Examination of nonresponse bias in a major residency follow‐up study