Objectives Identifying the factors that increase the likelihood of medical graduates choosing rural medical careers is key to addressing the global shortage of rural doctors. Using linked graduate-workforce outcomes data, this study aimed to identify predictors of rural medical practice in Aotearoa New Zealand (NZ). Design A national prospective cohort study linking data from the longitudinal Medical School Outcomes Database to workforce location data. Univariate and multivariate analyses were conducted to generate ORs for putative predictors of rural medical career. Setting and participants All NZ medical graduates from 2011 to 2019 were followed for a minimum of 3 years. Results During the study period, there were a total of 4152 medical graduates nationally. Included in the analysis were 3291 graduates who had linked longitudinal medical school and workforce data, of whom 133 (4%) doctors were classified as having decided on a career in rural medicine. Independent predictors of rural practice included being of rural origin (OR 2.13, 95% CI 1.19 to 3.81, p=0.011), age older than 25 years at entry to medical school (OR 2.88, 95% CI 1.54 to 5.36, p<0.001) and participation in an extended rural medical school placement (OR 2.51, 95% CI 1.48 to 4.25, p<0.001). Regional background and short regional medical experiences did not significantly improve rural medical workforce outcomes. Despite these predictors, students of urban origin and those with no rural career intentions at the start of medical school made up over half of the early career rural medical workforce. Conclusions This is the first national study linking medical school data to rural medical workforce outcomes. It demonstrates that previously known predictors of rural practice intention are borne out with actual career outcomes, and these also hold true at a national level. However, this research highlights that diverse pathways into rural practice are vital, given that urban-origin students and those with no early rural career intention make up a substantial number of the early-career rural medical workforce.
Costello et al. (2026) studied this question.