A programmed computer can objectively and reproducibly assign cardiovascular causes of death with agreement comparable to a panel of medical referees.
Computer agreement matches expert panels; supports scalable adjudication but leaves open prospective validation in contemporary cohorts.
Oalmann, M. C. (Louisiana State Univ. Medical Center, New Orleans, La. 70112), H. C. McGill, Jr. and R. H. Deupree. Cardiovascular mortality in a community: methodology and objective diagnostic criteria. Amer J Epidem 94: 531–545,1971.—Detailed methodology for collecting information regarding the final illness and for assigning deaths to categories of cardiovascular disease was developed and then tested on 514 male residents who died in Orleans Parish, Louisiana. The methodology was feasible in that cooperation was obtained from physicians, hospital personnel, coroner's office personnel, and personnel in the Orleans Parish Bureau of Vital Records. Assignments of cause of death by programmed computer agreed with the consensus opinion of five medical referees as frequently as the referees agreed with each other or with themselves. The objectivity and valid reproducibility of the programmed computer should be especially useful in defining real cardiovascular differentials among geographic regions and other population subgroups.
No takes yet. Share an insight, caveat, or question.
Oalmann et al. (1971) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: