OBJECTIVE: To investigate the influence of the performing surgeon's experience and degree of supervision on re-operation rate among patients admitted with a proximal femoral fracture (PFF). METHODS: Prospective study of 600 consecutive patients with proximal femoral fracture in our multimodal rehabilitation programme, between 2002 and 2004. Re-operation rate was assessed 6 months postoperatively. Surgeons were grouped as unsupervised junior registrars versus experienced surgeons operating or supervising. Fractures were stratified as technically undemanding or demanding. RESULTS: Unsupervised junior registrars operated on 23% (137/600) of all and 15% (56/365) of technically demanding proximal femoral fractures. The latter had a higher re-operation rate within 6 months, compared with the rate when more experienced surgeons were present. In logistic regression analysis combining age, gender, American Society of Anaesthesiologists score, New Mobility Score, time to surgery and type of implant, surgery by unsupervised junior registrars was still a significant independent risk factor for re-operation in technically demanding proximal femoral fractures. CONCLUSION: Unsupervised junior registrars should not operate on technically demanding proximal femoral fractures.