Article Text

Download PDFPDF

An analysis of the performance of UK medical graduates in the MRCOG Part 1 and Part 2 written examinations
  1. S Rushd1,
  2. A B Landau2,
  3. J A Khan3,
  4. V Allgar4,
  5. S W Lindow1
  1. 1Women and Children' Hospital, Hull Royal Infirmary, Hull, UK
  2. 2Examination Department, Royal College of Obstetricians & Gynaecologists, London, UK
  3. 3Academic Vascular Surgery, Hull Royal Infirmary, Hull, UK
  4. 4Hull York Medical School, Hull Royal Infirmary, Hull, UK
  1. Correspondence to Sophia Umber Rushd, Clinical Research Fellow in Urogynaecology, Women and Childrens' Hospital, Hull Royal Infirmary, Anlaby Road, Hull HU3 2JZ, UK; sofia_rushd{at}yahoo.co.uk

Abstract

Background There is a lack of evidence on whether graduates from different medical schools perform differently in postgraduate examinations.

Objective To evaluate the variations in performance of UK medical graduates in Member of the Royal College of Obstetricians and Gynaecologists (MRCOG) examination.

Methods A retrospective analysis of performance of 1335 doctors graduating in UK medical schools who entered the Part 1 MRCOG and 822 doctors taking the Part 2 MRCOG written examination for the first time between 1998 and 2008. The main outcome measures were to evaluate medical school effects, gender effects and academic performance effect.

Results Graduates of UK medical schools performed differently in the Part 1 and Part 2 written MRCOG examination. The graduates of Oxford (pass rate 82.6%), Cambridge (75%), Bristol (59.3%) and Edinburgh (57.5%) performed significantly better and the graduates of Liverpool (26.8%), Southampton (21.8%) and Wales (18.2%) performed significantly worse than the remaining cohort in the Part 1 examination. The candidates of Newcastle (88.9%), Oxford (82.4%), Cambridge (81%) and Edinburgh (78.2%) performed significantly better and the graduates of Glasgow (49.2%) and Leicester (36.2%) have significantly underperformed compared with the remaining cohort in Part 2 written examination. There was no difference in the success rates of male (47.5%) and female (42.0%) candidates in the Part 1; however, female candidates had a significantly better success rate in the Part 2 written examination than male candidates (65.6% vs 52.9%).

Conclusion These results show that there is variation in performance among the graduates from different medical schools in the Part 1 and Part 2 MRCOG written examination.

  • MRCOG examination
  • performance variation in MRCOG
  • MRCOG results evaluation
  • urogynaecology
  • thoracic medicine
  • statistics and research methods
  • education and training (see medical education and training)
  • medical education and training
  • internal medicine
View Full Text

Statistics from Altmetric.com

Background

During the last 2 decades, several reforms have been made in undergraduate education at most UK medical schools.1 2 Undergraduate and postgraduate medical education are now considered as a continuum in the training of a specialist in all fields of medicine, although little consideration is given to the effect that changes in style of learning, the curriculum and objectives of undergraduate education might have on postgraduate performance.3 4 There are a few studies comparing the outcomes of different medical curricula beyond the first year after graduation. A study in 1993 showed the pass rate for different medical schools across all UK for the Membership examination of the Royal College of General Practitioners.5 Recently, two more studies have evaluated and shown differences between the pass rates of graduates from UK medical schools, one for the Member of the Royal College of Physicians (MRCP (UK))4 and the other for the Fellowship of the Royal College of Anaesthetists (FRCA).6

In this study, we aim to compare the performance of graduates of UK medical schools in the two parts of the Member of the Royal College of Obstetricians and Gynaecologists (MRCOG) examinations. Membership is a prerequisite for doctors wishing to complete training in obstetrics and gynaecology. Part 1 MRCOG is an examination in the basic and clinical sciences relevant to obstetrics and gynaecology. It can be taken at any time after the completion of a primary medical degree. With the introduction of Modernising Medical Careers programme in the UK, it is a requirement for specialty trainees who are progressing from specialty training year (ST) 2 to ST3. Part 2 MRCOG can be taken once the Part 1 examination has been passed and the appropriate training successfully completed following this, and is now a requirement for specialty trainees in order to progress from ST5 to ST6.7

