{"paper_id":"1cbc8248-b3fe-48af-9345-6acd33ae1c0c","body_text":"Estimation of Years of Life Lost Due to Premature Mortality from Diabetes \nMellitus in Bahia - Brazil \n \n1 Luís Jesuino de Oliveira Andrade - https://orcid.org/0000-0002-7714-0330 \n2 Gabriela Correia Matos de Oliveira - https://orcid.org/0000-0002-3447-3143 \n3 Alcina Maria Vinhaes Bittencourt - https://orcid.org/0000-0003-0506-9210 \n3 João Cláudio Nunes Carneiro Andrade -  https://orcid.org/0009-0000-6004-4054 \n1 Janine Lemos de Lima - https://orcid.org/0009-0005-3271-6980 \n1 Luís Matos de Oliveira - https://orcid.org/0000-0003-4854-6910 \n \n1 Departamento de Saúde Universidade Estadual de Santa Cruz, Ilhéus, Bahia, Brazil. \n2 Programa Saúde da Família, Bahia, Brazil. \n3 Faculdade de Medicina Universidade Federal da Bahia, Salvador, Bahia, Brazil.  \n \nCorresponding Author: \nLuís Jesuino de Oliveira Andrade \nUniversidade Estadual de Santa Cruz - Campus Soane Nazaré de Andrade, Rod. Jorge \nAmado, Km 16 - Salobrinho, Ilhéus - BA, 45662-900.  \nE-mail: luis_jesuino@yahoo.com.br \n \nABSTRACT \nIntroduction: Diabetes mellitus (DM) in Bahia, Brazil, is a critical public health issue, \nintensified by socioeconomic inequalities. Years of life lost (YLL) highlight healthcare \naccess gaps, impacting vulnerable populations. Enhanced healthcare infrastructure, \neducation, and policy changes are essential to reduce DM morbidity and premature \nmortality. Objective: To quantify the societal burden of DM in Bahia, Brazil, by \nestimating YLL due to premature mortality between 2000 and 2023. Method: This \nstudy uses a quantitative epidemiological approach to estimate YLL due to DM in \nBahia, Brazil (2000–2023). Mortality data from the Mortality Information System \n(SIM) and demographic projections from IBGE were analyzed. YLL was calculated by \nmultiplying deaths by remaining life expectancy (using GBD 2019 reference tables), \nstratified by age and sex. Statistical analysis employed PSPP, Excel, and Python, with \nresults presented as rates per 100,000 population. Socioeconomic and healthcare factors \nwere contextualized using government reports and academic literature. Results: The \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\nstudy analyzed YLL due to DM in Bahia (2000–2023), revealing an upward trend, \npeaking in 2020 (422.6/100,000), likely exacerbated by COVID-19. Males showed \nhigher premature mortality than females (34.9 vs. 29.8/100,000 in 2020). Economic \nlosses reached ~41 million BRL (2020–2022). Rising YLL reflects growing DM \nburden, with socioeconomic impacts including lost productivity and household \ninstability. Gender disparities suggest differences in healthcare access, biological \nsusceptibility, or lifestyle factors, necessitating targeted interventions. Conclusion: \nBahia faces rising diabetes-related premature deaths, worsening post-2015 and spiking \nduring COVID-19. Men show higher mortality, with severe socioeconomic impacts, \nrequiring urgent public health and economic strategies. \nKeywords: Diabetes Mellitus, Years of Life Lost, Socioeconomic Disparities \n \nINTRODUCTION \nDiabetes mellitus (DM) is an important public health challenge globally, with its \nburden disproportionately affecting low- and middle-income regions such as Bahia, \nBrazil.\n1 As a metabolic disorder characterized by chronic hyperglycemia, DM \ncontributes significantly to premature mortality through complications like \ncardiovascular disease, renal failure, and neuropathy. Between 2000 and 2024, Bahia a \nstate marked by socioeconomic disparities and uneven healthcare access has \nexperienced rising DM prevalence, mirroring national trends but compounded by \nregional inequities.\n2 Quantifying the societal impact of DM-related deaths requires \nrobust epidemiological metrics, such as Years of Life Lost (YLL), which estimate the \ngap between observed age at death and life expectancy. \nYLL serves as an important tool for contextualizing premature mortality, \nemphasizing the societal cost of diseases like DM beyond crude mortality rates.\n3 Recent \nstudies in Brazil highlight YLL’s utility in uncovering hidden burdens in populations \nwith fragmented health data, particularly in the Northeast region, where Bahia is \nsituated .\n4 For instance, was demonstrated that YLL calculations in similar settings \nreveal up to 40% higher disease impacts than traditional mortality analyses, \nunderscoring the need for granular, region-specific assessments. 5 This approach is \nespecially pertinent in Bahia, where systemic gaps in chronic disease management \nexacerbate preventable DM complications. \n Socioeconomic determinants, including income inequality and limited primary \ncare coverage, disproportionately affect DM outcomes in Bahia. A cohort study linked \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nlow socioeconomic status to a 2.3-fold increase in DM-related mortality among Bahian \nadults under 60, highlighting premature death as a marker of healthcare inequity. 6 \nFurthermore, urbanization and shifts toward sedentary lifestyles have amplified DM \nincidence, yet public health responses remain fragmented. Chaves-Fonseca RM, et al.\n7 \nnoted that less than 60% of Bahia’s municipalities met national targets for diabetes care \ninfrastructure, perpetuating regional disparities in mortality.