Abstract
Background Tests of competence are written and clinical assessments taken by doctors under investigation by the General Medical Council (GMC) who have significant performance concerns. Male doctors on average perform more poorly in clinical assessments than female doctors, and are more likely to be sanctioned. It is unclear why.
Aim To examine sex differences in the tests of competence assessment scores of GPs under investigation by the GMC, compared with GPs not under investigation, and whether scores mediate any relationship between sex and sanction likelihood.
Design and setting Retrospective cohort study of GPs’ administrative tests of competence data.
Method Analysis of variance was undertaken to compare written and clinical tests of competence performance by sex and GP group (under investigation versus volunteers). Path analysis was conducted to explore the relationship between sex, written and clinical tests of competence performance, and investigation outcome.
Results On the written test, female GPs under investigation outperformed male GPs under investigation (Cohen’s d = 0.28, P = 0.01); there was no sex difference in the volunteer group (Cohen’s d = 0.02, P = 0.93). On the clinical assessment, female GPs outperformed male GPs in both groups (Cohen’s d = 0.61, P<0.0001). A higher clinical score predicted remaining on the UK medical register without a warning or sanction, with no independent effect of sex controlling for assessment performance.
Conclusion Female GPs outperform male GPs on clinical assessments, even among GPs with generally very poor performance. Male GPs under investigation may have particularly poor knowledge. Further research is required to understand potential sex differences in doctors who take tests of competence and how these impact on sex differences in investigation outcomes.
INTRODUCTION
Erasure or suspension from the UK medical register can result in loss of career and income; lesser sanctions (conditions and undertakings) and warnings can also limit doctors’ careers and hamper career progression.1,2 Male doctors are nearly three times more likely to be sanctioned than female doctors.3 The reasons underlying this difference are not clear.
Performance in academic assessments is associated with future disciplinary action,4,5 suggesting that sanctioned doctors may have deficient medical knowledge and clinical skills. Research has also demonstrated that female doctors perform better academically,6 particularly in clinical7–10 rather than written assessments.7,11,12
Role of the General Medical Council
The role of the General Medical Council (GMC) is to ensure proper standards in the practice of medicine in the UK, thereby protecting, promoting, and maintaining the health and safety of the patient population and the community as a whole.13 Under the Medical Act 1983, the GMC holds the powers to take action against a doctor’s registration if the doctor’s fitness to practise is impaired for one or more of a number of reasons, including deficient professional performance.14,15
The GMC triages information received about complaints that require investigation.16 If during an investigation a doctor’s professional performance is called into question, they may be required to undergo a performance assessment consisting of two main parts: a peer review and test of competence.17 The GMC’s decisions about the investigated doctor’s fitness to practise, and the outcome of the investigation, are based on the doctor’s performance in both parts.
At any point during an investigation it is possible for the doctor to apply to be removed from the medical register (voluntary erasure), which if granted means the doctor does not have to complete the investigation.18 However, should the doctor wish to return to the register and practise medicine, they would have to demonstrate that they are fit to practise.
Tests of competence are designed to identify gaps in the knowledge and clinical skills of doctors under investigation for poor performance.13 They comprise a written knowledge test (KT) and an objective structured clinical examination (OSCE). In general practice there is also simulated surgery (SS), which simulates a typical GP surgery, and assesses the doctor’s clinical, management, and communication skills.19
To ensure tests of competence are fair and fit for purpose, they are ‘piloted’ with doctors with no known fitness to practise concerns and who volunteer and are paid to complete them. There is no pass mark: the performance of the doctor under investigation is compared with reference groups of volunteers who have completed the same questions, and this contributes to the findings of the GMC investigation.20 Further information about the GMC’s tests of competence can be found in some of the references.13,21,22
How this fits in
Male doctors are more likely to face disciplinary action and have their medical registration acted on than female doctors. Doctors who are referred to the General Medical Council (GMC) because of concerns regarding their performance may be required to complete a set of assessments to assess their medical knowledge and clinical skills. This study has shown that, in this highly selected population of doctors, female doctors perform better than male doctors at the written and clinical assessments, and that performance at the clinical assessment predicts whether a GP being investigated by the GMC will remain on the medical register without receiving a warning or sanction. Sex has no independent effect on the outcome. Further research is required to better understand the reasons why female GPs perform better at the clinical assessment than male GPs.
