Annals of Epidemiology最新文献

{"title":"Beyond associations: From theory to interventions in health inequalities research using causal inference","authors":"Maria Gueltzow ,&nbsp;Maarten J. Bijlsma ,&nbsp;Frank J. van Lenthe","doi":"10.1016/j.annepidem.2025.10.001","DOIUrl":"10.1016/j.annepidem.2025.10.001","url":null,"abstract":"<div><div>One of the central goals of public health is not only to improve the health in the population overall, but also to reduce the unequal distribution of health and disease within the population. Even though a large amount of research is directed towards identifying and understanding health inequalities, much of this research is based on associations. This type of research can help to identify what groups in society are at risk of having worse health but cannot tell us how these inequalities may be reduced. In order to move beyond identifying who is at risk, we illustrate how we can combine the existing theoretical foundations with the counterfactual outcomes framework to understand how health inequalities can be tackled. We show how the Commission on Social Determinants of Health (CSDH) framework and the Diderichsen model can be translated into practice through the use of DAGs and notation. This will aid in generating more informative evidence on how certain interventions can reduce inequalities.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 1-7"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145253682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"COMBINE EBIs: A novel COllaborative Method for Building INterventions from Existing Evidence-Based Interventions","authors":"Justin Knox ,&nbsp;Kara Portier ,&nbsp;Christopher Magana ,&nbsp;Maggie Denning ,&nbsp;Christopher M. Ferraris ,&nbsp;Emily Dove-Medows ,&nbsp;Cho-Hee Shrader ,&nbsp;Ohemaa Poku ,&nbsp;Phil Kreniske ,&nbsp;Robert Remien ,&nbsp;Efrat Aharonovich ,&nbsp;Jennifer C. Elliott ,&nbsp;Denis Nash ,&nbsp;Kathryn Lancaster ,&nbsp;Till Baernighausen ,&nbsp;Kayo Fujimoto ,&nbsp;Adam Carrico ,&nbsp;John A. Schneider ,&nbsp;Alida Bouris ,&nbsp;D. Scott Batey ,&nbsp;Stefan D. Baral","doi":"10.1016/j.annepidem.2025.10.018","DOIUrl":"10.1016/j.annepidem.2025.10.018","url":null,"abstract":"<div><h3>Purpose</h3><div>Combining evidence-based interventions (EBIs) is a discrete process from adapting EBIs, and specific guidance for how to combine EBIs could be helpful amidst the proliferation of frameworks that combine and stage EBIs and calls for services to be combined or bundled. To address this gap, we developed and applied the <strong>CO</strong>llaborative <strong>M</strong>ethod for <strong>B</strong>uilding <strong>IN</strong>terventions from <strong>E</strong>xisting <strong>E</strong>vidence-<strong>B</strong>ased <strong>I</strong>nterventions (<strong>COMBINE-EBIs</strong>) approach, a five-step process for combining EBIs.</div></div><div><h3>Methods</h3><div>The five steps of COMBINE-EBIs are: (1) Identify and Select EBIs, (2) Develop a Shared Conceptual Model, (3) Evaluate the Conceptual Model, (4) Create a Single Combined Protocol, and (5) Refine through Further Input.</div></div><div><h3>Results</h3><div>We developed and applied the 5-step COMBINE-EBIs process to build a fully refined, pre-tested, combined multi-component intervention that leverages intrinsic social network support and mHealth technology to support people with HIV who drink heavily improve HIV care outcomes and reduce alcohol use.</div></div><div><h3>Conclusions</h3><div>COMBINE-EBIs is a rigorous, systematic and efficient approach for building multi-component, multi-modal interventions to address multiple, co-occurring health behaviors simultaneously. COMBINE-EBIs is a resource efficient approach that could facilitate the creation of additional multi-component interventions to address complex, co-occurring health conditions synergistically. Future research should evaluate the feasibility and utility of COMBINE-EBIs, including where adaptations are needed to maximize utility.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 64-75"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145477352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cumulative exposure to economic hardship and self-rated health among Korean women: An exploration of age heterogeneity","authors":"Gum-Ryeong Park Ph.D. ,&nbsp;Jinho Kim Ph.D.","doi":"10.1016/j.annepidem.2025.10.017","DOIUrl":"10.1016/j.annepidem.2025.10.017","url":null,"abstract":"<div><h3>Purpose</h3><div>Acknowledging the importance of subjective financial measures that objective indicators may not be able to fully capture, this study investigates whether and how perceived economic hardship influences self-rated health among women. Specifically, it examines the cumulative effects of perceived economic hardship while exploring variations across different age groups.