\n \n Temporal trends in DM mortality further complicate Bahia’s landscape. While \nnational data show a 22% decline in DM-related YLL from 2010 to 2019, Bahia’s \nreduction lagged at 12%, attributed to slower implementation of preventive strategies.\n8 \nEmerging evidence also suggests that the COVID-19 pandemic disrupted routine DM \nmanagement, potentially exacerbating premature mortality post-2020.9 \nThis study’s findings aim to inform targeted interventions by delineating the \nmagnitude and drivers of DM-related YLL in Bahia. By integrating mortality data with \ndemographic projections, the analysis will address gaps identified in the Global Burden \nof Disease framework, which often overlooks subnational heterogeneity.\n10 For Bahia \nhome to 14.850.513 people, such granularity is critical for aligning resource allocation \nwith the UN’s Sustainable Development Goals, particularly Target 3.4, which \nprioritizes reducing premature mortality from noncommunicable diseases.\n11 \n This study seeks to quantify the societal burden of DM in Bahia, Brazil, by \nestimating YLL due to premature mortality between 2000 and 2023. By integrating \nmortality data with demographic and epidemiologic projections, the analysis aims to \nelucidate the magnitude of preventable DM-related deaths, contextualizing disparities \ndriven by socioeconomic inequities, fragmented healthcare access, and evolving public \nhealth challenges. \n \nMETHODOLOGY \n This study will employ a quantitative, epidemiological approach to estimate \nYLL attributable to DM in Bahia, Brazil, between 2000 and 2023.  The methodology \nwill involve a multi-stage process integrating mortality data, demographic projections, \nand relevant epidemiological parameters. \nData Sources   \nMortality data for DM spanning the period from 2000 to 2023 were obtained \nthrough the Mortality Information System (SIM), developed by the Brazilian Ministry \nof Health in 1975. This system resulted from the unification of over forty different \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\ninstrument models used throughout the years to collect mortality data in the country \n(http://tabnet.datasus.gov.br/cgi/deftohtm.exe?sim/cnv/obt10uf.def). These data were \nstratified by age, sex, and underlying cause of death, specifically focusing on deaths \nwhere DM is listed as the primary or contributing cause. Demographic data, including \npopulation counts and age-sex distributions for Bahia, were obtained from the Brazilian \nInstitute of Geography and Statistics (IBGE).  These data were used to project the \npopulation at risk of DM-related mortality throughout the study period.  \nEpidemiological parameters, such as DM prevalence rates, were drawn from national \nhealth surveys and published literature specific to the Bahian population. \nYLL Calculation   \nYLL was calculated using the standard methodology, which involves subtracting \nthe age at death from a standard life expectancy: \nBasic Formula: YLL = \nΣ  (Number of deaths in each age group) × (Life expectancy in \nthat age group) \nThe choice of standard life expectancy was carefully considered, potentially \nutilizing the World Health Organization (WHO) standard or a Brazilian national life \nexpectancy table, ensuring consistency and comparability with other studies.  YLL was \ncalculated for each death attributed to DM and then aggregated to provide an overall \nestimate of DM-related YLL for Bahia.  The standard life expectancy was projected \nbased on the mean age at death, calculated with a 95% confidence interval. For this \npurpose, the global life table from the GBD (Global Burden of Disease Study) 2019 \nstudy was used as a reference to ensure comparability with existing studies. \nStatistical Analysis \nYLL due to premature mortality, as defined by the GBD study methodology, \naddresses the limitations of arbitrary age thresholds by quantifying lost time rather than \nraw death counts, using an individual’s maximum potential lifespan at each age as the \nfoundational metric.\n12 YLL is calculated by multiplying the number of deaths by the \nremaining life expectancy at the age of death. For DM, sex, and age group stratification, \nYLL estimates were derived as follows: \nThe core formula for YLL for a given cause c, sex s, age a, and year t is: \nYLL(c,s,a,t)=N(c,s,a,t)×L(s,a)YLL(c,s,a,t)=N(c,s,a,t)×L(s,a) \nwhere: \n• N(c,s,a,t)N(c,s,a,t) = number of deaths due to cause c for sex s and age a in \nyear t; \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n• L(s,a)L(s,a) = standard loss function specifying years of life lost per death at \nage a for sex s. \nAs described by Murray et al., 13 a discount rate may be applied to YLL to \nprioritize present health benefits over future ones, effectively reducing the value of each \nyear of life by a fixed annual percentage. For this study, a 3% discount rate \nwas not adopted. \nRegarding age weighting and time discounting, the 1990 GBD study by the \nWHO calculated Disability-Adjusted Life Years (DALY) using a 3% discount rate for \nfuture healthy life years lost, alongside an alternative 0% rate. Proponents of \ndiscounting argued its necessity to avoid decision-making paradoxes in cost-\neffectiveness analyses.