Making Sense of the Variation in Surgical Breast Cancer Treatment

Treatment variation across healthcare providers is seen as a sign of suboptimal care provision. However, it is often unclear whether such variation is actually provider-driven or rather related to (unobserved) patient characteristics, and even if it is provider-driven, the sources are rarely known. We study the variation in the surgical treatment of breast cancer in the Netherlands using individual claims data from a Dutch health insurer between 2016 and 2018. We estimate a case-mix corrected treatment propensity score that captures a hospital’s propensity to provide lumpectomy instead of mastectomy and use a distance-based instrument to exploit variation between a group of patients (quasi-)randomly assigned to a provider. We find substantial provider-driven treatment variation that, on average, is not driven by selection based on unobserved patient characteristics and impacts clinically important health outcomes for randomly selected patients. This variation is likely driven by differences in hospitals’ capabilities and allocative inefficiencies