Objectives This study aimed to determine if national licensing examinations that measure medical knowledge (QE1) and clinical skills (QE2) predict the quality of care delivered by doctors in future practice.
Methods Cohorts of doctors who took the Medical Council of Canada Qualifying Examinations Part I (QE1) and Part II (QE2) between 1993 and 1996 and subsequently entered practice in Ontario, Canada (n = 2420) were followed for their first 7–10 years in practice. The 208 of these doctors who were randomly selected for peer assessment of quality of care were studied. Main outcome measures included quality of care (acceptable/unacceptable) as assessed by doctor peer-examiners using a structured chart review and interview. Multivariate logistic regression was used to determine if qualifying examination scores predicted the outcome of the peer assessments while controlling for age, sex, training and specialty, and if the addition of the QE2 scores provided additional prediction of quality of care.
Results Fifteen (7.2%) of the 208 doctors assessed were considered to provide unacceptable quality of care. Doctors in the bottom quartile of QE1 scores had a greater than three-fold increase in the risk of an unacceptable quality-of-care assessment outcome (odds ratio [OR] 3.41, 95% confidence interval [CI] 1.14–10.22). Doctors in the bottom quartile of QE2 scores were also at higher risk of being assessed as providing unacceptable quality of care (OR 4.24, 95% CI 1.32–13.61). However, QE2 results provided no significant improvement in predicting peer assessment results over QE1 results (likelihood ratio test: χ2 = 3.21, P-value(1 d.f.) = 0.07).
Conclusions Doctor scores on qualifying examinations are significant predictors of quality-of-care problems based on regulatory, practice-based peer assessment.