</div></div><div><h3>Methods</h3><div>This study analyzed data from the Korean Longitudinal Survey of Women &amp; Families (2006–2022), including 12,800 participants who experienced varying levels of economic hardship. Economic hardship was assessed based on subjective perceptions reported across consecutive survey waves (ranging from 1 wave to over 4 waves), while self-rated health was measured on a five-point scale. To account for unmeasured individual-level heterogeneity, fixed effects models were employed in the analysis.</div></div><div><h3>Results</h3><div>Prolonged exposure to economic hardship is associated with greater declines in self-rated health, with longer durations of hardship leading to increasingly severe negative impacts. Also, age differences were observed, as older adults experienced significantly larger declines in self-rated health compared to their younger counterparts as the duration of hardship increased.</div></div><div><h3>Conclusion</h3><div>The findings on the cumulative effects of perceived economic hardship on health underscore the importance of incorporating subjective measures of economic conditions into research and policy discussions.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 46-52"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145419391","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comparison of methods for coding race in linear and logistic regression models","authors":"Melody S. Goodman ,&nbsp;Ariana Lopez ,&nbsp;Anarina L. Murillo ,&nbsp;Kristyn A. Pierce","doi":"10.1016/j.annepidem.2025.10.005","DOIUrl":"10.1016/j.annepidem.2025.10.005","url":null,"abstract":"<div><div>In many public health and clinical research studies that use regression models for analyses, race is often considered a confounder and \"controlled\" for in the regression model with simple indicators for race and non-Hispanic White as the reference group, without much introspection from the data analyst. From a health equity perspective, multiple issues exist with this approach. We examine and compare several methods for coding race in linear and logistic regression models. We compare several coding methods using a sample of 8097 participants (≥18 years old) from the 2020 New York City Community Health Survey. To illustrate the importance of coding methods for race, we conducted regression analyses to compare the results from six coding approaches: dummy, simple effect, difference (forward and backward), deviation, and analyst-defined coding. Body mass index measured continuously and diabetes status measured dichotomously were the outcome variables in the linear and logistic regression models. Results showed that selecting a coding method has implications for identifying racial health inequities. The reference group selection is critical to measuring racial inequities in health outcomes. This study emphasizes the need to consider the impact of coding techniques on research study design, particularly when racial health inequities are the research focus.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 15-22"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145318839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Capturing the implications of residential segregation for the dynamics of infectious disease transmission","authors":"Jon Zelner ,&nbsp;Danielle Stone ,&nbsp;Marisa C. Eisenberg ,&nbsp;Andrew F. Brouwer ,&nbsp;Krzysztof Sakrejda","doi":"10.1016/j.annepidem.2025.11.001","DOIUrl":"10.1016/j.annepidem.2025.11.001","url":null,"abstract":"<div><h3>Purpose</h3><div>Residential segregation is linked to racial and socioeconomic inequity in outcomes for numerous infections including SARS-CoV-2, influenza, STIs, and tuberculosis. Despite the importance of segregation as a driver of infection inequity, there are few mathematical models to inform our understanding of these dynamics.</div></div><div><h3>Methods</h3><div>We developed a transmission model including mechanistic relationships between residential segregation and infection inequity. We conceptualize segregation as a fundamental social cause of infection inequity that jointly impacts contact patterns and vulnerability to infection.