Shaping the future medical workforce: take care with selection tools.

INTRODUCTION Medical school selection is a first step in developing a general practice workforce. AIM To determine the relationship between medical school selection scores and intention to pursue a career in general practice. METHODS A longitudinal cohort study of students selected in 2006 and 2007 for The University of Auckland medical programme, who completed an exit survey on career intentions. Students are ranked for selection into year 2 of a six-year programme by combining grade point average from prior university achievement (60%), interview (25%) and Undergraduate Medicine and Health Sciences Admission Test (UMAT) scores (15%). The main outcome measure was level of interest in general practice at exit. Logistic regression assessed whether any demographic variables or admission scores predicted a 'strong' interest in general practice. RESULTS None of interview scores, grade point average, age, gender, or entry pathway predicted a 'strong' interest in general practice. Only UMAT scores differentiated between those with a 'strong' interest versus those with 'some' or 'no' interest, but in an inverse fashion. The best predictor of a 'strong' interest in general practice was a low UMAT score of between 45 and 55 on all three UMAT sections (OR 3.37, p=0.020). Yet, the academic scores at entry of students with these UMAT scores were not lower than those of their classmates. DISCUSSION Setting inappropriately high cut-points for medical school selection may exclude applicants with a propensity for general practice. These findings support the use of a wider lens through which to view medical school selection tools.