\n14 Critics, however, contended that there is no ethical justification \nfor devaluing future health outcomes, 15 a stance reinforced by expert consultations for \nthe 2010 GBD study, which discouraged discounting.16 \nData processing and variable management were performed using PSPP \nStatistics (open-source, public domain software). YLL calculations were executed \nin Excel (Office 2019 suite), the WHO deterministic DALY calculation \ntemplate (available: http://www.who.int/healthinfo/bodreferencedalycalculationtemplate\n.xls), and Python-Fiddle ( https://python-fiddle.com/?checkpoint=1742657084). Results \nare presented as rates per 100,000 population, stratified by sex and age group. \nProductivity loss in this study is analyzed through the human capital \ntheory framework, which views workers as economic investments generating returns. \nProductivity loss refers to diminished capacity of individuals or groups to engage in \neconomically productive activities due to health impairments or adverse conditions. \nThe economic cost of premature mortality due to productivity loss in Bahia was \nestimated using the following indicators: Expected annual income (RE); YLL; A 5% \ndiscount rate. \nThe formula for economic loss due to reduced productivity is: \nCost per death = (Expected annual income × Years of Life Lost) × Discount rate. \nSocioeconomic and Healthcare Contextualization  \nTo contextualize the observed YLL trends, data on socioeconomic indicators \n(e.g., income inequality, poverty rates), healthcare access (e.g., primary care coverage, \navailability of diabetes care facilities), and public health initiatives related to DM \nprevention and management in Bahia were gathered from relevant sources, such as \ngovernment reports, academic publications, and publicly available databases.  \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nQualitative contextualization was also be considered, drawing upon existing literature to \ndiscuss the complex interplay of social determinants, healthcare system factors, and \npublic health policies influencing DM-related mortality in Bahia. \nEthical Considerations   \nThis study utilized publicly available, anonymized data, and therefore, \nindividual patient consent was not be required.  So, according to Article 1, Sole \nParagraph, of Resolution No. 510/2016, the following will not be registered or \nevaluated by the Research Ethics Committee - Brazil: Research using databases in \nwhich the information is aggregated, with no possibility of individual identification. \n \nRESULTS \nDemographic data of population of Bahia \nThe demographic data for the population of Bahia were sourced from the Health \nInformation System of the Bahia State Health Department, with population estimates \ncompiled from the IBGE for the resident population in the region for the year 2020, as \noutlined in Table 1. \nTable 1. Estimated resident population, stratified by sex and age group – Bahia, 2020 to \n2022. \nAge Group Male Female Total \n0 to 4 years 525,056 500,834 1,025,890 \n5 to 9 years 527,814 504,220 1,032,034 \n10 to 14 years 567,736 543,627 1,111,363 \n15 to 19 years 611,873 591,578 1,203,451 \n20 to 24 years 636,964 626,936 1,263,900 \n25 to 29 years 590,619 612,126 1,202,745 \n30 to 34 years 576,263 621,096 1,197,359 \n35 to 39 years 597,392 648,826 1,246,218 \n40 to 44 years 543,080 588,278 1,131,358 \n45 to 49 years 452,349 493,962     946,311 \n50 to 54 years 409,589 449,741     859,330 \n55 to 59 years 344,211 384,668    728,879 \n60 to 64 years 279,734 323,490    603,224 \n65 to 69 years 211,058 254,732    465,790 \n70 to 74 years 159,226 197,803    357,029 \n75 to 79 years 106,076 143,344    249,420 \n80 years or older 115,162 191,171    306,333 \nSource: SESAB/SUVISA/DIVEP/GT - Demographics. IBGE/DataSUS/Ministry of \nHealth. March 2025. \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n \nStandard life expectancy \nAt present, the most recent iteration of the GBD study accessible for \ninvestigative purposes is the GBD 2019. This comprehensive research endeavor is \nconducted by the Institute for Health Metrics and Evaluation (IHME) at the University \nof Washington, with data undergoing periodic updates to reflect evolving health \nlandscapes. The GBD 2019 provides extensive evaluations concerning mortality, \nmorbidity, risk determinants, and life expectancy across more than 200 nations and \nterritories, encapsulating the period from 1990 to 2019. \nKey Attributes of the GBD 2019: 1) Broad Data Spectrum: It encompasses \nevaluations for over 350 distinct diseases and injuries, supplemented by 84 identified \nrisk factors; 2) Demographic Segmentation: The dataset is organized by age brackets, \ngender, and geographical location to facilitate granular analysis; 3) Health Metric \nInclusion: It integrates DALY, YLL, and Years Lived with Disability (YLD) as core \nhealth metrics; 4) Risk Factor Analysis: It scrutinizes the influence of variables such as \nobesity, tobacco consumption, and atmospheric pollution on the global disease burden. \nAccess to the data was made through the IHME website: \nhttps://www.healthdata.org/gbd/2019\n. \nThe standard life expectancy was projected based on the average age at death, \ncalculated with a 95% confidence interval. For this purpose, the Global Life Table from \nthe GBD 2019 study was used as a reference to ensure comparability with other \npreviously conducted studies (Table 2 and 3). \nTable 2. Global Life Expectancy, 2019 \nAGE GROUP                              2019 \n         Male                             Female \nUnder 1 year 70.85 75.87 \n1-4 years 72.03 76.88 \n5-9 years 68.70 73.59 \n10-14 years 63.96 68.86 \n15-19 years 59.16 64.04 \n20-24 years 54.43 59.23 \n25-29 years 49.79 54.47 \n30-34 years 45.17 49.72 \n35-39 years 40.59 45.00 \n40-44 years 36.10 40.34 \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n45-49 years 31.71 35.75 \n50-54 years 27.43 31.23 \n55-59 years 23.34 26.87 \n60-64 years 19.52 22.67 \n65-69 years 15.99 18.71 \n70-74 years 12.74 14.99 \n75-79 years 9.83 11.62 \n80-84 years 7.26 8.62 \n85+ Years 5.13 6.15 \nSource: GBD, 2019. \n \nTable 3. Age-Specific Data by Sex with 95% Confidence Intervals \nAge \nGroup \nMale \n95% Confidence Intervals \nFemale \n95% Confidence Intervals \nBoth Sexes \n95% Confidence Intervals \n Value LCI* UCI** Value LCI UCI Value LCI UCI \n< 1 71.01 70.08  71.95 76.14 75.37  76.94 73.52 72.78  74.29 \n1-4 72.16 71.40 72.98 77.16 76.48 77.90 74.62 74.04 75.26 \n5-9 68.84 68.11  69.63 73.89 73.23  74.60 71.32 70.77  71.94 \n10-14 64.05 63.33 64.84 69.07 68.42 69.76 66.52 65.96 67.12 \n15-19 59.20 58.49  59.99 64.21 63.56  64.89 61.67 61.12  62.26 \n20-24 54.47 53.76 55.24 59.41 58.77 60.07 56.90 56.37 57.49 \n25-29 49.85 49.16  50.60 54.66 54.04  55.30 52.22 51.69  52.79 \n30-34 45.23 44.56 45.97 49.92 49.31 50.54 47.54 47.03 48.10 \n35-39 40.67 40.00  41.40 45.20 44.61  45.81 42.91 42.41  43.45 \n40-44 36.19 35.55 36.90 40.54 39.97 41.13 38.35 37.86 38.88 \n45-49 31.81 31.19  32.49 35.95 35.41  36.52 33.87 33.42  34.38 \n50-54 27.55 26.96 28.18 31.44 30.91 31.97 29.49 29.07 29.97 \n55-59 23.47 22.93  24.05 27.08 26.60  27.58 25.29 24.90  25.72 \n60-64 19.66 19.18 20.18 22.88 22.44 23.34 21.30 20.96 21.69 \n65-69 16.13 15.71  16.59 18.92 18.53  19.32 17.58 17.27  17.91 \n70-74 12.88 12.53 13.26 15.21 14.88 15.55 14.12 13.86 14.41 \n75-79 10.00 9.73 10.29 11.88 11.61  12.15 11.03 10.83  11.25 \n80-84 7.50 7.31 7.71 8.96 8.74 9.17 8.34 8.19 8.50 \n85-89 5.56 5.45 5.69 6.68 6.53 6.84 6.24 6.14 6.36 \n90-94 4.31 4.24 4.38 4.86 4.76 4.95 4.68 4.61 4.75 \nSource: Institute for Health Metrics and Evaluation \n(https://gbd2019.healthdata.org/gbd-results/?params=gbd-api-2019-\npublic/f198e13d432a3a197b608d5f3e67099b). \n*LCI – Lower Confidence Interval \n**UCI – Upper Confidence Interval \n  \nYLL Attributable to Premature Mortality from DM in Bahia, Brazil \n The analyses revealed the following YLL attributable to premature mortality \nfrom DM in Bahia: 2000: 151.7/100,000 inhabitants; 2001: 178.9/100,000 inhabitants; \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n2002: 188.0/100,000 inhabitants; 2003: 190.9/100,000 inhabitants; 2004: 196.6/100,000 \ninhabitants; 2005: 189.2/100,000 inhabitants; 2006: 234.9/100,000 inhabitants; 2007: \n247.3/100,000 inhabitants; 2008: 262.0/100,000 inhabitants; 2009: 272.3/100,000 \ninhabitants; 2010: 284.6/100,000 inhabitants; 2011: 307.4/100,000 inhabitants; 2012: \n309.4/100,000 inhabitants; 2013: 309.6/100,000 inhabitants; 2014: 312.0/100,000 \ninhabitants; 2015: 330.0/100,000 inhabitants; 2016: 314.3/100,000 inhabitants; 2017: \n347.8/100,000 inhabitants; 2018: 341.0/100,000 inhabitants; 2019: 341.0/100,000 \ninhabitants; 2020: 422.6/100,000 inhabitants; 2021: 419,8/100,000 inhabitants; 2022: \n437,9/100,000 inhabitants; and 2023: 431,2/100,000 inhabitants (Graph 1).  \nGraph 1. Attributable to Premature Mortality from DM in Bahia \n \nSource: Study result \n \nThe increasing trend in YLL/100,000 indicates a growing burden of premature \nmortality due to DM in Bahia over the past two decades, with notable peaks in 2020. A \nsharp rise is observed between 2005 and 2007, followed by a plateau from 2012 to \n2019.  The spike in 2020 suggests that external factors, such as the COVID-19 \npandemic, may have contributed to increased diabetes-related deaths. \nInterpretive insights: \nAccelerated growth: Non-linear trend shows increasing slope after 2015 \n(\nβ 2015−2023 > β 2000−2015). \nPandemic amplification: 24.2% YLL increase from 2019 to 2020 \n(Δ =81.6/100kΔ =81.6/100k). \nSustained burden: Post-pandemic rates remain 26.4% above pre-2020 baseline. \n \nGender Disparity in YLL Due to Premature Diabetes Mellitus Mortality \n The age-standardized premature mortality rates, per 100,000 inhabitants, due to \nDM among individuals aged 30–69 years in Bahia, stratified by gender, reveal \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nsignificant trends and disparities over the period from 2000 to 2023. Among females, \nrates fluctuated moderately, peaking at 31.4 in 2009 and showing a slight increase to \n29.8 in 2020, likely influenced by the COVID-19 pandemic. In contrast, males \nexhibited higher and more pronounced rates, with a sharp peak of 34.9 in 2020, \nreflecting a greater vulnerability to diabetes-related complications during health crises. \nWhile both genders experienced a decline in mortality rates after 2009, males \nconsistently maintained higher rates, with a notable upward trend from 2014 onward, \nreaching 32.4 in 2015 and 34.5 in 2023. Females, on the other hand, demonstrated \nrelative stability, with rates hovering around 28–29 from 2016 to 2023. The data \nsuggests that males face a disproportionately higher burden of premature mortality due \nto Diabetes Mellitus, potentially due to differences in healthcare access, lifestyle factors, \nor biological susceptibility, as detailed in Graph 2 and Graph 3. \nGraph 2. YLL Due to Premature Death, DM, Bahia, Brazil 2000-2023, Female. \n \nSource: Study result \n \nGraph 3. YLL Due to Premature Death, DM, Bahia, Brazil 2000-2023, Male. \n \nSource: Study result \n \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n \nSocioeconomic Implications of Premature Diabetes Mellitus Mortality in Bahia \nIn 2020, the per capita income in Bahia was R$ 1,568, and in 2022, it was R$ \n1,613.17 Consequently, the economic loss attributable to premature mortality due to \ndiabetes mellitus was approximately 41 million Brazilian Reais. \nThe escalating premature mortality due to DM in Bahia, evidenced by the \nincreasing trend in YLL per 100,000 inhabitants, carries profound socioeconomic \nimplications for the state. Over the past 20 years, the surge in YLL, particularly the \nsharp increase observed between 2005 and 2007 and the peak in 2020, underscores the \ngrowing impact of DM on the working-age population. The peak in 2020, likely \nexacerbated by the COVID-19 pandemic, highlights how health crises can amplify the \nvulnerability of individuals with chronic conditions such as diabetes. Premature deaths \namong individuals aged 30 to 69 not only represent a tragic loss of lives but also \ndeprive families of their primary providers, disrupt household stability, and place \nadditional financial strain on surviving members. This loss of productivity and \nincreased reliance on social support systems further exacerbate economic inequalities, \nparticularly in low- and middle-income regions like Bahia (Graph 4). \nGraph 4. Trend YLL due to DM in Bahia. \n \nSource: Study result \n \nThe gender disparities in premature mortality rates reveal a disproportionate \nimpact between men and women, with rates peaking at 34.9 in 2020 and remaining \nelevated through 2023 (Graph 5).  \n \n \n \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n \nGraph 5. The gender disparities in premature mortality rates. \n \nSource: Study result \n \nThis trend suggests that men may face greater challenges in accessing \nhealthcare, adopting preventive measures, or managing DM-related complications, \npotentially due to social norms, occupational hazards, or biological factors. In contrast, \nwomen, although also affected, exhibit more stable mortality rates, indicating potential \nresilience or better engagement with health services. Addressing the socioeconomic \nimpact of premature DM mortality requires a comprehensive approach that integrates \npublic health strategies, economic support, and gender-sensitive policies to attenuate the \nlong-term consequences on families, communities, and the broader economy. \n \nDISCUSSION \n Our time-series analysis of premature DM mortality in Bahia, Brazil, reveals a \nconcerning trend of increasing YLL over the of more than two decades. This pattern \nunderscores the significant impact of DM on public health, particularly in terms of \npremature mortality. The escalating YLL rates suggest that DM continues to pose a \nsubstantial problem for the healthcare system and the community, highlighting the need \nfor targeted interventions and enhanced care strategies. Demographic characteristics \nsuch as age, sex, and socioeconomic factors play a significant role in understanding the \ndynamics of DM-related mortality. The observed trajectory of the study's results, \nmarked by a steady climb and punctuated by mortality spikes, demands a deeper \nexploration of the underlying factors driving this public health issue. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n Estimates of YLL represent a critical metric in public health, quantifying the \nimpact of premature mortality by calculating the difference between the age at death \nand a predetermined standard life expectancy.\n18 This index surpasses crude mortality \nrates, offering nuanced insights into the burden of diseases or injuries, particularly \nemphasizing deaths occurring at younger ages.\n19 Variations in standard life expectancy \nacross studies may affect comparability, necessitating methodological transparency and \ncontext-specific interpretations. Studies suggest that YLL varies significantly across \ndemographic data, highlighting disparities in healthcare access and socioeconomic \nfactors.\n20 YLL has been deemed indispensable for prioritizing public health \ninterventions and resource allocation. 21 This study evaluated the YLL attributable to \npremature mortality from 2000 to 2023 due to DM, aiming to assess the impact of DM \non population health. The evaluation focused on quantifying the YLL and its \nimplications for public health policy, which can provide important insights into the \ndisease burden and thereby lead to the development of strategies for targeted \ninterventions according to public health services. \n The demography of Bahia, a state with vast socioeconomic and cultural \ndiversity, presents a complex population profile. With an estimated population \nexceeding 14 million inhabitants, Bahia is the fourth most populous state in Brazil, \ncharacterized by an uneven distribution, featuring large urban concentrations and \nextensive rural areas.