Skip to main content
. 2020 Sep 22;106(1):153–167. doi: 10.1210/clinem/dgaa675

Table 5.

Genetic validation of polycystic ovary syndrome (PCOS) diagnosis. Logistic regressions were performed between PCOS case status and PCOS polygenic risk scores (PCOSPRS). Regressions were adjusted for the first 10 principal components, and each algorithm data set was case matched for age and race

Algorithm Cases Controls Estimate SE P OR LCI UCI PRS R2
European
Coded broad 1877 5052 0.06 0.03 .06 1.06 1.00 1.13 7.00E-04
Coded strict 366 5052 0.17 0.06 .01 1.18 1.05 1.33 0.004
Keyword broad 332 5052 0.25 0.06 6.97E-05 1.28 1.14 1.45 0.008a
Keyword strict 245 5052 0.20 0.07 .01 1.22 1.06 1.41 0.005
African
Coded broad 908 1663 0.11 0.06 .07 1.12 0.99 1.26 0.002
Coded strict 161 1663 0.34 0.12 .01 1.40 1.11 1.78 0.01
Keyword broad 129 1663 0.30 0.13 .03 1.34 1.03 1.75 0.007
Keyword strict 98 1663 0.40 0.15 .01 1.49 1.11 2.01 0.011

Abbreviations: LCI, lower confidence interval; OR, odds ratio; PRS, polygenic risk score; R2, pseudo-R2; UCI, upper CI.

a P < .002.