Informatics Research Seminar: Leveraging Secondary Data for Health Services Research and Policy Evaluations

February 18 @ 4:00 – 5:00 pm

Speaker: Stacie Dusetzina, PhD
Presented from UNC-CH

Broadcast Link: Seminar



Oral cancer medications are an increasingly important tool for treating cancer but are costly for patients and insurers. By late 2014, 34 states and the District of Columbia had passed laws to ensure that privately insured patients pay no more for oral cancer therapies than they would pay for intravenous cancer therapies offered by the same health plan. The effect of these state policies on utilization and spending on cancer treatments is unknown. We used administrative health plan claims from three nationwide insurers aggregated by the Health Care Cost Institute, together representing 25% of individuals under 65 with employer-sponsored insurance across the U.S. The adoption of cancer parity serves as a natural experiment, allowing us to evaluate the effect of state cancer parity laws using a rigorous difference-in-difference-in-differences modeling approach to determine their effect on oral chemotherapy use and costs.


Stacie B. Dusetzina, PhD is a pharmacoepidemiologist and health services researcher by training and has expertise in the design and conduct of secondary database analyses. She has worked extensively with large claims data sources, public use data files and registry-linked claims databases. Her work to date has focused on estimating changes in the utilization and costs of medication in large secondary data sources and on assessing the quality of medication prescribing and use in U.S. based samples. This includes evaluations of adherence and utilization of medications typically dispensed in outpatient pharmacies (oral drug therapies) and those delivered in inpatient or outpatient settings (injectable therapies, devices). Dr. Dusetzina’s primary research focus is on assessing the role of health system policies and costs on prescription drug utilization and the subsequent health outcomes for patients, particularly among individuals with cancer.