Selecting for a sustainable workforce to meet the future healthcare needs of rural communities in Australia.

An undersupply of generalists doctors in rural communities globally led to widening participation (WP) initiatives to increase the proportion of rural origin medical students. In 2002 the Australian Government mandated that 25% of commencing Australian medical students be of rural origin. Meeting this target has largely been achieved through reduced standards of entry for rural relative to urban applicants. This initiative is based on the assumption that rural origin students will succeed during training, and return to practice in rural locations. One aim of this study was to determine the relationships between student geographical origin (rural or urban), selection scores, and future practice intentions of medical students at course entry and course exit. Two multicentre databases containing selection and future practice preferences (location and specialisation) were combined (5862), representing 54% of undergraduate medical students commencing from 2006 to 2013 across nine Australian medical schools. A second aim was to determine course performance of rural origin students selected on lower scores than their urban peers. Selection and course performance data for rural (461) and urban (1431) origin students commencing 2006-2014 from one medical school was used. For Aim 1, a third (33.7%) of rural origin students indicated a preference for future rural practice at course exit, and even fewer (6.7%) urban origin students made this preference. Results from logistic regression analyses showed significant independent predictors were rural origin (OR 4.0), lower Australian Tertiary Admissions Rank (ATAR) (OR 2.1), or lower Undergraduate Medical and Health Sciences Admissions Test Section 3 (non-verbal reasoning) (OR 1.3). Less than a fifth (17.6%) of rural origin students indicated a preference for future generalist practice at course exit. Significant predictors were female gender (OR 1.7) or lower ATAR (OR 1.2), but not rural origin. Fewer (10.5%) urban origin students indicated a preference for generalist practice at course exit. For Aim 2, results of Mann-Whitney U tests confirmed that slightly reducing selection scores does not result in increased failure, or meaningfully impaired performance during training relative to urban origin students. Our multicentre analysis supports success of the rural origin WP pathway to increase rural student participation in medical training. However, our findings confirm that current selection initiatives are insufficient to address the continuing problem of doctor maldistribution in Australia. We argue for further reform to current medical student selection, which remains largely determined by academic meritocracy. Our findings have relevance to the selection of students into health professions globally.