\n22 The ongoing demographic transition, marked by population \naging and declining birth rates, poses significant challenges for public health and social \nsecurity policies.\n23 Furthermore, the high prevalence of non-communicable chronic \ndiseases, such as DM and hypertension, necessitates targeted interventions for health \npromotion and disease prevention.\n24 In our study, demographic data for the Bahian \npopulation were sourced from the Health Information System of the Bahia State Health \nDepartment, with population estimates systematically compiled by the IBGE. This \napproach ensured the utilization of authoritative and comprehensive datasets, facilitating \nrobust demographic analysis and enhancing the reliability of our study. \n Standard life expectancy in Brazil has exhibited an upward trajectory, albeit with \nsignificant regional disparities. In Bahia, specifically, the analysis reveals a complex \nscenario, influenced by socioeconomic factors and public health determinants. Studies \nindicate that, despite advancements, challenges persist regarding premature mortality \nfrom chronic diseases and external causes, such as urban violence.\n25 Furthermore, life \nexpectancy distribution varies considerably between urban and rural areas of the state, \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nreflecting inequalities in access to healthcare services and basic sanitation. 26 The GBD \n2019 study corroborates these findings, highlighting the substantial burden of non-\ncommunicable diseases and interpersonal violence on YLL in Bahia, underscoring the \nnecessity for targeted public health interventions.\n27 Our analysis, based on the GDB, \ndemonstrates a perceptible trend of declining life expectancy with advancing age in both \nsexes. Notably, women consistently exhibit a higher life expectancy compared to men \nwithin each age cohort. This disparity underscores the influence of biological and \nsociocultural determinants on mortality patterns. The progressive reduction in life \nexpectancy across all age groups suggests an increased vulnerability to mortality as \nindividuals age, highlighting the importance of age-specific public health interventions. \n The DM significantly contributes to premature mortality, with its impact \nextending beyond direct mortality to encompass substantial YLL.\n28 This condition \naccelerates the onset of comorbidities, such as cardiovascular diseases and renal failure, \nthereby reducing life expectancy.\n29 The burden of premature mortality attributable to \nDM varies across populations, influenced by socioeconomic factors and access to \nhealthcare.\n30 In our study, the temporal analysis of YLL attributable to premature \nmortality due to diabetes mellitus in Bahia, Brazil, revealed an upward trend over the \nstudy period. This escalation suggests a progressive increase in the disease's impact on \nthe population's life expectancy. The observed fluctuations indicate potential influences \nfrom evolving healthcare access, socioeconomic changes, and public health \ninterventions, particularly during the COVID-19 pandemic period. \n The gender disparities in YLL due to premature mortality from DM highlight \nsignificant variations in the disease's impact. Studies consistently report a higher \nprevalence of YLL among men, potentially attributable to differential risk factors and \nneglect in accessing healthcare.\n31 Biological factors, such as hormonal differences and \ngenetic predispositions, may also contribute to these disparities. 32 Additionally, \nsociocultural determinants, including lifestyle behaviors and occupational exposures, \ninfluence disease progression and mortality. 33 Addressing these gender-specific factors \nis of paramount importance for developing targeted interventions aimed at reducing \npremature mortality due to DM. Our analysis of age-standardized premature mortality \nrates due to DM in Bahia revealed a persistent gender disparity. In our study, men \nexhibited a higher prevalence of mortality compared to women, consistent with data \nreported in the literature. While both genders displayed fluctuations, men demonstrated \na more pronounced upward trend, particularly after 2014. This suggests a greater \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nvulnerability among men to DM-related complications, potentially influenced by \nneglect in seeking medical care or lifestyle-related factors. \n Premature mortality due to DM carries significant socioeconomic implications, \ndisproportionately affecting individuals in lower socioeconomic strata. Studies indicate \nthat DM-related premature deaths contribute to substantial productivity losses, as they \npredominantly occur during working-age years, exacerbating economic burdens on \nfamilies and healthcare systems.\n34 Additionally, individuals from disadvantaged \nbackgrounds often face barriers to accessing timely and effective healthcare, leading to \ndelayed diagnoses and suboptimal disease management. 35 This perpetuates a cycle of \nhealth inequities, where socioeconomic determinants such as education, income, and \noccupation directly influence disease outcomes. 