OUTCOME AFTER SEVERE HEAD-INJURY - AN ANALYSIS OF PREDICTION BASED UPON COMPARISON OF NEURAL-NETWORK VERSUS LOGISTIC-REGRESSION ANALYSIS

More reliable prediction of outcome would be helpful for clinicians wh o treat severely head-injured patients. To determine if neural network modeling would improve outcome prediction compared with standard logi stic regression analysis and to determine if data available 24 h after severe head injury allows better prediction than data obtained within 6 h, we tested the ability of both techniques at these two times to p redict outcome (dead versus alive) at 6 months. One thousand sixty-six consecutive patients with Glasgow Coma Scale scores of 8 or less duri ng the first 24 h after injury were randomly divided into two groups. Data from the first group (n = 799) were used to develop the models; d ata from the second group (n = 267) were used to test the accuracy, se nsitivity, and specificity of the models by comparing predicted and ac tual outcomes. The 6-month mortality race was 63.5%. Our findings conf irm the importance of age, Glasgow Coma Scale scores, and hypotension in predicting outcome. Using data available al 24 h improved the predi ctive power of both models compared with admission data; at both time points, however, the differences in the results obtained with the two models were negligible. We conclude that outcome (dead versus alive) a t 6 months after severe head injury can be predicted with logistic reg ression or neural network models based on data available at 24 h. Crit ical therapeutic decisions, such as cessation of therapy should be bas ed on the patient's status 1 day after injury and only rarely on admis sion status alone.