer, where no association was seen. In line with this, several meta-analyses and Mendelian randomization studies report inverse associations between BMI and lung cancer(\u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e), whereas measures of abdominal adiposity, such as WHR, show null or positive associations(\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). While results should be interpreted with caution due to differences in statistical power, there were no clear differences in the effect of interaction between the adiposity measures and PGS between subtypes.\u003c/p\u003e \u003cp\u003eUnderweight was also associated with higher risk of mortality, CVD, and CRD in the current study. In late life or among individuals with pre-existing chronic or degenerative conditions, a higher BMI is often linked to a reduced risk of adverse health outcomes, possibly due to reverse causation rather than a protective effect of higher BMI(\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Previous work indicates that associations between low BMI and adverse health also differ by genetic predisposition. We demonstrated that in late life, a higher BMI was associated with a lower risk of dementia, but only among those genetically predicted to a high BMI(\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Similarly, in a recent study of individuals with chronic obstructive pulmonary disease (COPD), those with a measured BMI lower than their genetically predicted BMI had the highest risk of mortality(\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e). In contrast, there was no evidence of an interaction between underweight and the PGS\u003csub\u003eBMI\u003c/sub\u003e in the current study. It is plausible that this lack of a difference is due to the relatively younger sample, where reverse causation due to unintentional weight loss is less evident than in older samples. However, associations between underweight and the outcomes were comparable in sensitivity analyses stratifying the sample into measures taken in midlife versus late-life, with no indications of interactions in either age group. Associations between higher BMI and disease outcomes as well as interactions with the PGS\u003csub\u003eBMI\u003c/sub\u003e were generally weaker in the late-life sample however, possibly reflecting the influence of pre-existing illness, loss of muscle mass, or selective survival among older individuals with obesity. In contrast, WHR showed more stable associations across age groups, which is expected given that fat distribution tends to shift with age, with relatively greater central adiposity(\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). These findings underscore the limitations of BMI as a measure of adiposity in older populations, where it may not fully capture body composition or metabolic risk.\u003c/p\u003e \u003cp\u003eThe main strengths of this study lie in its large, population-based design with detailed phenotypic and genotypic characterization, permitting robust evaluation of interactions between measured and genetically predicted adiposity across mortality and four prioritized NCDs defined by the WHO. The inclusion of both BMI and WHR, together with their respective PGS, allowed a comprehensive comparison between overall and central adiposity, providing novel insights into their distinct genetic and metabolic underpinnings. The availability of harmonized covariate data further strengthened the analyses, enabling multiple sensitivity and stratified models, including testing lifestyle influences. However, the observational design limits causal interpretation, and BMI and WHR were measured only at baseline and do not capture weight or fat distribution changes over time. While follow-up of disease diagnoses through healthcare registers is a strength, it should be noted that inpatient data may best capture more severe cases, while missing milder forms of disease diagnosed in primary care(\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e). Additionally, information about menopause and other relevant information was not available during follow-up, limiting investigations of the causes of observed sex-differences. In addition, despite the large sample size, statistical power was more limited for sensitivity analyses, particularly for certain subtypes, such as cancer outcomes, and comparisons should be made with caution. The PGS, while informative, explain only a fraction of the heritable component of adiposity, and residual confounding from unmeasured lifestyle or socioeconomic factors cannot be excluded. Nevertheless, results indicated clear differences between the PGS\u003csub\u003eBMI\u003c/sub\u003e and PGS\u003csub\u003eWHR\u003c/sub\u003e, which were robust in cross-trait analyses. It should also be noted that the PGS are most powerful to predict respective trait in samples of European ancestry. While ancestry-stratified analyses were overall robust, the UK Biobank primarily includes individuals of European genetic ancestry, and the generalizability of the findings to more diverse populations remain to be established.