Predicting fitness to practise events in international medical graduates who registered as UK doctors via the Professional and Linguistic Assessments Board (PLAB) system: a national cohort study

Abstract

Background: International medical graduates working in the UK are more likely to be censured in relation to fitness to practise compared to home graduates. Performance on the General Medical Council's (GMC's) Professional and Linguistic Assessments Board (PLAB) tests and English fluency have previously been shown to predict later educational performance in this group of doctors. It is unknown whether the PLAB system is also a valid predictor of unprofessional behaviour and malpractice. The findings would have implications for regulatory policy.

Methods: This was an observational study linking data relating to fitness to practise events (referral or censure), PLAB performance, demographic variables and English language competence, as evaluated via the International English Language Test System (IELTS). Data from 27,330 international medical graduates registered with the GMC were analysed, including 210 doctors who had been sanctioned in relation to at least one fitness to practise issue. The main outcome was risk of eventual censure (including a warning).

Results: The significant univariable educational predictors of eventual censure (versus no censures or referrals) were lower PLAB part 1 (hazard ratio [HR], 0.99; 95% confidence interval, 0.98 to 1.00) and part 2 scores (HR, 0.94; 0.91 to 0.97) at first sitting, multiple attempts at both parts of the PLAB, lower IELTS reading (HR, 0.79; 0.65 to 0.94) and listening scores (HR, 0.76; 0.62 to 0.93) and higher IELTS speaking scores (HR, 1.28; 1.04 to 1.57). Multiple resits at either part of the PLAB and higher IELTS speaking score (HR, 1.49; 1.20 to 1.84) were also independent predictors of censure. We estimated that the proposed limit of four attempts at both parts of the PLAB would reduce the risk in this entire group by only approximately two censures per 5 years in this group of doctors.

Conclusions: Making the PLAB, or any replacement assessment, more stringent and raising the required standards of English reading and listening may result in fewer fitness to practice events in international medical graduates. However, the number of PLAB resits permitted would have to be further capped to meaningfully impact the risk of sanctions in this group of doctors.

Keywords: Fitness to practise; International medical graduates; Medical regulation; Professionalism.

Fig. 1

Graph showing the risk of being eventually censured for fitness to practise issues in a sample of international medical graduates in relation to seven predictors. The coefficients depicted are hazard ratios derived from univariable survival analyses, with associated 95% confidence intervals. Blue triangles represent the estimated risk of being censured (versus not being censured, referred or referred without eventual censure n = 27,330). Red squares represent the risk of being eventually censured only in the group referred (n = 1168). Green circles represent the risk of being censured purely in relation to non-clinical (i.e. professionalism) concerns versus censure, which involves some clinical component (with or without professionalism issues n = 210)

Fig. 2

Graph showing the risk of being eventually censured for fitness to practise issues in a sample of international medical graduates in relation to the number of attempts at both parts of the Professional and Linguistic Assessments Board exam. The coefficients depicted are hazard ratios derived from univariable survival analyses, with associated 95% confidence intervals. Blue triangles represent the estimated risk of being censured (versus not being censured, referred or referred without eventual censure n = 27,330). Red squares represent the risk of being eventually censured only in the group referred (n = 1168). Green circles represent the risk of being censured purely in relation to non-clinical (i.e. professionalism) concerns versus censure, which involves some clinical component (with or without professionalism issues n = 210)