Ojectives-To develop, from simple clinical factors, criteria to identify appropriate patients for referral to a surgeon for consideration for arthroplasty, and to rank them in the queue once surgery is agreed. Design-Delphi process, with a panel including orthopaedic surgeons, rheumatologists, general practitioners, epidemiologists, and physiotherapists, who rated 120 case scenarios for appropriateness and 42 for waiting list priority. Scenarios incorporated combinations of relevant clinical factors, Ht was assumed that queues should be organised not simply by chronology but by clinical and social impact of delayed surgery. The panel focused on information obtained from clinical histories, to ensure the utility of the guidelines in practice, Relevant high quality research evidence was limited. Setting-Ontario, Canada. Main measures-Appropriateness ratings on a 7-point scale, and urgency rankings on a 4-point scale keyed to specific waiting times. Results-Despite incomplete evidence panellists agreed on ratings in 92.5% of appropriateness and 73.8% of urgency scenarios versus 15% and 18% agreement expected by chance, respectively. Statistically validated algorithms in decision tree form, which should permit rapid estimation of urgency or appropriateness in practice, were compiled by recursive partitioning. Rating patterns and algorithms were also used to make brief written guidelines an how clinical factors affect appropriateness and urgency of surgery. A summary score was provided for each case scenario; scenarios could then be matched to chart audit results, with scoring for quality management. Conclusions-These algorithms and criteria can be used by managers or practitioners to assess appropriateness of referral for hip or knee replacement and relative rankings of patients in the queue for surgery.
