SIMULATED NEURAL NETWORKS TO PREDICT OUTCOMES, COSTS, AND LENGTH OF STAY AMONG ORTHOPEDIC REHABILITATION PATIENTS

Our purpose was to develop a set of simulated neural networks that wou ld predict functional outcomes, length of stay, and costs among orthop edic patients admitted to an inpatient rehabilitation hospital. We use d retrospective data for a sample of 387 patients between the ages of 60 and 89 who had been admitted to a single rehabilitation facility ov er a period of 12 months. Using age and data on functional capacity at admission from the Functional Independence Measure, we were successfu l in constructing networks that were 86%, 87%, and 91% accurate in pre dicting functional outcome, length of stay, and costs to within +/-15% of the actual value. In each case the accuracy of the network exceede d that of a multiple regression equation using the same variables. Our results show the feasibility of using simulated neural networks to pr edict rehabilitation outcomes, and the advantages of neural networks o ver conventional linear models. Networks of this kind may be of signif icant value to administrators and clinicians in predicting outcomes an d resource usage as rehabilitation hospitals are faced with capitation and prospective payment schemes.