Background Musculoskeletal complaints in music students are common. Little is known about effectiveness of interventions. Aims To assess whether a biopsychosocial prevention course is better at reducing disability due to musculoskeletal disorders compared with physical activity promotion. Methods This was a multicentre randomized controlled trial with intention-to-treat analysis. Participants were first- and second-year students from five conservatories, randomized to experimental or control groups. The experimental group participated in 11 classes on body posture playing the instrument according to postural exercise therapy, and performance-related psychosocial aspects. The control group participated in five classes promoting physical activity according to national guidelines. The primary outcome was disability using the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire, including the performing arts module. Secondary outcomes were pain, quality of life, playing-related musculoskeletal disorders and health behaviour. Outcomes were assessed at six points, from baseline until 2-year follow-up. Results One hundred and seventy participants were randomized to experimental (n = 84) or control (n = 86) groups. Loss to follow-up was 40% during the trial and 69% at 2-year follow-up. The dropout rate was equal in both groups. Overall, there were no significant differences between groups for any outcome adjusted for baseline characteristics: percentage disability, odds ratio was 1.31, 95% confidence interval (CI) 0.69-2.51; general DASH, β = -0.57, 95% CI -3.23 to 2.09; and performing arts module, β = -0.40, 95% CI 5.12-4.32. Conclusions A biopsychosocial prevention course tailored for musicians was not superior to physical activity promotion in reducing disability. Large numbers lost to follow-up warrant cautious interpretation.