Barriers to exercise prescription among Australian general practitioners and endocrinologists: a cross‐sectional survey

BACKGROUND: Physical inactivity is a leading preventable cause of chronic disease in Australia. Exercise prescription by physicians increases patients' physical activity, but implementation is inconsistent, and barriers among Australian physicians are not well characterised. AIMS: To investigate barriers and facilitators to exercise prescription among Australian general practitioners (GPs) and endocrinologists and to identify physician characteristics associated with prescription frequency. METHODS: Cross-sectional survey of 118 Australian GPs (n = 59) and endocrinologists (n = 59) recruited via purposive and snowball sampling (December 2021-December 2022). Chi-squared tests examined associations with prescription frequency; logistic regression adjusted key associations for specialty, age and physicians' own activity levels. RESULTS: Of 118 physicians, 59% prescribed physical activity to ≥61% of eligible patients (high prescribers). The strongest correlate was systematic assessment of patients' physical activity (81% vs 43% high prescribers; adjusted odds ratio (OR) 5.50, 95% confidence interval (CI): 2.04-14.80). Physicians doubting guideline feasibility were less likely to prescribe (46% vs 76%; adjusted OR 3.20, 95% CI: 1.17-8.75), and feasibility scepticism was associated with physicians' own activity: 74% of those meeting guidelines believed them feasible versus 27% of those who did not (P < 0.001). Written prescriptions and perceived patient disinterest remained associated with prescription frequency after adjustment; associations for training and time barriers attenuated. Knowledge of guidelines was not associated with prescription frequency. CONCLUSIONS: Systematic assessment practices and physicians' feasibility beliefs may be more important to exercise prescription than guideline knowledge. Physicians' own exercise habits are associated with their beliefs about what is feasible for patients, with implications for the design of medical education.