Identifying factors that impact patient length of stay metrics for healthcare providers with advanced analytics

Managing patients' length of stay is a critical task for healthcare organizations. In order to better manage the processes impacting this performance metric, providers can leverage data resources describing the network of activities that impact a patient's stay with analytic methods. Interdependencies between departmental activities exist within the patient treatment process, where inefficiency in one element of the patient care network of activities can adversely affect process outcomes.This work utilizes the method of neural networks to analyze data describing inpatient cases that incorporate radiology process variables to determine their effect on patient length of stay excesses for a major NJ based healthcare provider. The results indicate that inefficiencies at the radiology level can adversely extend a patient's length of stay beyond initial estimations. Proactive analysis of networks of activities in the patient treatment process can enhance organizational efficiencies of healthcare providers by enabling decision makers to better optimize resource allocations to increase throughput of activities.