Detecting Dysglycemia Using the 2015 United States Preventive Services Task Force Screening Criteria: A Cohort Analysis of Community Health Center Patients

Abstract

Background: In 2015, the United States Preventive Services Task Force (USPSTF) recommended targeted screening for prediabetes and diabetes (dysglycemia) in adults who are aged 40 to 70 y old and overweight or obese. Given increasing prevalence of dysglycemia at younger ages and lower body weight, particularly among racial/ethnic minorities, we sought to determine whether the current screening criteria may fail to identify some high-risk population subgroups.

Methods and findings: We investigated the performance of the 2015 USPSTF screening recommendation in detecting dysglycemia among US community health center patients. A retrospective analysis of electronic health record (EHR) data from 50,515 adult primary care patients was conducted. Longitudinal EHR data were collected in six health centers in the Midwest and Southwest. Patients with a first office visit between 2008 and 2010 were identified and followed for up to 3 y through 2013. We excluded patients who had dysglycemia at baseline and those with fewer than two office visits during the follow-up period. The exposure of interest was eligibility for screening according to the 2015 USPSTF criteria. The primary outcome was development of dysglycemia during follow-up, determined by: (1) laboratory results (fasting/2-h postload/random glucose ≥ 100/140/200 mg/dL [5.55/7.77/11.10 mmol/L] or hemoglobin A1C ≥ 5.7% [39 mmol/mol]); (2) diagnosis codes for prediabetes or type 2 diabetes; or (3) antidiabetic medication order. At baseline, 18,846 (37.3%) participants were aged ≥40 y, 33,537 (66.4%) were overweight or obese, and 39,061 (77.3%) were racial/ethnic minorities (34.6% Black, 33.9% Hispanic/Latino, and 8.7% Other). Overall, 29,946 (59.3%) patients had a glycemic test within 3 y of follow-up, and 8,478 of them developed dysglycemia. Only 12,679 (25.1%) patients were eligible for screening according to the 2015 USPSTF criteria, which demonstrated the following sensitivity and specificity (95% CI): 45.0% (43.9%-46.1%) and 71.9% (71.3%-72.5%), respectively. Racial/ethnic minorities were significantly less likely to be eligible for screening yet had higher odds of developing dysglycemia than whites (odds ratio [95% CI]: Blacks 1.24 [1.09-1.40]; Hispanics 1.46 [1.30-1.64]; and Other 1.33 [1.16-1.54]). In addition, the screening criteria had lower sensitivity in all racial/ethnic minority groups compared to whites. Limitations of this study include the ascertainment of dysglycemia only among patients with available test results and findings that may not be generalizable at the population level.

Conclusions: Targeted diabetes screening based on new USPSTF criteria may detect approximately half of adult community health center patients with undiagnosed dysglycemia and proportionately fewer racial/ethnic minorities than whites. Future research is needed to estimate the performance of these screening criteria in population-based samples.

Fig 1. Flow diagram.

Fig 2. Development of clinically detected dysglycemia within 3 y among patients who received screening by race/ethnicity and USPSTF eligibility (n = 29,946).

P-values were derived from pairwise comparisons of USPSTF eligibility among racial/ethnic groups using Chi-square tests. Numbers displayed within the bars represent the probability of being eligible for screening among those who developed dysglycemia (i.e., sensitivity). USPSTF, US Preventive Services Task Force.