Using patient-reportable clinical history factors to predict myocardial infarction

Using a derivation data set of 1253 patients, we built several logistic reg ression and neural network models to estimate the likelihood of myocardial infarction based upon patient-reportable clinical history factors only. The best performing logistic regression model and neural network model had C-i ndices of 0.8444 and 0.8503, respectively, when validated on an independent data set of 500 patients. We conclude that both logistic regression and ne ural network models can be built that successfully predict the probability of myocardial infarction based on patient-reportable history factors alone. These models could have important utility in applications outside of a hos pital setting when objective diagnostic test information is not yet be avai lable. (C) 2000 Elsevier Science Ltd. All rights reserved.