paperKB
coga / coga-kb
Processing
Help
Sign in

Chunk #18 — Results — Discrimination using phenotype based risk scores

Source
Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort study.
Embedded
yes

Text

The odds ratio for developing diabetes was 8.4 (95% confidence interval 5.1 to 13.9) for people in the top fifth of the Cambridge risk score distribution compared with those in the bottom fifth, and the area under the receiver operating characteristics curve was 0.72 (95% confidence interval 0.69 to 0.76) at phase 7 (table 1). The odds ratio for type 2 diabetes in participants in the top fifth of the Framingham offspring risk score distribution compared with those in the bottom fifth was 18.2 (9.8 to 33.8), with a significantly greater area under the receiver operating characteristics curve of 0.78 (0.75 to 0.82) (P=0.01) (table 1 and fig 1). Detection rates (sensitivity) for a 5% false positive rate were 19.7 (95% confidence interval 14.1 to 25.3) for the Cambridge risk score (cut-off point 0.34) and 30.6 (24.1 to 37.1) for the Framingham offspring risk score (cut-off point 0.08). The respective detection rates for a 10% false positive rate were 34.2 (27.5 to 40.9; cut-off point 0.26) and 43.0 (36.0 to 50.0; cut-off point 0.05) (table 1).