Mental health and learning conditions among French physiotherapy students: a nationwide cross-sectional study of individual, educational and contextual factors

Learning conditions in health professions education may be associated with students’ mental health and academic experience. Physiotherapy training combines academic, clinical and relational demands, yet nationwide evidence on potentially modifiable educational factors remains limited. This study assessed clinically significant depressive, anxiety and insomnia symptoms among French physiotherapy students and examined their associations with individual, educational, social and contextual factors. A nationwide cross-sectional online survey using voluntary non-probability sampling was conducted from 18 November 2025 to 19 January 2026 among students aged 18 years or older enrolled in years 2–5 of physiotherapy education in France. Symptoms were assessed using the Patient Health Questionnaire-9, Generalized Anxiety Disorder-7 and Insomnia Severity Index, with prespecified thresholds of 10, 10 and 15, respectively. Separate multivariable logistic regression models were fitted using cluster-robust standard errors at the training-institution level. Sensitivity analyses assessed the robustness of the main findings. The descriptive sample included 1,717 students from 48 of 63 French physiotherapy training institutions; 1,711 were included in regression analyses. Clinically significant depressive, anxiety and insomnia symptoms were reported by 46.8%, 35.1% and 18.1% of participants, respectively. The strongest and most consistent educational association concerned an unfavourable balance between personal life and training, which was associated with depressive symptoms (adjusted odds ratio [aOR] 3.67, 95% confidence interval [CI] 2.86–4.71), anxiety symptoms (aOR 2.89, 95% CI 2.38–3.51) and insomnia symptoms (aOR 2.70, 95% CI 1.95–3.72). Unfavourable support from teachers or supervisors and peers was also associated with all three outcomes, while high or very high perceived workload was independently associated with anxiety symptoms. In exploratory contextual analyses, institutional status and geographical setting were not independently associated with any outcome. Sensitivity analyses using continuous symptom scores and alternative approaches to clustering largely supported the main findings. Mental health symptoms were frequent, but the most consistent associations concerned proximal and potentially modifiable features of the learning environment, particularly work–life balance and perceived educational and peer support. These findings support a shift from documenting symptom burden alone towards examining how training organisation, supervision and support structures may contribute to student well-being. Physiotherapy programmes should prioritise these educational conditions while prospectively evaluating the effects of interventions.