Differential attainment in UK postgraduate medical examinations: examining the relationship between sociodemographic...

Differential attainment (DA), or differences in performance of groups (rather than individual differences), has been observed in a number of postgraduate medical specialty examinations used in UK medical training. Until now, much of the published research on DA has been limited in scope and size to one specialty, one examination or one type of assessment. This retrospective cohort study addressed this gap by examining the relationship between numerous sociodemographic differences and performance in almost all UK postgraduate medical examinations using a dataset of more than 180,000 examination attempts by UK and international medical graduates, adjusting for prior academic attainment.

This retrospective cohort study used the UK Medical Education Database (UKMED) to analyse the impact of a range of sociodemographic factors on performance in all UK postgraduate medical examinations aggregated into written and clinical exams. Pass/fail data at the first examination attempt were analysed for all candidates (UK medical school graduates (UKG) and those from non-UK schools (IMG)) sitting an examination between 2014 and 2020. Univariate analyses identified variables to carry forward into multivariate logistic regression models. Informed by previous research, all models were adjusted for prior academic attainment.

180,890 examination first-attempts were made by UKG and IMG candidates, and 121,745 (67.3%) passed at the first attempt. Multivariate regression models showed that place of primary qualification (UKG vs IMG), gender, age, ethnicity, religion, sexual orientation, disability status and working less than full-time were all statistically significant, independent predictors of examination outcomes for all examination candidates. Additionally, there were significant associations between socioeconomic backgrounds and performance for UKGs alone. The strongest independent predictors of failing written and clinical examinations were graduating from a non-UK medical school, having a minority ethnic background and having a registered disability.

This, the largest study of UK postgraduate medical examination outcomes, identified sociodemographic differences that were independently predictive of performance in written and clinical postgraduate medical examinations. Further analysis is now required to ascertain whether these group-level differences exist in each postgraduate medical examination, the majority or a select few.

Not applicable.