</div></div><div><h3>Results</h3><div>We show that the basic reproduction number, <span><math><msub><mrow><mi>R</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>, and equilibrium prevalence are sensitive to interactions between these factors. Our results show that separation alone is insufficient to explain segregation-associated differences in infection risks. Increasing separation only results in concentration of risk in segregated populations when accompanied by increasing vulnerability.</div></div><div><h3>Conclusions</h3><div>This work shows why it is important to consider causal linkages between high-level social determinants - like segregation - and more-proximal transmission mechanisms when crafting and evaluating public health policies. While the framework in this analysis is stylized, it lays the groundwork for data-driven explorations of the mechanistic impact of residential segregation on infection inequities.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 102-109"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145524773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Planning for the sustainability of a youth suicide prevention program in Native American contexts: A modeling study","authors":"Lauren Yan ,&nbsp;Zhixi Chen ,&nbsp;Novalene Goklish ,&nbsp;Kristin Mitchell ,&nbsp;Charity Watchman ,&nbsp;Meredith Stifter ,&nbsp;Victoria O’Keefe ,&nbsp;Allison Barlow ,&nbsp;Mary Cwik ,&nbsp;Takeru Igusa ,&nbsp;Emily E. Haroz","doi":"10.1016/j.annepidem.2025.10.020","DOIUrl":"10.1016/j.annepidem.2025.10.020","url":null,"abstract":"<div><h3>Purpose</h3><div>We aimed to identify actionable, effective sustainment strategies for a community-based suicide prevention program implemented in Tribal contexts through a participatory process of system dynamics modeling.</div></div><div><h3>Methods</h3><div>Through a series of workshops with teams implementing a suicide prevention program, we prioritized strategies for sustaining implementation and related health outcomes. We used system dynamics modeling and microsimulations to assess the impact of key strategies (<em>increased funding, enhanced program management, and leadership development</em>) on program sustainment outcomes and suicidal behavior among youth.</div></div><div><h3>Results</h3><div><em>Enhanced program management</em> impacted sustainment by increasing simulated levels of community trust by 31 % and community and external collaborators by 10 %. <em>Increased funding</em> increased simulated resource availability by 46 % and staffing levels by 13 % over the simulated time frame. Among the three simulated sustainment strategies, <em>enhanced program management</em> most improved clinical outcomes, decreasing non-suicidal self-injury by 1.4 % and suicide attempts by 1.3 %.</div></div><div><h3>Conclusions</h3><div>We collaboratively developed a simulation model that assessed the relative impact of stakeholders’ prioritized sustainment strategies. A multifaceted intervention for <em>enhanced program management</em> had the greatest impact on program sustainment outcomes. This approach can benefit Tribal communities who are considering methods to support vital community-based programs.</div></div><div><h3>Keywords</h3><div>System dynamics; Microsimulation; Community-based participatory research; Native American; Suicide prevention; Surveillance; Sustainability</div></div><div><h3>List of abbreviations and acronyms</h3><div>AI/AN, American Indian and Alaska Native; WMAT, White Mountain Apache Tribe; JHCIH, Johns Hopkins Center for Indigenous Health; C-CASA, Columbia Classification Algorithm for Suicide Assessment; GLS, Garrett Lee Smith Memorial Act; CDC, Centers for Disease Control and Prevention; NHS, National Health Service; SBIRT, Screening, Brief Intervention, Referral to Treatment; PHPCS, Public Health Program Capacity for Sustainability; PSAT, Program Sustainability Assessment Tool; SD, system dynamics; IHS, Indian Health Service</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 84-93"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145507878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Availability of sexual orientation and gender identity (SOGI) information in a cohort of transgender and gender diverse people: An analysis of electronic