How EHRs Help Identify Diabetes Risk in Families of Adults With Prediabetes

Diabetes doesn’t just affect individuals; it runs in families. In a cohort of adults with prediabetes, researchers found that 76% of multiresident households included at least 1 other family member with risk factors for diabetes.1 By leveraging electronic health record (EHR) data, health systems can identify at-risk household members and implement targeted prevention strategies, offering a new frontier in family-centered diabetes care.

Recent findings support the use of EHRs not only for individual patient care but also for family-centered interventions to curb the growing diabetes epidemic. | Image credit: abidika - stock.adobe.com

This observational index cohort study is published in JAMA Network Open.

“In a cohort of patients with prediabetes and their household members, 75.9% of multiresident households had at least 1 additional household member with diabetes risk factors,” wrote the researchers of the study.

Family history is a well‑established risk factor for developing diabetes, reflecting both inherited genetic predisposition and shared environmental or lifestyle influences.2 Individuals with close relatives who have diabetes—especially type 2 diabetes—are at higher risk than those without a family history, as genes and behaviors like diet and activity patterns often run in families. For type 1 diabetes, having a parent with the condition significantly increases a child’s likelihood of developing it compared with the general population, though genetics alone do not determine disease onset.

The study leveraged EHR data from a large health system to identify adults with prediabetes (fasting plasma glucose 100-125 mg/dL or hemoglobin A1c 5.7%-6.4%) and a body mass index of 25 or higher between January 1 and December 31, 2023.1 Adults with existing diabetes, end-stage kidney disease, or less than 12 months of continuous health system membership were excluded. Household members were identified through shared subscriber medical record numbers and address history.

Diabetes risk factors were assessed for individuals aged 10 years or older, with adult risk defined as overweight/obesity or a history of gestational diabetes, hypertension, dyslipidemia, cardiovascular disease, prediabetes, or diabetes; pediatric risk included overweight/obesity or prediabetes/diabetes.

Among 356,626 adults with prediabetes (mean [SD] age of 50.5 [12] years; 51.7% women), 58.5% were classified as obese. Households were nearly evenly split between single-person (48.1%) and multiresident (51.9%) households. The cohort was racially and ethnically diverse: 27.9% Asian American, 26.6% Hispanic or Latino, 9.1% Black or African American, 30.1% White, 1.1% Hawaiian or Pacific Islander, 0.4% Native American or American Indian, 0.8% multiracial, and 4.1% unknown.

In multiresident households, 75.9% included at least 1 additional household member with risk factors for diabetes. These findings highlight that a substantial proportion of family members living with adults with prediabetes also carry diabetes risk factors, underscoring the potential for EHR-enabled household-level identification and prevention strategies.

However, the researchers acknowledged several study limitations. Because it relied on EHR data from a single health system, the findings may not be fully generalizable to the broader population. Risk factor assessments for household members may be incomplete, potentially underestimating the true prevalence of diabetes risk. Additionally, the study could identify only coinsured household members, which may have underestimated household size and, conversely, overestimated the proportion of family members with risk factors. These factors should be considered when interpreting the findings and assessing the potential impact of household-level diabetes screening interventions.

Despite these limitations, the study demonstrates that adults with prediabetes often live with other household members who carry diabetes risk factors, with more than three-quarters of multiresident households affected.

“Our study highlights that EHR data can be used to identify households at risk for diabetes,” wrote the researchers. “Health systems could use EHRs to screen family members for risk factors and support care coordination among families for cardiometabolic risk reduction. Family care coordination could reduce health system costs; however, few prevention programs enroll households at risk, reflecting a missed opportunity for population-level diabetes prevention.”

References

1. Thomas TW, Finertie H, Silverberg A, et al. Identifying household diabetes risk for family diabetes prevention using electronic health records. JAMA Netw Open. 2026;9(1):e2551823. doi:10.1001/jamanetworkopen

2. Genetics of diabetes. American Diabetes Association. Accessed January 12, 2026. https://diabetes.org/about-diabetes/genetics-diabetes