\u003c/p\u003e \u003cp\u003eIn summary, while it is important to note that obesity was associated with increased risk of all outcomes regardless of genetic predisposition, the association with CVD, CRD, and diabetes weakened with genetic predisposition to higher BMI. This indicates that obesity influenced by genetic predisposition may be less detrimental than obesity influenced by environmental factors. For cancer, the direction of the interaction differed from the other outcomes and indicated a higher risk in those with obesity and genetic predisposition to a higher BMI. When adiposity was measured as high WHR, the associations with mortality, CVD, CRD, and diabetes were instead stronger with genetic predisposition to higher WHR. These findings emphasize that BMI and measure of body fat distribution represent distinct phenotypes, both on the phenotypic and genetic level. Future research should further disentangle the shared and unique genetic influences on BMI and WHR to clarify how genetic susceptibility interacts with environmental exposures in shaping the heterogeneity of adiposity and its consequences for health.\u003c/p\u003e"},{"header":"Declarations","content":" \u003ch2\u003eEthics approval and consent to participate:\u003c/h2\u003e \u003cp\u003e All individuals in the sample gave written informed consent to participate in UKB and for data to be used in future research. The UKB study has ethical approval from the Northwest Multi-Centre Research Ethics Committee. The current study was approved by the Swedish Ethical Review Authority, DNR 2024-03706-01.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication:\u003c/strong\u003e \u003cp\u003eNot applicable.\u003c/p\u003e \u003ch2\u003eCompeting interests:\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding:\u003c/h2\u003e \u003cp\u003eThe study was funded by the Swedish Research Council for Health, Working Life and Welfare (Forte; 2022\u0026thinsp;\u0026minus;\u0026thinsp;00672); the Strategic Research Program in Epidemiology (SFOepi) at Karolinska Institutet; Karolinska Institutet's Research Foundation (2022\u0026thinsp;\u0026minus;\u0026thinsp;01718, 2024\u0026ndash;02898); Loo and Hans Osterman's Foundation (2022\u0026thinsp;\u0026minus;\u0026thinsp;01222, 2023\u0026thinsp;\u0026minus;\u0026thinsp;01855, 2024\u0026ndash;02197); and the Foundation for Geriatric Diseases at Karolinska Institutet (2022\u0026thinsp;\u0026minus;\u0026thinsp;01296, 2023\u0026thinsp;\u0026minus;\u0026thinsp;01854, 2024\u0026ndash;02116).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eEOL, CAR, YZ, ADA, JJ, and IK contributed to conception or design of the work. SH contributed to data acquisition. EOL, SM, and IK contributed to the statistical analyses. All authors contributed to interpretation of the results. EOL, MN, and IK drafted the first version of the manuscript, and all authors contributed to critical revision and approved the final version for submission.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe are grateful to UK Biobank participants and research team, without whom the research would not have been possible.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThis research was conducted using the UK Biobank resource under the Application Number 22224. UK Biobank data are available upon application to the UK Biobank (https:/www.ukbiobank.ac.uk ). The summary statistics used for the polygenic score construction are publicly available from the GWAS publications(19, 20).The analytical plan for the current study was preregistered on the Open Science Framework (OSF; https:/osf.io/k3qxy ) in January 2025. The analysis plan includes descriptions of the study aims, hypotheses, exposure and outcome definitions, and planned statistical analyses. Final codes used for data processing and analysis are also available at the Open Science Framework, along with the preregistration.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHildebrand S, Pfeifer A. The obesity pandemic and its impact on non-communicable disease burden. Pflugers Arch. 2025;477(5):657\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilventoinen K, Jelenkovic A, Sund R, Yokoyama Y, Hur Y-M, Cozen W, et al. Differences in genetic and environmental variation in adult BMI by sex, age, time period, and region: an individual-based pooled analysis of 40 twin cohorts. Am J Clin Nutr. 2017;106(2):457\u0026ndash;66.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMin J, Chiu DT, Wang Y. Variation in the heritability of body mass index based on diverse twin studies: a systematic review. Obes Rev. 2013;14(11):871\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOjalehto E, Zhan Y, Jylh\u0026auml;v\u0026auml; J, Reynolds CA, Dahl Aslan AK, Karlsson IK. Genetically and environmentally predicted obesity in relation to cardiovascular disease: a nationwide cohort study. EClinicalMedicine. 2023;58:101943.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavidson T, Vinneau-Palarino J, Goode JA, Boardman JD. Utilizing genome wide data to highlight the social behavioral pathways to health: The case of obesity and cardiovascular health among older adults. Soc Sci Med. 2021;273:113766.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRhee T-M, Choi J, Lee H, Merino J, Park J-B, Kwak SH. Discrepancy Between Genetically Predicted and Observed BMI Predicts Incident Type 2 Diabetes. Diabetes Care. 2024;47(10):1826\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson IK, Lehto K, Gatz M, Reynolds CA, Dahl Aslan AK. Age-dependent effects of body mass index across the adult life span on the risk of dementia: a cohort study with a genetic approach. BMC Med. 2020;18(1):131.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson IK, Gatz M, Arpawong TE, Dahl Aslan AK, Reynolds CA. The dynamic association between body mass index and cognition from midlife through late-life, and the effect of sex and genetic influences. Sci Rep. 2021;11(1):7206.