Factors affecting the probability of first-year medical student dropout in the UK : a logistic analysis for the entry cohorts of 1980-1992

Objectives To assess the extent to which various factors influence the probability that an individual medical student will drop out of medical school during their first year of study, focusing on the influence both of prior qualifications, such as A-level subjects taken and scores attained, and of type of school and family background. Design Individual-level administrative data for entire cohorts of medical students from 1980/81 to 1992/93 (that is, the graduating cohorts of 1985/86 to 1997/98), yielding data on 51,810 students in 20 medical schools in the UK. The nature of the data permits two different approaches to the modeling of medical school dropout. First, by focusing on the determinants of first-year dropout we are able to exploit data for cohorts up to and including the entry cohort of 1992. Second, as an alternative, for any given cohort one can analyse the probability of a student dropping out in any one of the 5 years of their programme : but in this case one is restricted to data on cohorts no more recent than the 1986 entry cohort.1 In the current paper, we follow the first of these research strategies. Statistical analysis is by logistic regression. Main outcome measures First-year dropout from medical school versus continuation into second year of study. A (first-year) dropout is anyone who left their medical school programme before the end of their first year of study. Results The probability that a student will drop out of medical school during their first year of study is influenced significantly both by the subjects studied at A-level and by the scores achieved. Among students who took Biology, Chemistry and Physics at A-level, each extra grade achieved reduces the probability of dropping out by about one-third of a percentage point. There is an additional effect for students with the maximum A-level score of 30 points in their best 3 A-levels (that is, three grade As): such a student is almost one percentage point less likely to drop out of medical school, ceteris paribus, compared to a student with 28 points. Furthermore, this estimated effect of A-level performance on dropout behavior is very similar for each of the 13 cohorts. In general, indicators of both the social class and the previous school background of the student are largely insignificant, with the exception that students with a parent who is a medical doctor are significantly less likely to drop out. There are significant differences by gender, with males more likely to drop out. There is also evidence of significant age effects, with a tendency for the dropout probability to fall with age. Conclusions Policies aimed at increasing the size of the medical student intake in the UK and of widening access to students from non-traditional backgrounds should be informed by evidence that student dropout probabilities are sensitive to measures of A-level attainment such as subject studied and scores achieved. If traditional entry requirements or standards are relaxed, then this is likely to have detrimental effects on medical schools’ retention rates unless accompanied by appropriate measures such as targeted admissions and focused student support.