Efficient Cutoff Points for Three Screening Tests for Detecting Undiagnosed Diabetes and Pre-Diabetes

Abstract

OBJECTIVE—Opportunistic screening for undiagnosed type 2 diabetes and pre-diabetes (either impaired glucose tolerance or impaired fasting glucose) is recommended by the American Diabetes Association. The aim of this study was to determine efficient cutoff points for three screening tests for detecting undiagnosed diabetes alone or both undiagnosed diabetes and pre-diabetes. RESEARCH DESIGN AND METHODS— We estimated the number of individuals with undiagnosed diabetes alone or with both undiagnosed diabetes and pre-diabetes that could be detected by using different cutoff points for each screening test as the product of the prevalence of each condition, the sensitivity of the tests at each cutoff point for identifying each condition, and the number of individuals who would be eligible for screening in the U.S. We estimated the total cost of opportunistic screening by multiplying the cost for screening one person by the number of individuals screened. RESULTS—The most efficient cutoff points for both detecting pre-diabetes and undiagnosed diabetes (100 mg/dl for the fasting plasma glucose test, 5.0% for the HbA1c test, and 100 mg/dl for the random capillary blood glucose test) were less than those for detecting undiagnosed diabetes alone (110 mg/dl for the fasting plasma glucose test, 5.7% for the HbA1c test, and 120 mg/dl for the random capillary blood glucose test). CONCLUSIONS—A lower cutoff value should be used when screening for pre-diabetes and undiagnosed diabetes together than when screening for undiagnosed diabetes alone.