Presentation Description
Miriam Armstrong1
David Black1, Liliana Chis2, Hristina Evans1, David Hope3, Abigail Schneider1 and Mike Jones1
1 Federation of Royal Colleges of the Physicians of the United Kingdom
2 MRCP(UK)
3 University of Edinburgh
David Black1, Liliana Chis2, Hristina Evans1, David Hope3, Abigail Schneider1 and Mike Jones1
1 Federation of Royal Colleges of the Physicians of the United Kingdom
2 MRCP(UK)
3 University of Edinburgh
Background:
We examined the contribution of sex, ethnicity, age and PMQ to variation in UK physician recruitment scores and MRCP Part 1 results, extending the evidence-base for interactions between demographic characteristics and performance. We evaluated self- assessment as a component of selection and associations with interview and exam performance. Using continuous score outcomes measures attainment gaps among passing candidates and lets us explore nonlinear effects.
We examined the contribution of sex, ethnicity, age and PMQ to variation in UK physician recruitment scores and MRCP Part 1 results, extending the evidence-base for interactions between demographic characteristics and performance. We evaluated self- assessment as a component of selection and associations with interview and exam performance. Using continuous score outcomes measures attainment gaps among passing candidates and lets us explore nonlinear effects.
Summary of work:
Using a retrospective longitudinal cohort sample of N=11,791 physicians recruited into UK Core Medical Training in 2012-2018, we applied multilevel modelling to test the associations of self-assessment and interview scores and MRCP Part 1 exam scores, and then measure the effects of sex, ethnicity, age and PMQ on those associations.
Using a retrospective longitudinal cohort sample of N=11,791 physicians recruited into UK Core Medical Training in 2012-2018, we applied multilevel modelling to test the associations of self-assessment and interview scores and MRCP Part 1 exam scores, and then measure the effects of sex, ethnicity, age and PMQ on those associations.
Results:
Self-assessment and interview scores were non-linearly associated. Medium-large associations were observed for low-scorers, who scored poorly on self-assessment and interview, but there was effectively no association for those who provided a high self- assessment. The association between recruitment scores and MRCP Part 1 performance was linear, with a large effect. The multilevel model showed that sex, ethnicity, age and PMQ impacted the strength of the associations between these variables, indicating demographic factors influence scores in postgraduate assessment. Age correlated moderately and non- linearly with recruitment and exam results, with younger physicians (<30) scoring slightly higher.
Self-assessment and interview scores were non-linearly associated. Medium-large associations were observed for low-scorers, who scored poorly on self-assessment and interview, but there was effectively no association for those who provided a high self- assessment. The association between recruitment scores and MRCP Part 1 performance was linear, with a large effect. The multilevel model showed that sex, ethnicity, age and PMQ impacted the strength of the associations between these variables, indicating demographic factors influence scores in postgraduate assessment. Age correlated moderately and non- linearly with recruitment and exam results, with younger physicians (<30) scoring slightly higher.
Discussion:
All four examined demographic characteristics impact performance in postgraduate assessment. The support required by different physicians in training should be tailored according to the requirements of their specific demographic characteristics.
All four examined demographic characteristics impact performance in postgraduate assessment. The support required by different physicians in training should be tailored according to the requirements of their specific demographic characteristics.
Conclusions:
Sex, ethnicity, age and PMQ intersect in strong and complex ways with recruitment scores and examinations performance. Interactions between demographic variables need further examination and understanding before their implications can be fully understood.
Sex, ethnicity, age and PMQ intersect in strong and complex ways with recruitment scores and examinations performance. Interactions between demographic variables need further examination and understanding before their implications can be fully understood.
Take-home messages:
These interactions in postgraduate training data must be further investigated if we are to minimise attainment gaps.
These interactions in postgraduate training data must be further investigated if we are to minimise attainment gaps.
References (maximum three)
- Patterson, F., S. Lopes, S. Harding, E. Vaux, L. Berkin, and D. Black. "The Predictive Validity of a Situational Judgement Test, a Clinical Problem Solving Test and the Core Medical Training Selection Methods for Performance in Specialty Training." Clin Med (Lond) 17, no. 1 (Feb 2017): 13-17. https://doi.org/10.7861/clinmedicine.17-1-13. https://www.ncbi.nlm.nih.gov/pubmed/28148572
- Pyne Y, Ben-Shlomo Y. Older doctors and progression through specialty training in the UK: a cohort analysis of General Medical Council data. BMJ Open 2015; 5:e005658. doi:10.1136/bmjopen-2014-005658
- Evans, A. W., C. McKenna, and M. Oliver. "Self-Assessment in Medical Practice." J R
Soc Med 95, no. 10 (Oct 2002): 511-3. https://doi.org/10.1177/014107680209501013. https://www.ncbi.nlm.nih.gov/pubme d/12356978