health records","authors":"Cynthia N. Ramirez ,&nbsp;Michael Goodman ,&nbsp;Kristine Magnusson ,&nbsp;Wendy Leyden ,&nbsp;Alexandra N. Lea ,&nbsp;Darios Getahun ,&nbsp;Courtney McCracken ,&nbsp;Suma Vupputuri ,&nbsp;Lee Cromwell ,&nbsp;Timothy L. Lash ,&nbsp;Oumaima Kaabi ,&nbsp;Guneet K. Jasuja ,&nbsp;Michael J. Silverberg","doi":"10.1016/j.annepidem.2025.10.016","DOIUrl":"10.1016/j.annepidem.2025.10.016","url":null,"abstract":"<div><h3>Purpose</h3><div>Electronic health records (EHR) offer a unique opportunity to systematically collect sexual orientation and gender identity (SOGI) data. This study examined the prevalence and determinants of SOGI reporting in an EHR-based cohort of transgender and gender diverse (TGD) individuals.</div></div><div><h3>Methods</h3><div>We identified TGD people with and without SOGI documentation across four Kaiser Permanente health plans from January 1, 2022–2024. TGD status was determined through clinical notes, diagnostic codes, and SOGI data based on a previously established cohort. Factors associated with SOGI reporting were assessed using log-binomial regression, yielding prevalence ratios (PR) and the 95 % confidence intervals (CI).</div></div><div><h3>Results</h3><div>Among 23,060 TGD individuals, 71 % had SOGI documentation in the EHR. Reporting varied by sociodemographic and clinical characteristics. For example, compared to those &lt; 20 years, SOGI reporting was higher for those aged 21–59 (PRs 1.10–1.21; 95 % CIs 1.06–1.24) and lower for those &gt; 60 (0.93; 0.88–0.99). Documentation was slightly lower for those assigned male at birth (0.98; 0.97–1.00) and varied by race and ethnicity (e.g., Hispanic: 0.97; 0.95–0.99; Other: 1.02; 0.98–1.05 vs. White).</div></div><div><h3>Conclusions</h3><div>KP’s EHRs captured SOGI data for over 70 % of TGD individuals, though more research is needed to understand factors associated with missing data not captured in structured fields.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 23-27"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145314098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Piloting the novel Evidence Synthesis Equity Companion (ESEC) tool","authors":"Maria Benkhalti,&nbsp;Talia Salzman,&nbsp;Prinon Rahman,&nbsp;Ayan Hashi,&nbsp;Laura Boland,&nbsp;Mallory Drysdale","doi":"10.1016/j.annepidem.2025.11.002","DOIUrl":"10.1016/j.annepidem.2025.11.002","url":null,"abstract":"<div><h3>Introduction</h3><div>The need for greater evidence on equity-denied population groups continues to be recognized across public health, epidemiology, and community medicine. Despite a number of existing frameworks and tools on the inclusion of equity considerations in evidence syntheses, there was a gap in tools providing step-wise guidance and facilitating the documentation of the process.</div></div><div><h3>Methods</h3><div>We adapted the Equity Checklist for Health Technology Assessment (ECHTA) to the more specific steps of an evidence synthesis. We complemented the guidance provided from seminal resources with articles from a comprehensive literature search and input from expert methodologists. We piloted the tool with a rapid review on substance-related mortality across racialized and ethic groups. We reported on the methodological changes following the use of the tool and the limitations we have observed.</div></div><div><h3>Results</h3><div>The Evidence Synthesis Equity Companion (ESEC) tool provides guidance and structure to considering equity and the methods to use at each step of an evidence synthesis. It allows review authors to transparently document their process and explicitly highlight its limitations. During the pilot, using the ESEC tool resulted in enhanced approaches to answer the equity-focused review question. Certain limitations remain with the tools, notably a more thorough guidance development for incorporating equity in the knowledge translation process.</div></div><div><h3>Conclusion</h3><div>The ESEC tool provides structured guidance, supporting the incorporation of equity considerations throughout the evidence synthesis process. The pilot demonstrated its benefits both for the methods used and the evidence uncovered. Additionally, the process was valuable to the review team. Using the tool in different types of evidence syntheses will allow for greater insight into its value-added.