Evaluating Use and Expenditures on Low Quality Breast Cancer Care in the Medicare Program

Project Summary The provision of low-value (quality) care, defined as services for which the potential for harm outweighs the potential for benefit, is estimated to account for more than $300 billion in health care expenditures annually and thus has critical consequences for patients and the overall health care system (Koziara 2020). Over the past decade, initiatives such as the Choosing Wisely (CW) campaign and the Less is More series sponsored by JAMA Internal Medicine have raised awareness of the magnitude of the problem. (Because most published studies use the term “low-value” we adopt this terminology as equivalent to low-quality). Several studies constructed indicators of low-value care from claims data; most measures were diagnostic test or imaging procedures (Schwartz et al. 2014; Mafi et al. 2021; Segal et al. 2014; Colla et al. 2015). Research to date shows that the CW campaign had little impact on reducing the use of low-value care. Early-stage breast cancer (BC) is a highly prevalent condition for which several metrics of low-value care have been identified by oncology and surgical specialty societies. Thus, efforts to reduce the use of these low-value services will likely yield substantial cost savings. Private Medicare Advantage (MA) plans have a unique toolkit which can be used to reduce the use of low-value care. For example, MA plans may require prior authorization for high-cost procedures, provide incentives for beneficiaries to seek care from a network of “high value” providers and employ financial incentives to influence physicians’ prescribing patterns. In contrast, fee-for-service or “traditional” Medicare (TM) does not impose such constraints and restrictions on patients’ choice of providers or procedures deemed to be of low-value. In addition, different types of MA plans (health maintenance organizations (HMOs) and preferred provider organizations (PPOs)) face varying financial incentives that are likely to influence efforts to reduce use of low-value care. Yet, most previous research either compares all MA plans in aggregate to TM or focuses solely on MA-HMO plans. Comparisons of TM vs MA-PPO plans are nonexistent. Our aims to examine early-stage BC (stages 0, I, IIA, IIB and IIIA) are: Aim 1: To assess differences in receipt of low-value (quality) BC care for women newly diagnosed with early-stage BC across TM and specific types of MA plans, including TM vs MA-HMO plans and TM vs MA- PPO plans, after controlling for individuals’ self-selection into a TM, MA-HMO plan, or MA-PPO plan, patient demographic and clinical characteristics and breast surgeon practice structure. Aim 2: To evaluate incremental spending on low-value BC care among women age 70 and older with early-stage BC enrolled in TM and comparable measures of spending (resource use) among similar women enrolled in MA-HMO plans and MA-PPO plans, after controlling individuals’ self-selection into a TM, MA-HMO plan, or MA-PPO plan, patient demographic and clinical characteristics and breast surgeon practice structure.