Fig 1.
Study design inclusion and exclusion criteria.
Table 1.
Attrition in all databases among A) cases and B) controls.
Table 2.
Model relatedness between the original study and each of the validation databases.
Fig 2.
Receiver operating characteristic (ROC) curve for IQVIA Medical Research Data (IMRD) in A) UK B) France C) Germany D) Belgium E) Australia.
Table 3.
Measures of predictive performance of the model at ~80% sensitivity.