Table 1. Number of Screenings and Diagnostic Tests Needed to Identify One Undiagnosed Diabetes Case

Category of Cases Age at Screening, y Prevalence of Undiagnosed Diabetes at Screeninga Sensitivity of CBG Screening Testb Specificity of CBG Screening Testb Number Needed to Screenc Number of Diagnostic Tests Needed, nd
Men with hypertension 35 0.05302 0.912 0.961 20.7 2.8
45 0.05302 0.938 0.905 20.1 3.8
55 0.08178 0.938 0.905 13.0 3.1
65 0.03403 0.938 0.905 31.3 4.9
75 0.04047 0.938 0.905 26.3 4.4
Women with hypertension 35 0.05302 0.74 0.973 25.5 2.7
45 0.05302 0.796 0.928 23.7 3.6
55 0.08178 0.796 0.928 15.4 3.0
65 0.03403 0.796 0.928 36.9 4.6
75 0.04047 0.796 0.928 31.0 4.1
Men without hypertension 35 0.05534 0.912 0.961 19.8 2.7
45 0.05534 0.938 0.905 19.3 3.7
55 0.05742 0.938 0.905 18.6 3.7
65 0.02606 0.938 0.905 40.9 5.8
75 0.04779 0.938 0.905 22.3 4.0
Women without hypertension 35 0.05534 0.74 0.973 24.4 2.6
45 0.05534 0.796 0.928 22.7 3.5
55 0.05742 0.796 0.928 21.9 3.5
65 0.02606 0.796 0.928 48.2 5.4
75 0.04779 0.796 0.928 26.3 3.8

aData from National Health and Nutrition Examination Survey III.1

bSensitivity and specificity of CBG screen based on test results of <120 mg/dL, >8 h postprandial time, as reported in Figure 2 of Rolka 2001.22

cNumber needed to screen to identify one undiagnosed diabetes case = [(prevalence of undiagnosed diabetes)(CBG sensitivity)].1

dNumber of diagnostic tests to identify one undiagnosed diabetes case = 2(1 true positive) + (number of false positive screening results undiagnosed diabetes case per undiagnosed diabetes case) = 2 + number needed to screen (1 - diabetes prevalence) (1 - CBG test specificity).

Note: CBG = capillary blood glucose.

Return to Document