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVinneau JM, Huibregtse BM, Laidley TM, Goode JA, Boardman JD. Mortality and Obesity Among U.S. Older Adults: The Role of Polygenic Risk. J Gerontol B Psychol Sci Soc Sci. 2021;76(2):343\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, et al. Meta-analysis of genome-wide association studies for height and body mass index in \u0026sim;700000 individuals of European ancestry. Hum Mol Genet. 2018;27(20):3641\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePulit SL, Stoneman C, Morris AP, Wood AR, Glastonbury CA, Tyrrell J, et al. Meta-analysis of genome-wide association studies for body fat distribution in 694 649 individuals of European ancestry. Hum Mol Genet. 2019;28(1):166\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWinkler TW, G\u0026uuml;nther F, H\u0026ouml;llerer S, Zimmermann M, Loos RJ, Kutalik Z, et al. A joint view on genetic variants for adiposity differentiates subtypes with distinct metabolic implications. Nat Commun. 2018;9(1):1946.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. Global status report on noncommunicable diseases 2014. 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarslake D, Davey Smith G, Gunnell D, Davies N, Nilsen TIL, Romundstad P. Confounding by ill health in the observed association between BMI and mortality: evidence from the HUNT Study using offspring BMI as an instrument. Int J Epidemiol. 2018;47(3):760\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDye L, Boyle NB, Champ C, Lawton C. The relationship between obesity and cognitive health and decline. Proc Nutr Soc. 2017;76(4):443\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12(3):e1001779.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKivim\u0026auml;ki M, Batty GD, Singh-Manoux A, Britton A, Brunner EJ, Shipley MJ. Validity of Cardiovascular Disease Event Ascertainment Using Linkage to UK Hospital Records. Epidemiology. 2017;28(5):735\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLocke AE, Kahali B, Berndt SI, Justice AE, Pers TH, Day FR, et al. Genetic studies of body mass index yield new insights for obesity biology. Nature. 2015;518(7538):197\u0026ndash;206.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShungin D, Winkler TW, Croteau-Chonka DC, Ferreira T, Locke AE, M\u0026auml;gi R, et al. New genetic loci link adipose and insulin biology to body fat distribution. Nature. 2015;518(7538):187\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLloyd-Jones LR, Zeng J, Sidorenko J, Yengo L, Moser G, Kemper KE, et al. Improved polygenic prediction by Bayesian multiple regression on summary statistics. Nat Commun. 2019;10(1):5086.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhuang P, Liu X, Li Y, Wan X, Wu Y, Wu F, et al. Effect of Diet Quality and Genetic Predisposition on Hemoglobin A1c and Type 2 Diabetes Risk: Gene-Diet Interaction Analysis of 357,419 Individuals. Diabetes Care. 2021;44(11):2470\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team. R: A Language and Environment for Statistical Computing [Internet]. R Foundation for Statistical Computing. 2025. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.r-project.org\u003c/span\u003e\u003cspan address=\"https://www.r-project.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePosit team. RStudio: Integrated Development Environment for R [Internet]. Boston, MA: Posit Software, PBC. 2025. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://posit.co\u003c/span\u003e\u003cspan address=\"https://posit.co\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H, Miller E, Smith D. haven: Import and Export 'SPSS', 'Stata' and 'SAS' Files. R package version 2.5.5 ed2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H, Fran\u0026ccedil;ois R, Henry L, M\u0026uuml;ller K, Vaughan D. dplyr: A Grammar of Data Manipulation. R package version 1.1.4 ed2025.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTherneau TM, until TL, Elizabeth A, survival Cynthia C. Survival Analysis. 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H. ggplot2: Elegant Graphics for Data Analysis. 2016.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuguie B, Antonov A, gridExtra. Miscellaneous Functions for Grid Graphics. 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoscardin CK, Sewell JL, Tolsgaard MG, Pusic MV. How to Use and Report on p-values. Perspect Med Educ. 2024;13(1):250\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim D, Highland HM, Smit RAJ, Hysong MR, Buchanan VL, Young KL, et al. Genetic underpinnings of the heterogeneous impact of obesity on lipid levels and cardiovascular disease. Genome Med. 2025;17(1):113.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChami N, Wang Z, Svenstrup V, Obrero VD, Hemerich D, Huang Y, et al. Genetic subtyping of obesity reveals biological insights into the uncoupling of adiposity from its cardiometabolic comorbidities. Nat Med. 2025;31(11):3801\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026uuml;ller MJ, Geisler C, Heymsfield SB, Bosy-Westphal A. Recent advances in understanding body weight homeostasis in humans. F1000Res. 2018;7:F1000-Faculty Rev-25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSulc J, Winkler TW, Heid IM, Kutalik Z. Heterogeneity in Obesity: Genetic Basis and Metabolic Consequences. Curr Diab Rep. 2020;20(1):1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArehart CH, Lin M, Gibson RA, Colorado Center for Personalized M, Raghavan S, Gignoux CR, et al. Modeling the genomic architecture of adiposity and anthropometrics across the lifespan. Nat Commun. 2025;16(1):7494.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eToth MJ, Tchernof A, Sites CK, Poehlman ET. Menopause-related changes in body fat distribution. Ann N Y Acad Sci. 2000;904:502\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRenehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 2008;371(9612):569\u0026ndash;78.