Grant Number:	5R01CA172639-05 Interpret this number
Primary Investigator:	Wharam, James
Organization:	Harvard Pilgrim Health Care, Inc.
Project Title:	High Deductible Insurance: Impact on Breast Cancer Care and Outcomes
Fiscal Year:	2017

DESCRIPTION (provided by applicant): Breast cancer affects 12% of women during their lifetimes and is the third leading cause of death among women age 45-64. Patients being evaluated for breast cancer require a spectrum of advanced medical care including diagnostic imaging and biopsy procedures. Breast cancer patients also need access to surgical treatments and life-saving medications. A rapidly growing form of health coverage - high-deductible insurance - substantially increases patients' out-of-pocket costs for such services. Families in high-deductible health plans must pay up to $12,000 per year before more comprehensive coverage begins. Such high cost sharing levels could substantially reduce access to life-saving care, especially among underserved women such as the poor. It is essential to understand how high-deductible health plans affect breast cancer care and outcomes. Suboptimal access to care could reduce or delay detection, accelerate disease progression, and increase deaths. This proposal seeks to assess the impact of high-deductible health plans on breast cancer diagnostic testing, treatment, and outcomes in a nationally representative population. Measures of diagnostic testing will include diagnostic mammography, breast ultrasound, and breast biopsy. We also will assess changes in surgical tumor resection and adjuvant hormonal therapy use. Our primary outcome will be all-cause mortality among women screened for breast cancer. In addition, we will determine whether select high-deductible health plans with low drug cost-sharing improve outcomes compared with otherwise similar high drug cost- sharing high-deductible plans. Finally, the research will quantify the degree to which transition to high- deductible health plans affects socioeconomic disparities in these measures. This study will draw from a 15-year rolling sample of members from a large national health insurer whose employers mandated a switch from traditional to high-deductible health plans. We will use employer- and member-level propensity score matching to minimize selection bias. The study will employ strong quasi- experimental designs including interrupted time-series with comparison series and Kaplan-Meier survival curves to examine outcomes of interest. This project will be the first to examine these research questions on a national scale. Policy makers will be able to use results to design value-based insurance plans that optimize breast cancer care.





None