Objective: We aimed to investigate registrar, practice and consultation characteristics associated with varying degrees of GP registrars’ practice rurality.

Design: A cross-sectional analysis of 12 rounds of data collection (2010-2015) from the longitudinal Registrar Clinical Encounters in Training study, an ongoing, cohort study of Australian GP registrars. The principal analysis used was a generalised ordered logistic regression.

Setting/Participants: GP registrars in training practices within five of 17 GP regional training providers in five Australian states.

Main outcome measure: Degree of rurality of the practice in which the registrar undertook training terms was calculated from the practice postcode using the Australian Standard Classification-Remoteness Area classification.

Results: A total of 1161 registrars contributed data for 166 998 patient consultations (response rate 95.5%). Of these, 56.9% were in major city practices (ASGC-RA1), 25.7% were in inner-regional practices (ASGC-RA2) and 17.4% were in outer-regional/rural practices (ASGC-RA3-5). Several statistically significant associations (P < .001) were found within regional/rural practices (ASGC-RA2-5), when compared with major city practices (ASGC-RA1). These included registrar characteristics such as being in Term 1, being medically trained overseas, and having worked at the practice previously; patient characteristics such as the patient being an existing patient, being older and being Aboriginal or Torres Strait Islander; and consultation characteristics such as performance of procedures.

Conclusion: Our findings suggest that registrars are undertaking rural practice early in their GP training and are being exposed to a rich and challenging mix of clinical and educational practice.