Influence of the Provider-Patient Relationship on Patient Health-Related Behavior and Outcomes

University Of California Berkeley, Berkeley CA

Investigators

Abstract

Project Summary/Abstract The provider-patient relationship impacts healthcare access and production. In healthcare access, the relationship affects whether a patient has a usual source of care, successfully seeks care, and receives timely care. In healthcare production, the relationship determines the quality, effectiveness, and efficiency of care. This project will examine the influence of the relationship on patient health-related behavior and outcomes. The first subproject will investigate effects of continuity in the provider-patient relationship on patient behavior and outcomes. Provider continuity occurs when the same provider treats a patient over time. Provider discontinuities, defined as breaks in continuity, may affect the quantity and quality of healthcare. The project will determine the causal effects of provider discontinuities that occur with (a) individual physicians in multi- physician groups, (b) individual physicians in single-physician practices, and (c) entire multi-physician groups. The primary analysis will use event study methods to study discontinuity events due to physician exits from the profession and physician relocations outside provider markets. The secondary analysis will add instrumental variable methods to study potential discontinuity events due to physician relocations within provider markets. The second subproject will investigate how factors in the provider-patient relationship contribute to healthcare variation. Provider-specific factors generate effects a provider would have on all patients, whereas patient-specific factors generate effects a patient would have at all physicians. The approach will first decompose healthcare variation into provider-specific factors, patient-specific factors, and idiosyncratic provider-patient match effects using methods from the firm-worker match literature. Second, it will identify (a) organization and physician characteristics that explain provider-specific effects and (b) provider-patient relationship characteristics that explain idiosyncratic provider-patient match effects. Third, it will characterize provider-patient sorting based on patient-specific and provider-specific factors. Finally, it will determine effects of healthcare provider organizations on referrals and associated provider choice using event study methods. The analyses will use claims data from Original Medicare and Massachusetts insured, non-Medicare patients. The provider sample will consist of physicians in outpatient specialties, including primary care. The outcomes will be efficiency (Prevention Quality Indicators), major events (hospitalizations, mortality), spending, new diagnoses, medical care use (preventive, primary, specialty, and emergency care), and drug use. The results will provide evidence on what changes to provider-patient relationships could improve healthcare access, quality, and efficiency. Potential policy and practice changes could increase provider continuity (e.g., regulate narrow network plans), strengthen provider-patient relationships in accountable care organizations (ACOs) or managed care plans, and promote efficient provider choice (e.g., offer more provider information, improve online searches, assign patient-specific default providers, restructure referral incentives).