The University of Washington Tacoma (UWT) and MultiCare Health System (MHS) research partnership leverages the local expertise available in both organizations, to solve big healthcare data challenges.
Congestive Heart Failure (CHF) is one of the leading causes of hospitalization, and studies have shown that many of these admissions are readmissions. Identifying the patients who are at a greater risk of hospitalization can guide implementation of appropriate plans to prevent these readmissions. In the field of medical science, prediction of such outcomes is an extremely challenging task because it involves integration of various variables asso- ciated with patients such as patients socio-demographic factors, health conditions, health care utilization and factors related to health care providers. This work aims at analyzing and building an effective predictive model for this problem that could successfully iden- tify patients who are at a greater risk of future hospital admissions.
Through the use of shared insights, advanced data models and machine learning algorithms, the UWT MHS partnership will explore methods for reducing the instances of readmission for congestive heart failure.
Specifically, as part of this research, the team will develop clustering modules types of possible readmissions and to explore the discriminating role of various factors that impact each readmission type. Next, we develop a classification model to classify each case as high or low likelihood for readmission. We use historical data to validate the robustness of these models and report on the performance on test data. The models can then be deployed to provide results and gain insights.