Evaluation of family history as a risk factor and screening tool for detecting undiagnosed diabetes in a nationally representative survey population.

Abstract

We examined the utility of a three-level familial risk stratification system as a screening tool for diabetes in a nationally representative sample of the U.S. adult population. National Health and Nutrition Examination Survey data were used to assess the prevalence and distribution of familial risk for diabetes, the association between three levels of familial risk and undiagnosed diabetes, and the use of familial risk as a screening tool for diabetes, alone and in combination with body mass index and age. The prevalence of undiagnosed diabetes was 3% and increased with increasing familial risk (average = 2%, moderate = 4%, high = 10%). High familial risk was significantly associated with undiagnosed diabetes (adjusted odds ratio = 4.6; 95% confidence interval: 1.9-11.3). The use of a three-tiered familial risk stratification for diabetes screening yielded higher specificity (94%) and positive predictive value (9.9%) for high familial risk than body mass index > or = 25 (specificity = 38%, positive predictive value = 4.2%). High familial risk and body mass index > or = 25 combined had higher specificity (97%) and positive predictive value (13.4%); the addition of age > or = 45 years further improved positive predictive value (21.0%) without reducing specificity. There was a strong and proportional association between familial risk and undiagnosed diabetes, suggesting that a three-tiered assessment of familial diabetes risk may increase the effectiveness of diabetes screening.