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBai T, Wu C. Association of cardiovascular disease on cancer: observational and mendelian randomization analyses. Sci Rep. 2024;14(1):28465.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLarsson SC, Spyrou N, Mantzoros CS. Body fatness associations with cancer: evidence from recent epidemiological studies and future directions. Metabolism. 2022;137:155326.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLauby-Secretan B, Scoccianti C, Loomis D, Grosse Y, Bianchini F, Straif K, et al. Body Fatness and Cancer\u0026ndash;Viewpoint of the IARC Working Group. N Engl J Med. 2016;375(8):794\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Y, Dong J, Sun K, Zhao L, Zhao F, Wang L, et al. Obesity and incidence of lung cancer: a meta-analysis. Int J Cancer. 2013;132(5):1162\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHidayat K, Du X, Chen G, Shi M, Shi B. Abdominal Obesity and Lung Cancer Risk: Systematic Review and Meta-Analysis of Prospective Studies. Nutrients. 2016;8(12):810.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou W, Liu G, Hung RJ, Haycock PC, Aldrich MC, Andrew AS, et al. Causal relationships between body mass index, smoking and lung cancer: Univariable and multivariable Mendelian randomization. Int J Cancer. 2021;148(5):1077\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Moll M, Hobbs BD, Bakke P, Regan EA, Xu H, et al. Genetically Predicted Body Mass Index and Mortality in Chronic Obstructive Pulmonary Disease. Am J Respir Crit Care Med. 2024;210(7):890\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuk JL, Saunders TJ, Davidson LE, Ross R. Age-related changes in total and regional fat distribution. Ageing Res Rev. 2009;8(4):339\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilkinson T, Schnier C, Bush K, Rannikm\u0026auml;e K, Henshall DE, Lerpiniere C, et al. Identifying dementia outcomes in UK Biobank: a validation study of primary care, hospital admissions and mortality data. Eur J Epidemiol. 2019;34(6):557\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Obesity, Adiposity, Polygenic score, Genetic predisposition, Non-communicable disease, Mortality, Epidemiology","lastPublishedDoi":"10.21203/rs.3.rs-8296230/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8296230/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePrevious research indicate that obesity is less harmful in people with genetic predisposition to high body mass index (BMI), compared to obesity driven by other factors such as environment or lifestyle. Yet differences between genetically and environmentally influenced adiposity in relation to health outcomes remain unexplored, and have not examined adiposity measures beyond BMI. Therefore, we examined differences between genetically versus environmentally influenced adiposity, measured with BMI as well as waist-hip ratio (WHR), in relation to risk of mortality and key non-communicable diseases.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003e We followed 484,858 UK Biobank participants (aged 40\u0026ndash;69 years at baseline) over on average 13 years. Baseline BMI and WHR categories were tested in interaction with polygenic scores (PGS) for respective trait, to distinguish between high adiposity influenced by genetic predisposition (obesity or high WHR and high PGS) versus by environmental factors (obesity or high WHR but low PGS). Risk of all-cause mortality, cardiovascular disease (CVD), cancers, chronic respiratory disease (CRD), and diabetes were modelled in Cox proportional hazard regression, adjusting for age, sex, ethnicity, and socioeconomic factors..\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe PGS for BMI interacted with obesity, such that genetic predisposition to higher BMI attenuated obesity associations with CVD, diabetes, and CRD, but strengthened that with cancer. In contrast, the PGS for WHR had the opposite effect, and genetic predisposition to higher WHR instead amplified associations between high WHR and mortality, CVD, diabetes, and CRD.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eObesity was linked to lower risk for most outcomes in people with genetic predisposition to higher BMI, compared to those with genetic predisposition to a low BMI. This indicates that genetically influenced obesity may be less detrimental than obesity influenced by environmental factors sucha as lifestyle. In contrast, the opposite was seen when adiposity was measured as WHR, where the association between high WHR and most outcomes was stronger in people with genetic predisposition to higher WHR. This highlights that BMI and WHR capture distinct adiposity profiles with opposing genetic effects, and underscores the heterogeneity in obesity.\u003c/p\u003e","manuscriptTitle":"Genetically versus environmentally influenced obesity and risk of mortality and non-communicable diseases: A cohort study from the UK Biobank","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-07 10:29:56","doi":"10.21203/rs.3.rs-8296230/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4eb0aa82-ef4d-4bb6-aa43-ae780538db2a","owner":[],"postedDate":"January 7th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-30T15:27:11+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-07 10:29:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8296230","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8296230","identity":"rs-8296230","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}