Human reliability assessment in surgery--a new approach for improving surgical performance and clinical outcome.

Abstract Surgical operative performance is currently assessed by audit of morbidity, mortality and, especially in patients with cancer, in terms of long-term outcome. Its chief merit is the identification of problems and sub-optimal results by individual surgeons/centres. There is one aspect of audit that constitutes its intrinsic weakness, the verdict on performance it gives is always retrospective--the problem is identified because of the bad results thrown up by analysis of the data. As a result, there is a distinct possibility that surgeons might exclude patients with potentially curative conditions because of increased operative risk due to co-morbid disease from major surgery because of a fear, conscious or otherwise, of comparative under-performance. There is a methodology in established use by industry that is both prospective and prescriptive in ensuring optimal performance--human reliability assessment (HRA), which can be translated into clinical practice. This paper explains the nature of HRA and reports on its initial use in surgery.