Predictive validity of osteopathic medical licensing examinations for osteopathic medical knowledge measured by graduate written examinations.

The purpose of this study was to evaluate the predictive validity of osteopathic medical licensing examinations for osteopathic medical knowledge measured by graduate written medical examinations. Performances on the three osteopathic initial licensing examinations, the three osteopathic internal medicine in-service examinations, and the osteopathic internal medicine board certification examinations were analyzed for a cohort of the most recent osteopathic internal medicine board certification examination candidates (N = 82). Multiple regressions were performed for the predictive value of licensing examination scores for the late examination scores. Logistic regressions were used for the prediction of pass/fail status on the licensing examinations for that on the board certification. A longitudinal performance profile was constructed to assess the rank changes in decile on the examinations at different times. All correlation coefficients between the licensing examinations and other examinations were significant and higher than .70. The licensing examinations together predicted at least 60% of the variance of any of the other examination scores. The pass/fail status on the licensing examinations predicted 89% of the pass/fail status on the certification examination. Decile ranks since the first licensing examination were consistent and stable over at least 5 years. The osteopathic licensing examinations had high predictive validity for the late written osteopathic internal medicine examinations. Generalization of the findings to other disciplines needs a caution, as a specialty bias may exist for these types of studies.