CONTEXT: Conventional continuing medical education (CME) has been shown to have modest effects on doctor performance. New educational approaches based on the review of routine practices have brought better results. Little is known about factors that affect the outcomes of these approaches, especially in middle-income countries. This study aimed to investigate factors that influence the learning and quality of clinical performance in CME based on reflection upon experiences. METHODS: A questionnaire and a clinical performance test were administered to 165 general practitioners engaged in a CME programme in Brazil. The questionnaire assessed behaviours related to four input variables (individual reflection on practices, peer review of experiences, self-regulated learning and learning skills) and two mediating variables (identification of learning needs and engagement in learning activities, the latter consisting of self-study of scientific literature, consultations about patient problems, and attendance at courses). Structural equation modelling was used to test a hypothesised model of relationships between these variables and the outcome variable of clinical performance, measured by the clinical performance test. RESULTS: After minor adjustments, the hypothesised model fit the empirical data well. Individual reflection fostered identification of learning needs, but also directly positively influenced the quality of clinical performance. Peer review did not affect identification of learning needs, but directly positively affected clinical performance. Learning skills and self-regulation did not help in identifying learning needs, but self-regulation enhanced study of the scientific literature, the learning activity that most positively influenced clinical performance. Consultation with colleagues, the activity most frequently triggered by the identification of learning needs, did not affect performance, and attendance of courses had only limited effect. CONCLUSIONS: This study shed light on the factors that influence learning and performance improvement in continuing education based on the review of routine practices in middle-income settings. The findings support the importance of reflection on practices as an instrument for enhancing clinical performance.
CONTEXT: Conventional continuing medical education (CME) has been shown to have modest effects on doctor performance. New educational approaches based on the review of routine practices have brought better results. Little is known about factors that affect the outcomes of these approaches, especially in middle-income countries. This study aimed to investigate factors that influence the learning and quality of clinical performance in CME based on reflection upon experiences. METHODS: A questionnaire and a clinical performance test were administered to 165 general practitioners engaged in a CME programme in Brazil. The questionnaire assessed behaviours related to four input variables (individual reflection on practices, peer review of experiences, self-regulated learning and learning skills) and two mediating variables (identification of learning needs and engagement in learning activities, the latter consisting of self-study of scientific literature, consultations about patient problems, and attendance at courses). Structural equation modelling was used to test a hypothesised model of relationships between these variables and the outcome variable of clinical performance, measured by the clinical performance test. RESULTS: After minor adjustments, the hypothesised model fit the empirical data well. Individual reflection fostered identification of learning needs, but also directly positively influenced the quality of clinical performance. Peer review did not affect identification of learning needs, but directly positively affected clinical performance. Learning skills and self-regulation did not help in identifying learning needs, but self-regulation enhanced study of the scientific literature, the learning activity that most positively influenced clinical performance. Consultation with colleagues, the activity most frequently triggered by the identification of learning needs, did not affect performance, and attendance of courses had only limited effect. CONCLUSIONS: This study shed light on the factors that influence learning and performance improvement in continuing education based on the review of routine practices in middle-income settings. The findings support the importance of reflection on practices as an instrument for enhancing clinical performance.