Key points are not available for this paper at this time.
CONTEXT: Continuous assessment of individual performance of doctors is crucial for life-long learning and quality of care. Policy-makers and health educators should have good insights into the strengths and weaknesses of the methods available. The aim of this study was to systematically evaluate the feasibility of methods, the psychometric properties of instruments that are especially important for summative assessments, and the effectiveness of methods serving formative assessments used in routine practise to assess the performance of individual doctors. METHODS: We searched the MEDLINE (1966-January 2006), PsychINFO (1972-January 2006), CINAHL (1982-January 2006), EMBASE (1980-January 2006) and Cochrane (1966-2006) databases for English language articles, and supplemented this with a hand-search of reference lists of relevant studies and bibliographies of review articles. Studies that aimed to assess the performance of individual doctors in routine practise were included. Two reviewers independently abstracted data regarding study design, setting and findings related to reliability, validity, feasibility and effectiveness using a standard data abstraction form. RESULTS: A total of 64 articles met our inclusion criteria. We observed 6 different methods of evaluating performance: simulated patients; video observation; direct observation; peer assessment; audit of medical records, and portfolio or appraisal. Peer assessment is the most feasible method in terms of costs and time. Little psychometric assessment of the instruments has been undertaken so far. Effectiveness of formative assessments is poorly studied. All systems but 2 rely on a single method to assess performance. DISCUSSION: There is substantial potential to assess performance of doctors in routine practise. The longterm impact and effectiveness of formative performance assessments on education and quality of care remains hardly known. Future research designs need to pay special attention to unmasking effectiveness in terms of performance improvement.
Overeem et al. (Thu,) studied this question.