Part 1 MRCOG consists of two papers of 2 h duration each. Each paper consists of two question formats, extended matching questions (EMQs) and true/false multiple choice questions (MCQs). Paper 1 includes questions from the fields of anatomy, embryology, endocrinology, microbiology, virology, pharmacology, statistics, epidemiology, clinical trial design and analysis. Paper 2 includes questions from the fields of biochemistry, molecular and cell biology, biophysics, genetics, genomics, regulation of gene expression, immunology, pathology and physiology.

Part 2 MRCOG written examination consists of three formats: true/false MCQs 25%, EMQs 15% and short answer questions 60%. A candidate must pass the written examination before proceeding to the oral examination. Part 1 and Part 2 MRCOG papers use a pass mark that has been standard set, which is recognition that some examination courses are more difficult than others. Pass marks thus fluctuate and bear no relation to pass rates. Standard setting is a complex process which varies for each type of paper, but essentially involves assessment of questions on an individual basis in terms of their difficulty. A large panel of carefully trained representative consultants implement the standard setting procedures. They are asked to review the questions bearing in mind the standard that a competent trainee should achieve by the end of core training in obstetrics and gynaecology. A modified Angoff method is used for the MCQ and EMQ papers.7 Many doctors attend various preparation courses for this examination and the Royal College of Obstetricians and Gynaecologists (RCOG) also organises revision courses. The overall pass rate for candidates who attended the RCOG Part 1 revision courses in 2005 was 47% in contrast to the overall pass rate of 32%.7 These pass rates are similar to the candidates attending revision courses for the (MRCP) (UK) examination.8

This study was undertaken to determine the pass rate for first time candidates in the MRCOG Part 1 and Part 2 examinations stratified by the university to obtain primary medical qualification. It is unknown whether graduates of different universities fare differently and if there are any potential issues that need to be addressed.

Methods

The primary data for the analysis were retrieved from the RCOG database. Candidates included had taken Part 1 and Part 2 MRCOG written examinations for the first time between 1998 and 2008. Candidates were identified by their numbers only thus maintaining confidentiality and authors had no access to the names of the candidates. Medical school descriptors in the RCOG database follow the same format as used by the General Medical Council and only the university awarding the degree was recorded. As a result medical schools of the University of London are all recorded as ‘London’.

Data on mean A level grades for each university for 1998 were purchased from the Higher Education Statistics Agency (HESA) (HESA, 95 Promenade, Cheltenham, UK) database.

All retrieved data were stored on Microsoft Excel worksheets.

Data analysis

To analyse the performance of each medical school, they were compared using χ2 tests for categorical data (eg, pass rate). Second, the difference for each candidate was calculated from the pass mark. The differences were compared between medical schools using analysis of variance. Logistic regression and analysis of variance were undertaken to control for the effects of gender on the pass rate and difference from pass rate. A p value of <0.05 was considered to be statistically significant. Data were analysed using SPSS V.17 (SPSS Inc.).

Results

A total of 1335 UK graduates registered for the Part 1 MRCOG examination for the first time between 1998 and 2008 (table 1) and 1334 were included in the study as the one candidate from Exeter was excluded.

Table 1

The MRCOG part 1 and part 2 written examination results for first time candidates between 1998 and 2008 for graduates from each United Kingdom university

Regarding the Part 2 MRCOG written examination, 822 UK graduates sat for the first time between 1998 and 2008 (table 1).

Results are presented as the number (%) of passes and the mean (SD) of the difference away from the pass mark.

Candidates were stratified by the university of primary medical qualification and mean A level points achieved by first degree entrants to medicine at each university for 1998.

Medical school effects

In the Part 1 MRCOG examination the overall pass rate was 43.1%. There was a statistically significant difference in the pass rates between the medical schools (χ2 (18)=66.7, p<0.0001). Graduates of Oxford, Cambridge, Bristol and Edinburgh performed better, while the graduates of Liverpool, Southampton and Wales performed worse than the remaining cohort as shown in table 1 (figure 1).

