A prescriptive approach to surgical inpatient discharges
Taghi Khaniyev, MIT Sloan School of Management
December 4, Friday
16:00 via Zoom
Predicting inpatient discharges has become an important operational problem as hospitals continue to experience frequent capacity crises caused by, among other factors, an imbalance of the timing of patient admissions and discharges. Often during capacity crises, what is needed beyond a list of discharge-ready patients is another list of potentially dischargeable patients. Having this short list can have a tremendous impact on the level of disruption to the hospital operations. We propose a prescriptive approach to identify such a list of dischargeable patients and prescribe associated interventions for their timely transition out of the hospital. The proposed approach starts with representing patients' clinical and administrative barriers to discharge in a clinically interpretable way, followed by building a neural network model to predict discharge likelihood of a given patient within 24 hours. Finally, using the trained neural network parameters, we employ a mixed integer programming model to identify the minimal subset of barriers that needs to be resolved, i.e., prescribed interventions, in order to lift a patient's discharge likelihood above a given threshold. We show the effectiveness of the proposed approach in a series of retrospective analyses conducted at the Massachusetts General Hospital. Specifically, the fraction of patients who are discharged within 48 hours (a measure of clinical readiness) were shown to be significantly higher (63.2%) among the patients identified by the model as actionable compared to that among non-actionable patients (54.0%). Furthermore, the fraction of prescribed barriers resolved by the time of discharge (a measure of the importance of the barrier) was shown to be significantly higher (50.9%) compared to that of non-prescribed barriers (37.4%).
Taghi Khaniyev is a postdoctoral fellow at MIT Sloan School of Management working in collaboration with Massachusetts General Hospital (MGH) on hospital operations management. His current focus is to develop a machine learning tool that can predict the discharge likelihood of patients and prescribe personalized interventions for the timely discharge of patients. Prior to joining MIT/MGH, he acquired his PhD in Management Science at University of Waterloo. His main research interests are hospital operations management, deep neural networks, data-driven optimization, structure detection and decomposition in mathematical programs, and brain connectivity networks. His research has been published in prestigious scientific journals, an algorithm he developed for decomposition and parallel processing of large-scale optimization problems have been adopted by the software company SAS, the machine learning tool he developed for discharge prediction has become an integral part of the Capacity Coordination Center’s workflow at MGH (Harvard), a surgery duration prediction model he developed was used by Lucile Packard Children’s Hospital (Stanford), and his paper on the network optimization approach for identifying the hub regions in the human brain won the best student paper award by Canadian Operations Research Society (CORS).