Scalable Diabetes Preventive Care Through Household Risk Stratification Using Electronic Health Records

Diabetes prevention is among the most effective population health strategies, yet the current approaches focus largely on individuals, despite strong evidence that diabetes risk clusters within families. 1 Diabetes affects more than 38 million US adults and costs an estimated $412 billion annually, yet participation in programs such as the National Diabetes Prevention Program (DPP) remains low. 2 In a new study, Thomas et al 3 demonstrate that electronic health records (EHRs) can be used to algorithmically identify households at elevated risk of diabetes, offering a scalable strategy to support family-centered prevention.This study illustrates how existing clinical data can automate household-level risk stratification and broaden preventive outreach. Evolving Approaches to Family Linkage in Health DataLinking family members within health data has long been a priority, but its evolution has been incremental.Early efforts relied on manual collection of family history, survey-based linkages, or genetic registries, later supplemented by computational approaches. 4Some health systems created algorithms linking mothers and newborns for perinatal care or used emergency contact information to connect parents and children. 5Consortia such as SUPREME-DM (Surveillance, Prevention, and Management of Diabetes Mellitus) demonstrated multisystem EHR integration for diabetes surveillance, and integrated delivery systems have examined neighborhood-level diabetes burden. 6re recently, health care-integrated research databases have enabled family linkage, although preventive applications remain limited. 7International work, including UK initiatives using property reference identifiers, shows promise when accurate address infrastructures exist. 8Thomas et al 3 build on this progress by demonstrating a practical, clinical-grade method to identify households at risk of diabetes, without new data collection.Taken together, these efforts reflect steady progress toward harnessing family context in health data. Why Households Matter for PreventionConsider a 52-year-old patient with prediabetes at a routine visit.In traditional care models, clinicians focus on the individual's risk factors and may recommend enrollment in a diabetes prevention program.In contrast, a household-based approach would recognize that this patient lives with a spouse who has undiagnosed prediabetes and an adolescent with obesity, creating an opportunity for coordinated, family-centered intervention rather than 3 separate, disconnected encounters.However, despite growing awareness of the value of family-based prevention, most health systems still lack automated tools to identify household structures within EHRs.As a result, individuals who share environments, risks, and opportunities for prevention often remain invisible to care systems.This gap is notable because behaviors, food environments, cultural practices, and shared stressors shape diabetes risk as powerfully as genetics.Family-based diabetes prevention programs remain relatively rare.Barriers include logistical constraints such as time and transportation, inconsistent family support, program costs, and social determinants of health, including food access.Although these challenges affect family-based interventions, individual-level strategies have similarly struggled to achieve broad participation and