Self-reported confidence in prescribing skills correlates poorly with assessed competence in fourth-year medical students

The objective of this study was to investigate the relationship between students' self-reported confidence and their objectively assessed competence in prescribing. We assessed the competence in several prescribing skills of 403 fourth-year medical students at the VU University Medical Center, the Netherlands, in a formative simulated examination on a 10-point scale (1 = very low; 10 = very high). Afterwards, the students were asked to rate their confidence in performing each of the prescribing skills on a 5-point Likert scale (1 = very unsure; 5 = very confident). Their assessments were then compared with their self-confidence ratings. Students' overall prescribing performance was adequate (7.0 ± 0.8), but they lacked confidence in 2 essential prescribing skills. Overall, there was a weak positive correlation (r = 0.2, P < .01, 95%CI 0.1-0.3) between reported confidence and actual competence. Therefore, this study suggests that self-reported confidence is not an accurate measure of prescribing competence, and that students lack insight into their own strengths and weaknesses in prescribing. Future studies should focus on developing validated and reliable instruments so that students can assess their prescribing skills.