To the Editor.
—An article published by Haynes and Walker1 in the July 1987 issue of the Archives provides for a timely appraisal of studies to date on the efficacy of computer systems in aiding in the clinical management of patients. I believe, however, that the authors place too much emphasis on clinical outcome as the raison d'être of clinical support systems. As pointed out, the majority of the reliable studies were conducted in outpatient settings. The current interest in integrated hospital systems will provide a different emphasis on what is perceived as a beneficial outcome. As hospital administrations feel more the financial pinch of Medicare payments based on diagnosis related groups (the corollary in Canada being global hospital budgets), a new form of cooperation must inevitably evolve where clinicians are, through hospital systems, made cost-aware of their proposed diagnostic and therapeutic protocols at the time of order entry,