Spring 2016 Seminar Series

DATE: 4/28/16
LOCATION: E51-325
TIME: 4:15pm
Reception immediately following

SPEAKER:
Nicoleta Serban

TITLE
Health Analytics in Action: Modeling Heterogeneity in Healthcare Utilization Using Massive Medical Claims Data

ABSTRACT
Health Analytics is a means for understanding, predicting, optimizing and evaluating the healthcare system using real or synthetic data. I will begin this seminar with an introduction of a high level implementation of health analytics for a large database of medical claims for millions of people insured under the Medicaid program. I will illustrate the challenges of such implementation along with the policy making relevance of deriving knowledge from the Medicaid claims data using a modeling approach for characterizing heterogeneity in healthcare utilization. The focus is on pediatric asthma, one of the major respiratory chronic conditions in children. A first step in the modeling approach is the translation of the medical claims observed for a large number of patients and across multiple years into patient-level discrete point-of-care events called utilization sequences. I will contrast two models applied to the patient-level utilization sequences, both with advantages and limitations. The first approach is a Markov clustering model, useful in summarizing and visualizing a large number of utilization sequences. The second approach is an exponential proportional hazards mixture model. The objective of both models is to cluster patients according to their longitudinal utilization behaviors while the second model also allows to determine the main drivers of variation in healthcare utilization while controlling for the demographic, geographic, and health characteristics of the patients longitudinally. I will conclude with policy implications for targeted interventions to improve adherence to recommended care practices for pediatric asthma.