Figure 1

Pass rate (95% CI) for fist time candidates in the MRCOG part 1 examination for graduates of each UK university.

There was a significant difference between the medical schools and the difference in the mark obtained from the overall pass mark (table 1, figure 2; F(18)=5.219, p<0.001).

Figure 2

Mean difference (95% CI) away from the pass mark for first time candidates in the MRCOG part 1 examination for graduates of each UK university.

In the Part 2 MRCOG written examination, the overall pass rate was 61.3%. There was a statistically significant difference in the pass rate between the medical schools (χ2 (18)=46.49, p<0.0001). The candidates of Cambridge, Edinburgh, Newcastle and Oxford performed significantly better. The candidates of Glasgow and Leicester underperformed compared with the remaining cohort (table 1, figure 3).

Figure 3

Pass rate (95% CI) for fist time candidates in the MRCOG part 2 written examination for graduates of each UK university.

There was a significant difference between the medical schools and the difference in the mark obtained and the overall pass mark (Table 1, figure 4; F (18)=2.462, p=0.001).

Figure 4

Mean difference (95% CI) away from the pass mark for first time candidates in the MRCOG part 2 written examination for graduates of each UK university.

There is a significant correlation between the success rate of individual universities in the Part 1 and Part 2 written examination (Spearman's correlation coefficient 0.64, p=0.003).

Gender effects

The study also showed that 80% of candidates who took the Part 1 MRCOG between 1998 and 2008 were female candidates. There was a decline in the number of female candidates from 1998 to 2001, and then a steady increase from 2002 to 2008. Similar variation is seen in male candidates taking the examination.

The pass percentage remained steady, being unaffected by the number of candidates (figure 4). There were 261 male candidates, out of which 124 (47.5%) passed the examination. There were 1075 female candidates, out of which 452 (42%) passed the examination. There was no significant difference between the pass rate between male and female candidates (χ2 (1)=2.57, p=0.11) (table 2).

Table 2

Male and female first time candidates and success rates in the Member of the Royal College of Obstetricians and Gynaecologists (MRCOG) Part 1 and Part 2 written examinations

There was no significant difference between the difference from pass mark and gender. The mean difference for female candidates was −1.75 (6.79 SD) compared with −1.43 (6.72 SD) for male candidates (t(1334)=0.674, p=0.500).

After controlling for gender, there was still a significant difference between universities and the per cent pass rate (p<0.001), and the difference from the pass mark (p<0.001).

Overall, 66% of candidates who took the MRCOG Part 2 between 1998 and 2008 were female candidates. There was no significant difference in the proportion of male and female candidates who took the Part 2 MRCOG between medical schools (χ2 (18)=25.099, p=0.122). However, female candidates always outnumbered their male counterparts.

There was a significant difference in the pass rate between male and female candidates (χ2 (1)=12.6, p=0.0001). The total number of male candidates taking MRCOG Part 2 written between 1998 and 2008 was 278, out of which 147 (52.9%) passed the examination. There were 543 female candidates, out of which 355 (65.4%) passed the examination (table 2).

There was a significant difference between the difference from pass mark and gender. The mean difference for female candidates was 1.66 (6.4 SD) compared with −0.72 (6.67 SD) for male candidates.

However, even after controlling for gender, there was still a significant difference between universities and the per cent pass rate (p=0.001), and the difference from the pass mark (p=0.016).

Academic performance effects

HESA student records demonstrate the mean A level scores for 1998 first degree medical school entrants (table 1).

The mean academic performance of the individual universities correlates with the pass rate in the Part 1 examination (Spearman's correlation coefficient 0.67, p<0.01) but there was no significant correlation at the Part 2 written examination (Spearman's correlation coefficient 0.45, p=NS).