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 94-101"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145530885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trends and cyclical patterns of dengue disease in Mexico: A 40-year time series analysis","authors":"Jaime Briseno-Ramirez ,&nbsp;Judith Carolina De Arcos-Jiménez ,&nbsp;Ana María López-Yáñez ,&nbsp;Roberto Miguel Damián-Negrete ,&nbsp;Patricia Noemi Vargas-Becerra ,&nbsp;Laura Karina Salas-Salazar ,&nbsp;Berenice Martínez-Melendres ,&nbsp;Ismael Caballero-Quirarte ,&nbsp;Pedro Martínez-Ayala","doi":"10.1016/j.annepidem.2025.10.022","DOIUrl":"10.1016/j.annepidem.2025.10.022","url":null,"abstract":"<div><h3>Purpose</h3><div>To quantify long-term trends and multi-year cycles in dengue in Mexico (1985–2025) and examine recent serotype–severity patterns.</div></div><div><h3>Methods</h3><div>We integrated open national surveillance data from the Morbidity Yearbook (1985–2024) and Open Data (2020–2025) using the official estimated-cases definition. Monthly series underwent STL decomposition, Butterworth filtering, periodogram, and autocorrelation analyses. Stationarity was assessed in prespecified windows (1985–2000; 2000–2015; 2015–2025) and by geography (coastal/high-burden vs inland). Serotype distribution and hospitalizations (2020–2025) were summarized among laboratory-confirmed cases and compared across years with χ² tests.</div></div><div><h3>Results</h3><div>Over 40 years we identified 1715,456 estimated cases; annual burden increased (Mann–Kendall z = 3.25; Sen’s slope ≈ +1275 cases/year). A recurrent multi-year oscillation overlaid the seasonal cycle, with a spectral peak near 5.8 years and an ACF secondary maximum at ∼4.3 years; the signal persisted in coastal/high-burden states and across historical windows. Since 2023, DENV-3 predominated (86 % in 2024; 94.9 % by May-2025) and coincided with higher hospitalization proportions.</div></div><div><h3>Conclusions</h3><div>Dengue in Mexico shows a significant long-term rise and a reproducible ∼5-year cycle. Cycle-aware baselines, paired with serotype surveillance, could sharpen early-warning thresholds and targeting of vector control.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 53-63"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145446558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Guiding artificial intelligence in public health and medicine with epidemiology: A lifecycle framework for mitigating AI misalignment","authors":"Ahmed Hassoon MD, MPH ,&nbsp;Christine Lin MPH ,&nbsp;Hyun Yi (Jacqualine) Woo MD, MPH ,&nbsp;Ruxandra Irimia MD, PHD, MPH ,&nbsp;Jill A. Marsteller PhD, MPP ,&nbsp;Anthony Li MD, PMH ,&nbsp;Antonio Banderia MD, MPH ,&nbsp;Hubert Leo BS ,&nbsp;Xiaoyi Peng ScM ,&nbsp;David Rastall MD ,&nbsp;Mark Dredze PhD","doi":"10.1016/j.annepidem.2025.11.004","DOIUrl":"10.1016/j.annepidem.2025.11.004","url":null,"abstract":"<div><div>Artificial Intelligence (AI) holds immense promise for public health, yet its potential is undermined by alignment failures where systems act contrary to human values, often exacerbating health disparities. This paper challenges the narrow view that algorithmic bias is solely a data problem, arguing instead that misalignment arises at every stage of the AI development lifecycle. We introduce a comprehensive seven-stage framework, spanning problem definition, team assembly, study design, data acquisition, model training, validation, and post-deployment implementation, viewed through an epidemiological lens. This approach systematically integrates core principles such as population representativeness, rigorous study design, bias characterization, and causal reasoning to identify and mitigate alignment risks. For each stage, we define specific alignment failures, from flawed problem formulation to post-market performance degradation, and propose actionable, evidence-based solutions. By embedding epidemiological rigor throughout the entire AI lifecycle, this framework provides a structured, proactive pathway for researchers, developers, and policymakers to create trustworthy, safe, and fair AI systems. This systemic approach is critical to harnessing AI's transformative benefits for population health while preventing the perpetuation of inequity and harm.</div></div>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":"112 ","pages":"Pages 119-126"},"PeriodicalIF":3.0,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145571772","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}