OBJECTIVE: To evaluate the efficacy of carbon-fibre shoe-stiffening inserts in individuals with first metatarsophalangeal joint osteoarthritis. DESIGN: This was a randomised, sham-controlled, participant- and assessor-blinded trial. One hundred participants with first metatarsophalangeal joint osteoarthritis received rehabilitation therapy and were randomised to receive either carbon fibre shoe-stiffening inserts (n = 49) or sham inserts (n = 51). The primary outcome measure was the Foot Health Status Questionnaire (FHSQ) pain domain assessed at 12 weeks. RESULTS: All 100 randomised participants (mean age 57.5 (SD 10.3) years; 55 (55%) women) were included in the analysis of the primary outcome. At the 12 week primary endpoint, there were 13 drop-outs (7 in the sham insert group and 6 in the shoe-stiffening insert group), giving completion rates of 86 and 88%, respectively. Both groups demonstrated improvements in the FHSQ pain domain score at each follow-up period, and there was a significant between-group difference in favour of the shoe-stiffening insert group (adjusted mean difference of 6.66 points, 95% CI 0.65 to 12.67, P = 0.030). There were no between-group differences for the secondary outcomes, although global improvement was more common in the shoe-stiffening insert group compared to the sham insert group (61 vs 34%, RR 1.73, 95% CI 1.05 to 2.88, P = 0.033; number needed to treat 4, 95% CI 2 to 16). CONCLUSION: Carbon-fibre shoe-stiffening inserts were more effective at reducing foot pain than sham inserts at 12 weeks. These results support the use of shoe-stiffening inserts for the management of this condition, although due to the uncertainty around the effect on the primary outcome, some individuals may not experience a clinically worthwhile improvement.