Association Between Licensing Examination Scores and Resource Use and Quality of Care in Primary Care Practice

CONTEXT: Clinical competence is a determinant of the quality of care delivered, and may be associated with use of health care resources by primary care physicians. Clinical competence is assumed to be assessed by licensing examinations, yet there is a paucity of information on whether scores achieved predict subsequent practice. OBJECTIVE: To determine if licensing examination scores were associated with selected aspects of quality of care and resource use in initial primary care practice. DESIGN: Prospective cohort study of recently licensed family physicians, followed up for the first 18 months of practice. SETTING: The Quebec health care system. PARTICIPANTS: A total of 614 family physicians who passed the licensing examination between 1991 and 1993 and entered fee-for-service practice in Quebec. MAIN OUTCOME MEASURES: All patients seen by physicians were identified by the universal health insurance board and all health services provided to these patients were retrieved for the 18 months prior to (baseline) and after (follow-up) the physicians' entry into practice. Medical service and prescription claims files were used to measure rates of resource use (specialty consultation, symptom-relief prescribing compared with disease-specific prescribing) and quality of care (inappropriate prescribing, mammography screening). Baseline data were used to adjust for differences in practice population. RESULTS: Study physicians saw a total of 1116389 patients, of whom 113535 (10.2%) were elderly and 83391 (7.5%) were women aged 50 to 69 years. Physicians with higher licensing examination scores referred more of their patients for consultation (3.8/1000 patients per SD increase in score; 95% confidence interval CI, 1.2-7.0; P = .005), prescribed to elderly patients fewer inappropriate medications (-2.7/1000 patients per SD increase in score; 95% CI, -4.8 to -0.7; P=.009) and more disease-specific medications relative to symptom-relief medications (3.9/1000 patients per SD increase in score; 95% CI, 0.3 to 7.4; P= .03), and referred more women aged 50 to 69 years (6.6/1000 patients per SD increase in score; 95% CI, 1.2-11.9; P = .02) for mammography screening. If patients of physicians with the lowest scores had experienced the same rates of consultation, prescribing, and screening as patients of physicians with the highest scores, an additional 3027 patients would have been referred, 179 fewer elderly patients would have been prescribed symptom-relief medication, 912 more elderly patients would have been prescribed disease-specific medication, 189 fewer patients would have received inappropriate medication, and 121 more women would have received mammography screening. CONCLUSIONS: Licensing examination scores are significant predictors of consultation, prescribing, and mammography screening rates in initial primary care practice.