The GP tests of competence assessment: which part best predicts fitness to practise decisions?

Abstract Background The General Medical Council (GMC) conducts Tests of Competence (ToC) for doctors referred for Fitness to Practise (FtP) issues. GPs take a single best answer knowledge test, an Objective Structured Clinical Examination (OSCE), and a Simulated Surgery (SimSurg) assessment which is a simulated GP consultation. The aim of this study was to examine the similarities between OSCEs and SimSurg to determine whether each assessment contributed something unique to GP ToCs. Methods A mixed methods approach was used. Data were collated on 153 GPs who were required to undertake a ToC as a part of being investigated for FtP issues between February 2010 and October 2016. Using correlation analysis, we examined to what degree performance on the knowledge test, OSCE, and SimSurg related to case examiner recommendations and FtP outcomes, including the unique predictive power of these three assessments. The outcome measures were case examiner recommendations (i) not fit to practise; ii) fit to practise on a limited basis; or iii) fit to practise) as well as FtP outcomes (i) erased/removed from the register; ii) having restrictions/conditions; or iii) be in good standing). For the qualitative component, 45 GP assessors were asked to rate whether they assess the same competencies and which assessment provides better feedback about candidates. Results There was significant overlap between OSCEs and SimSurg, p < 0.001. SimSurg had additional predictive power in the presence of OSCEs and the knowledge test (p = 0.030) in distinguishing doctors from different FtP categories, while OSCEs did not (p = 0.080). Both the OSCEs (p = 0.004) and SimSurg (p < 0.001) had significant negative correlations with case examiner recommendations when accounting for the effects of the other two assessments. Inductive thematic analysis of the responses to the questionnaire showed that assessors perceived OSCEs to be better suited to target specific knowledge and skills. SimSurg was thought to produce a more global picture as the scenarios more accurately portray a patient consultation. Conclusion While all three assessments are strong predictors of both case examiner recommendations and FtP outcomes, our findings suggest that the efficiency of GP ToCs can be improved by removing some of this overlapping content