Discussion

The results show that candidates who have trained at different medical schools perform differently in the Part 1 and Part 2 MRCOG examinations. A graduate from Oxford, Cambridge, Bristol, Edinburgh, Nottingham and London was more likely to pass Part 1 MRCOG examination at the first attempt than the remaining graduates. A graduate from Liverpool, Southampton and Wales was less likely to pass first time than the remaining graduates. In the Part 2 MRCOG written examination, candidates from Cambridge, Edinburgh, Newcastle and Oxford were significantly more likely to pass the examination in first attempt and the graduates from Glasgow and Leicester less likely to pass than the remaining graduates. The MRCOG results are very similar to medical school graduates' performance in the primary FRCA examination; as far as better than average performance is concerned, the five medical schools which have shown better performance in that study were Oxford, Cambridge, Edinburgh, Bristol and Newcastle-upon-Tyne. The MRCOG results are in contrast for the medical schools who have underperformed in the primary FRCA examination, which were Sheffield, Aberdeen, Leicester, Dundee and Belfast.6 There are similar medical school performances in the MRCP (UK) examination, with Newcastle, Oxford and Cambridge being on the upper quartile and Liverpool in the lower quartile of both series.6

The study showed that 38.9% of female graduates passed the Part 1 MRCOG examination in first attempt compared with 44.8% of male graduates; however, this result was not statistically significant. Female graduates in the primary FRCA examination and the MRCP (UK) Parts 1 and 2 have performed less well than male candidates.4 6 The one common feature of Part 1 MRCOG, primary FRCA and MRCP Parts 1 and 2 is that they are all MCQ/EMQ type papers. In contrast female candidates have performed better than male candidates in the clinical assessment examination (PACES) of the MRCP (UK)4 as both examinations have clinical component which proves that female candidates do perform better if the exam includes face to face appearance with the examiner.4 The same is true for performance at medical schools where female candidates tend to outperform male candidates and are more likely to get an honours degree.9 10

The cause of differences found in the performance of candidates appears to be multifactorial and should not be considered as the sole indicator of excellence of a medical school. One explanation for the better performance in postgraduate exams of medical graduates from Oxford and Cambridge is the selection process and entrance requirements of these two universities: they select the graduates most able to pass these exams. The better qualified entrants (by mean A level score) demonstrated a significant correlation in the Part 1 but non-significant correlation in Part 2 examination success. This finding is of debatable validity as the A level scores quoted only refers to first degree entrants and there are a number of second degree medical school entrants who are not, therefore, included. In addition, there is an assumption that obstetrics and gynaecology trainees are recruited from all levels of academic achievement uniformly. To examine whether the A level performance is an independent predictor of postgraduate examination success it would be necessary to know each candidate's A level grades. It has also been shown that A level grades offered correlate with performance in MRCP (UK) Part 1.11 Another factor which is important at all levels of education is the difference in education or training at medical school.12 Institutions can be different in terms of teaching techniques, course emphasis or focus which can have an impact on future careers.

There is a limitation in this study; we have only analysed results of Part 1 and Part 2 written MRCOG written examinations, which represent only a small percentage of medical graduates who will appear in all postgraduate examinations and so the results of this study cannot be generalised as an overall indicator of performance of a medical school.

Conclusion

Our results show that there is variation in the performance among the graduates of different medical schools in the Part 1 and Part 2 MRCOG written examinations. Female candidates performed significantly better than male candidates in the Part 2 and there were no gender differences in the Part 1 examination. There is a significant correlation between the success rate of individual universities in the Part 1 and Part 2 written examinations.

Main messages

  • Member of the Royal College of Obstetricians and Gynaecologists examinations success rates are significantly different according to the university of medical graduation.

  • Success rates of individual universities in the Part 1 and Part 2 written examinations are significantly correlated.

  • Female candidates performed significantly better in the Part 2 written examinations than male candidates.

  • The mean A level scores of first degree entrants to individual universities correlated with success rates in the Part 1 but not the Part 2 written examinations.

Current research question

  • What are the components of the undergraduate curriculum which predict success in the Member of the Royal College of Obstetricians and Gynaecologists examinations?

References

View Abstract

Supplementary materials

  • Press release

    Press release

    Files in this Data Supplement:

Footnotes

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.