Study aims
The authors hypothesised that male doctors on average have greater gaps in their medical knowledge and clinical skills than female doctors, which contributes to sex differences in disciplinary action. This study aimed to explore:
the presence and magnitude of sex differences in performance on the written and clinical components of the GMC’s tests of competence;
whether tests of competence sex differences are comparable for doctors currently under investigation by the GMC and doctors not currently under investigation and without restrictions on their medical registration, who had volunteered to complete the tests of competence; and
whether any relationship between sex and likelihood of disciplinary action is mediated by tests of competence performance.
METHOD
Study design, setting, and source of data
This was a retrospective cohort study of GPs in the UK involving GPs undergoing tests of competence as part of a GMC fitness to practise investigation, and GPs not currently under investigation and with no restrictions on their medical registration who volunteered to take tests of competence. Data were obtained from the GMC and the Research Department of Medical Education at University College London.
Outcome of the GMC fitness to practise investigation
This study examined whether the outcome of the GMC investigation was related to GP sex, written assessment score and SS score, and whether tests of competence performance mediated the relationship between sex and outcome.
Investigation outcome was collapsed into a nominal variable: no sanction imposed, warning or sanction imposed, and no longer registered on the List of Registered Medical Practitioners (Box 1). Warnings and sanctions were combined because warnings are recorded on the List of Registered Medical Practitioners for a period of time and can affect a doctor’s career progression.1
Box 1. Definition of outcome types after General Medical Council fitness to practise investigation
| Outcome type | Definition |
|---|
| No sanction imposed | No impairment found during the investigation and no restrictions imposed on the doctor’s medical registration |
| Warning/sanction imposed | Warning issued, but no restrictions placed on the doctor’s medical registration OR Sanction imposed resulting in a restriction on the doctor’s medical registration. Sanctions include undertakings, conditions, suspension from the LRMP, and erasure from the LRMP3,23 |
| No longer registered on the LRMP | Administrative erasure from LRMP, voluntary erasure from LRMP, or doctor deceased23 |
Population
The study included all GPs between 2008 and 2013 under investigation by the GMC and required to complete tests of competence between 2008 and 2014. The study focused on general practice because it is the largest medical specialty and attracts the most complaints.23
The comparison group comprised GPs not currently under investigation and with no restrictions on their registration, who voluntarily completed the KT and/or SS assessment as part of the tests of competence pilots.
There were two versions of the SS test in circulation during the timeframe of this study. Most doctors completed version 1, whereas version 2 was used primarily to reassess doctors. The OSCE was not used for this study because the OSCE and simulated surgery have been found to have significant overlap.19
Statistical methods
The KT scores were z-transformed for both GP groups (under investigation; volunteers). Analysis of variance was performed to test the performance at the KT and SS test by doctors’ sex and GP group, checking for interaction effects. Multinomial and binomial logistic regression models were built, and then a path analysis model was built using multiple regression, with each variable being set as the dependent variable in turn. Paths were included in the model if they were significant at P<0.05. Statistical analyses used Stata (version 12/SE). The STROBE statement24 guided the reporting.
DISCUSSION
Summary
Among GPs under investigation by the GMC, females outperformed males on the written and clinical components of the tests of competence. Among GPs not under investigation, females outperformed males on the SS test only. Among GPs under investigation, low SS score increased the likelihood of having a warning or sanction imposed, but male GPs were no more likely than female GPs to have a warning or sanction imposed despite lower test scores. Warnings or sanctions are imposed based on a range of evidence of which the test of competence is one component; the outcome ‘warning or sanction imposed’ does not demonstrate the severity, which may differ by test of competence score levels.
Strengths and limitations
A strength of this study was the inclusion of nearly all GPs investigated by the GMC for fitness to practise concerns who had completed a test of competence. The comparator group of doctors not under investigation enabled the examination of sex differences at different levels of performance. A weakness of the study is that the volunteer GPs are a self-selecting group who may not be representative of the overall GP population. It is also not possible to generalise the findings to doctors in other specialties.
Unmeasured factors may have influenced the findings. Data from other specialties show that volunteer doctors differ from doctors under investigation in terms of sex, ethnicity, world region of primary medical qualification, and seniority:20 factors known to influence performance at clinical exams.25–27 Demographic data (aside from sex) were missing for many volunteer GPs so it was not possible to adjust for these confounders.
In the group of investigated GPs it would have been interesting to explore whether ethnicity was associated with performance or outcome, but unfortunately ethnicity data were missing for nearly one-third of these GPs. It would also have been of interest to explore the relationship between number of years since primary medical qualification (or age), sex, and performance to see whether recent changes in the sex distribution of GPs,28,29 and the role and format of the examinations required to gain Membership of the Royal College of General Practitioners (MRCGP),30 have influenced the performance of GPs. However, because of the small numbers of female GPs and GPs who had been qualified for 10 years or less, it was not possible to explore this further. Data were not included on the OSCE test of competence, although the SS score is a better predictor of investigation outcome.19
Comparison with existing literature
Female doctors generally outperform male doctors in postgraduate medical exams in general practice,8,10,25,31 and other specialties.7,9,32 The current study has shown that this holds even in a group with overall very poor performance; however, there was no sex difference in the written test for GPs not under investigation.
