Undiagnosed Diabetes in US Adults: Prevalence and Trends

OBJECTIVE Confirmatory testing is recommended for diabetes diagnosis in clinical practice. However, national estimates of undiagnosed diabetes are based on single elevated test measures, potentially resulting in overestimation. Our objective was to update trends in undiagnosed diabetes using definitions consistent with clinical practice. RESEARCH DESIGN AND METHODS We included 30,492 adults (aged >= 20 years) from the National Health and Nutrition Examination Survey (1988-2020). Among adults without diagnosed diabetes, confirmed undiagnosed diabetes was defined as having both elevated levels of fasting plasma glucose (FPG) (>= 126 mg/dL) and elevated glycated hemoglobin (HbA(1c); >= 6.5%), and persistent undiagnosed diabetes was defined as having elevated HbA(1c) or FPG levels, adjusted for the within-person variability in HbA(1c) and FPG tests. RESULTS From the periods 1988-1994 to 2017 to March 2020, there was an increase in the prevalence of diagnosed diabetes (from 4.6% to 11.7%), but no change in prevalence of persistent undiagnosed diabetes (from 2.23% to 2.53%) or confirmed undiagnosed diabetes (from 1.10% to 1.23%). Consequently, the proportion of all undiagnosed diabetes cases declined from 32.8% to 17.8% (persistent undiagnosed diabetes) and from 19.3% to 9.5% (confirmed undiagnosed diabetes). Undiagnosed diabetes was more prevalent in older and obese adults, racial/ethnic minorities, and those without health care access. Among persons with diabetes, Asian Americans and those without health care access had the highest proportion of undiagnosed cases, with rates ranging from 23% to 61%. CONCLUSIONS From 1988 to March 2020, the proportion of undiagnosed diabetes cases declined substantially, suggesting major improvements in diabetes screening and detection. Undiagnosed diabetes currently affects 1-2% of US adults; up to 90% of all cases are diagnosed.