Outlier analysis in orthopaedics: use of CUSUM: the Scottish Arthroplasty Project: shouldering the burden of improvement.

National joint registries have become well established across the world. Most registries track implant survival so that poorly performing implants can be removed from the market. The Scottish Arthroplasty Project was established in 1999 with the aim of encouraging continual improvement in the quality of care provided to joint replacement patients in Scotland. This aim has been achieved by using statistics to engage surgeons in the process of audit. We monitor easily identifiable end points of public concern and inform surgeons if they breach our statistical limits and become "outliers." Outlier status is often associated with poor implants, and our methods are therefore applicable for indirect implant surveillance. The present report describes the evolution of our statistical methodology, the processes that we use to promote positive changes in practice, and the improvements in patient outcomes that we have achieved. Failure need not be fatal, but failure to change almost always is. We describe the journey of both the Scottish Arthroplasty Project and the orthopaedic surgeons of Scotland to this realization.