Previous research has shown that male doctors receive more sanctions after controlling for time since primary medical qualification, non-domestic primary medical qualification, and specialty.3 However, in the current study there was no evidence to suggest a sex difference in warning/sanction rates among this highly selected group of GPs who completed a test of competence as part of an investigation. This suggests that the final decision on warnings or sanctions showed no sex bias beyond the differences in test performance.
Implications for research and practice
The reasons for sex differences are uncertain. Further research is required to examine whether male and female doctors differ in ways the current study was unable to measure and which influenced their performance.
Lack of insight — being unaware of and not addressing deficiencies — is common among investigated doctors.33 Previous research found that male volunteer doctors tended to overestimate their written and clinical test of competence scores, suggesting less insight.34 If these sex differences are present in doctors under investigation, more poorly performing female doctors may remove themselves from the medical register before taking a test of competence. It would be interesting to investigate further those GPs who are no longer registered following an investigation into their fitness to practise, particularly those GPs who were voluntarily erased from the UK medical register, not only in terms of demographics and performance, but also the reasons behind their decision to apply for voluntary erasure.
Female doctors in general have a more patient-centred approach and ask more psychosocial questions than male doctors, which stimulates more patient disclosure.35 Female doctors, including those under investigation, may therefore learn more information from patients, and perform better at the clinical assessment. Females on average have been found to score higher on dutifulness (a facet of conscientiousness),36 which is a predictor of performance.37–40 Female doctors may therefore be higher on personality traits that lead to them maintaining their skills and knowledge, and performing better at assessments. Female doctors have also been shown to have higher person-related values.41–43 It has been shown that performance in a clinical setting is predicted by the person-related values held by a doctor.42 It is plausible that female GPs’ higher performance is therefore due to differences in skills and attitudes, as well as knowledge.
Another factor that may have influenced doctors’ examination performance is dyslexia or another specific learning difficulty. Doctors with a specific learning difficulty may also face extra challenges in the effective performance of their duties, especially if unrecognised or undisclosed.44 It is uncertain whether there are sex differences in specific learning difficulties such as dyslexia among doctors, but, given that nearly 2% of medical students have dyslexia,45 it would be important to explore not only how specific learning difficulties may affect learning and performance, but also how to better identify and support those doctors with specific learning difficulties.
Organisational factors can also affect performance;37 for example, professional isolation from peers and colleagues can limit opportunities for feedback and development.33 A possible hypothesis is that male GPs may be more at risk of professional isolation. Evidence suggests that male GPs are more likely to work in a single-handed practice than female GPs.46 There is also evidence to suggest that professional isolation is more prevalent in rural areas.47
It is unclear why it is that, among volunteer GPs, female doctors performed better on the clinical but not the written assessment. It may be that the male doctors in the self-selected group of volunteer GPs had better medical knowledge than the average male GP population.
Future work should explore whether reasons for referral to the GMC (allegation type) differ between the sexes and whether certain allegation types are associated with a higher risk of sanctions.
This study has implications for support for GPs undergoing a fitness to practise investigation and future work could explore how the medical profession could better support those doctors undergoing an investigation.
Notes
Funding
Katherine Woolf is funded by a National Institute for Health Research (NIHR) career development fellowship. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, or the Department of Health.
Ethical approval
This study is part of a research project that has received ethical approval from the University College London (UCL) Research Ethics Committee (Project ID: 5025/001).
Provenance
Freely submitted; externally peer reviewed.
Competing interests
Emily Unwin is supported by a UCL Impact studentship. Katherine Woolf is supported by an NIHR Fellowship and is a Higher Education Funding Council for England (HEFCE)-supported staff member at UCL Medical School. She received funding from the General Medical Council (GMC) for two unrelated research studies during the course of the current study. She is educational advisor to the Membership of the Royal Colleges of Physicians of the United Kingdom. Jane Dacre was the President of the Royal College of Physicians during the course of the current study, and was a GMC council member and chaired the GMC Education and Training Committee (2008–2012). She was the principal investigator for the Fitness to Practise contract with the GMC. Henry WW Potts is an HEFCE-supported staff member at UCL. He has received research funding from the GMC.
Discuss this article
Contribute and read comments about this article: bjgp.org/letters