36 Addressing these disparities requires \ntargeted public health interventions, including improved healthcare access, education on \npreventive measures, and policies aimed at reducing socioeconomic inequities, to \nattenuate the broader societal impact of premature DM mortality. Based on the results of \nour study, it is evident that financial loss, coupled with declining per capita income, \nunderscores economic strain. The disproportionate impact on the working-age \npopulation, particularly exacerbated during the 2020 pandemic, highlights the fragility \nof chronic disease management during health crises. This results in productivity loss, \nfamily instability, and increased reliance on social support, exacerbating pre-existing \neconomic disparities in the region. Thus, premature mortality due to DM in Bahia, as \nevidenced by the rise in YLL, imposes substantial socioeconomic burdens. \n Thus, our analysis of YLL due to premature mortality from DM in Bahia, Brazil, \nunderscores a public health challenge. The rising YLL rates, coupled with pronounced \ngender disparities and socioeconomic impacts, demand immediate and targeted \ninterventions. By acknowledging the complex interplay of demographic, biological, and \nsociocultural factors, public health services can devise strategies to attenuate premature \nmortality among diabetic individuals, thereby improving population health and reducing \neconomic and social costs. \n \nCONCLUSION \nAnalysis of Bahia's demographic data, utilizing IBGE population estimates and \nGBD 2019 metrics, reveals an escalating trend in YLL due to premature DM mortality. \nThis increase, particularly pronounced post-2015 and exacerbated by the 2020 \npandemic, highlights a significant public health challenge. Gender disparities, with \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\nmales exhibiting higher mortality, and substantial socioeconomic impacts, including lost \nproductivity and economic strain, necessitate integrated public health and economic \ninterventions to attenuate this burden. \n \nCompeting Interests : No potential conflict of interest relevant to this article was \nreported. \n \nREFERENCES \n1. Lovic D, Piperidou A, Zografou I, Grassos H, Pittaras A, Manolis A. The \nGrowing Epidemic of Diabetes Mellitus. Curr Vasc Pharmacol. 2020;18(2):104-\n109. \n2. Revista Baiana de Saúde Pública Plano Estadual de Saúde 2024-2027./ \nSecretaria da Saúde do Estado da Bahia. - v. 47, supl.1, out./dez. 2023. \n3. Bracco PA, Gregg EW, Rolka DB, Schmidt MI, Barreto SM, Lotufo PA, et al. \nLifetime risk of developing diabetes and years of life lost among those with \ndiabetes in Brazil. J Glob Health. 2021;11:04041. \n4. Flor LS, Campos MR, Oliveira AF, Schramm JM. Diabetes burden in Brazil: \nfraction attributable to overweight, obesity, and excess weight. Rev Saúde \nPública. 2015;49:29. \n5. Costa AF, Flor LS, Campos MR, Oliveira AF, Costa MF, Silva RS, et al. Burden \nof type 2 diabetes mellitus in Brazil. Cad Saúde Pública. 2017;33(2):e00197915. \n6. Lima LO, Palmeira CS. Mortalidade por Diabetes Mellitus no estado da Bahia \nno período de 2012 a 2021. Ver Enferm Contemp. 2024;13:e5455. \n7. Chaves-Fonseca RM, Matos OS, Lordelo RA, Abreu M, Farias MG, Coutinho \nJF, et al. Implementation of a systematic approach to diabetes in primary care in \nBahia, Brazil improves metabolic outcomes: PRODIBA-Programa de \nInteriorização da Assistência ao Diabetes na Bahia (Project for Dissemination of \nDiabetes Care in the State of Bahia). Diabet Med. 2009;26(3):286-92. \n8. de Lima LO, Palmeira CS. Mortalidade por Diabetes Mellitus no estado da \nBahia no período de 2012 a 2021. Rev Enf Contemp [Internet]. 2024;13:e5455. \n9. Khunti K, Aroda VR, Aschner P, Chan JCN, Del Prato S, Hambling CE, et al. \nThe impact of the COVID-19 pandemic on diabetes services: planning for a \nglobal recovery. Lancet Diabetes Endocrinol. 2022;10(12):890-900. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n10. GBD 2021 Risk Factors Collaborators. Global burden and strength of evidence \nfor 88 risk factors in 204 countries and 811 subnational locations, 1990-2021: a \nsystematic analysis for the Global Burden of Disease Study 2021. Lancet. \n2024;403(10440):2162-2203. \n11. NCD Countdown 2030 collaborators. NCD Countdown 2030: worldwide trends \nin non-communicable disease mortality and progress towards Sustainable \nDevelopment Goal target 3.4. Lancet. 2018;392(10152):1072-1088. \n12. Martinez R, Soliz P, Caixeta R, Ordunez P. Reflection on modern methods: \nyears of life lost due to premature mortality-a versatile and comprehensive \nmeasure for monitoring non-communicable disease mortality. Int J Epidemiol. \n2019;48(4):1367-1376. \n13. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJL. Measuring the \nGlobal Burden of Disease and Risk Factors, 1990-2001. In: Lopez AD, Mathers \nCD, Ezzati M, Jamison DT, Murray CJL, editors. Global Burden of Disease and \nRisk Factors. Washington (DC): The International Bank for Reconstruction and \nDevelopment / The World Bank; 2006. Chapter 1. \n14. Lomas JRS. Incorporating Affordability Concerns Within Cost-Effectiveness \nAnalysis for Health Technology Assessment. Value Health. 