INTRODUCTION: Exercises are commonly prescribed to improve shoulder range following axillary burns, but the effect of additional splinting is unclear. AIM: To compare splinting and exercise to exercise alone in adults with axillary burns. METHOD: Prospective randomised study allocating participants to a splinting (n=27) or no splinting group (n=25). Outcomes measured at six and twelve weeks were shoulder abduction and flexion range, quality of life using the Burn Specific Health Scale-Brief (BSHS-B) questionnaire and upper limb function using the Upper Extremity Functional Index (UEFI) and the Grocery Shelving Task (GST). RESULTS: At week twelve, there was no difference between groups for shoulder abduction (mean difference 0°, 95% CI -22 to 22°), flexion (mean difference 2°, 95% CI -18 to 23°), BSHS-B (mean difference -2 points, 95% CI -23 to 18 points), UEFI (mean difference -3 points, 95% CI -19 to 14 points) and GST (mean difference -9s, 95% CI -20 to 3s). Adherence to splinting decreased from 77% of participants at week one to 16% at week twelve. CONCLUSION: Shoulder splints did not improve clinical outcomes in this study population and low adherence rates suggest splinting may be unacceptable to patients and makes drawing firm conclusions difficult.