ADA and USPSTF recommended diabetes screening starting at age 35 recently. We estimated the percentage of adults with undiagnosed prediabetes or diabetes who can be detected by USPSTF and ADA guidelines in all and by racial/ethnicity groups in the All of Us cohort. We then identified predictors of incident T2D. Among 50,705 eligible participants, we identified 20,150 with undiagnosed prediabetes and 1,860 with undiagnosed diabetes at enrollment, >99% of whom met the ADA 2022 screening criteria, which outperformed USPTSF ones (57.4% and 67.4%). All results were similar to findings from a recent NHANES study. Substantial racial disparities were identified in stratified analyses. The percentage of undiagnosed prediabetes was 37.7%, 50.8% and 38.2% in White, Black and Hispanic populations, respectively and it was 2.6%, 7.8% and 3.8%, respectively, for undiagnosed diabetes (all p<2e-16). However, the ADA 2022 guideline can detect almost all (>98.8%) the undiagnosed, particularly in the understudied populations. During the 4 year follow up, 23% of the undiagnosed diabetes were subsequently diagnosed, with race, income, education, and insurance being the strongest predictors. In summary, the new ADA guideline can effectively facilitate early detection of diabetes across all populations, but barriers still exist to effectively screen and diagnose equitably.


C.Zeng: None. T.Cassini: None. T.M.Ferrara: None. D.Schlueter: None. T.C.Tran: None. J.Denny: None.


National Institutes of Health (HG200417)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at