2019;22(8):898-\n905. \n15. Tsuchiya A, Dolan P, Shaw R. Measuring people's preferences regarding ageism \nin health: some methodological issues and some fresh evidence. Soc Sci Med. \n2003;57(4):687-96. \n16. Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al, Global \nand regional mortality from 235 causes of death for 20 age groups in 1990 and \n2010: a systematic analysis for the Global Burden of Disease Study 2010. \nLancet. 2012;380(9859):2095-128. \n17. BRASIL. Instituto Brasileiro de Geografia e Estatística (IBGE). PNAD \nContinua. 2022.\n Available \nat:<https://www.ibge.gov.br/estatisticas/sociais/trabalho/17270-pnad-\ncontinua.html?edicao=36796&t=publicacoes> Accessed: March 22, 2025 \n18. Chudasama YV, Khunti K, Gillies CL, Dhalwani NN, Davies MJ, Yates T, et al. \nEstimates of years of life lost depended on the method used: tutorial and \ncomparative investigation. J Clin Epidemiol. 2022;150:42-50. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n19. Bailey HM, Zuo Y, Li F, Min J, Vaddiparti K, Prosperi M, et al. Changes in \npatterns of mortality rates and years of life lost due to firearms in the United \nStates, 1999 to 2016: A joinpoint analysis. PLoS One. 2019;14(11):e0225223. \n20. Chen-Xu J, Varga O, Mahrouseh N, Eikemo TA, Grad DA, Wyper GMA, et al. \nSubnational inequalities in years of life lost and associations with \nsocioeconomic factors in pre-pandemic Europe, 2009-19: an ecological study. \nLancet Public Health. 2024;9(3):e166-e177. \n21. Duque-Molina C, García-Rodríguez G, Zaragoza-Jiménez CA, Torre-Rosas A, \nHerrera-Canales M, Loera-Rosales MJ, et al. Impact on Fatality Rates and Years \nof Life Lost During the COVID-19 Pandemic: The Experience of the Mexican \nPublic Health Incident Management Command. Arch Med Res. \n2025;56(1):103073. \n22. Instituto Brasileiro de Geografia e Estatística (IBGE). (2022). Censo \nDemográfico 2022. \n23. Carvalho JA, Rodríguez-Wong LL. The changing age distribution of the \nBrazilian population in the first half of the 21st century. Cad Saude Publica. \n2008;24(3):597-605. \n24. Teixeira RA, Ishitani LH, Marinho F, Pinto Junior EP, Katikireddi SV, Malta \nDC. Methodological proposal for the redistribution of deaths due to garbage \ncodes in mortality estimates for Noncommunicable Chronic Diseases. Rev Bras \nEpidemiol. 2021;24(suppl 1):e210004. \n25. Freeman T, Gesesew HA, Bambra C, Giugliani ERJ, Popay J, Sanders D, et al. \nWhy do some countries do better or worse in life expectancy relative to income? \nAn analysis of Brazil, Ethiopia, and the United States of America. Int J Equity \nHealth. 2020;19(1):202. \n26. De Carvalho JA, Wood CH. [Mortality, distribution of income and rural-urban \nresidence in Brazil]. Estud Poblac. 1978;3(7-12):7-19. \n27. GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases \nand injuries in 204 countries and territories, 1990-2019: a systematic analysis for \nthe Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204-1222. \n28. Wang F, Wang W, Yin P, Liu Y, Liu J, Wang L, et al. Mortality and Years of \nLife Lost in Diabetes Mellitus and Its Subcategories in China and Its Provinces, \n2005-2020. J Diabetes Res. 2022;2022:1609267. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint \n\n29. Lee AK, Steinman MA, Lee SJ. Improving the American Diabetes Association \nFramework for individualizing treatment in older adults: evaluating life \nexpectancy. BMJ Open Diabetes Res Care. 2020;8(1):e001624. \n30. Bommer C, Heesemann E, Sagalova V, Manne-Goehler J, Atun R, \nBärnighausen T, et al. The global economic burden of diabetes in adults aged \n20-79 years: a cost-of-illness study. Lancet Diabetes Endocrinol. 2017;5(6):423-\n430. \n31. Costa AF, Flor LS, Campos MR, Oliveira AF, Costa MF, Silva RS, et al. Burden \nof type 2 diabetes mellitus in Brazil. Cad Saude Publica. 2017;33(2):e00197915. \n32. Al-Shamsi S, Govender RD, Soteriades ES. Mortality and potential years of life \nlost attributable to non-optimal glycaemic control in men and women with \ndiabetes in the United Arab Emirates: a population-based retrospective cohort \nstudy. BMJ Open. 2019;9(9):e032654. \n33. Medina-Gómez OS, Medina-Reyes IS. Mortality from type 2 diabetes and \nimplementation of the PREVENIMSS program: a time series study in Mexico, \n1998-2015. Cad Saude Publica. 2018;34(5):e00103117. \n34. Park J, Bigman E, Zhang P. Productivity Loss and Medical Costs Associated \nWith Type 2 Diabetes Among Employees Aged 18-64 Years With Large \nEmployer-Sponsored Insurance. Diabetes Care. 2022;45(11):2553-2560. \n35. Giebel C, Corcoran R, Goodall M, Campbell N, Gabbay M, Daras K, et al. Do \npeople living in disadvantaged circumstances receive different mental health \ntreatments than those from less disadvantaged backgrounds? BMC Public \nHealth. 2020;20(1):651. \n36. Liu C, He L, Li Y, Yang A, Zhang K, Luo B. Diabetes risk among US adults \nwith different socioeconomic status and behavioral lifestyles: evidence from the \nNational Health and Nutrition Examination Survey. Front Public Health. \n;2023;11:1197947. \n . CC-BY 4.0 International licenseIt is made available under a \n is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)\nThe copyright holder for this preprint this version posted March 25, 2025. ; https://doi.org/10.1101/2025.03.24.25324575doi: medRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}