Examining the Continuity of the Patient-Provider Relationship Among Maryland Medicaid Enrollees

Abstract

Additional research on primary care can make an important contribution to inform efforts to shift away from fee-for-service care models and toward greater provider accountability of care. Initiatives such as Accountable Care Organizations or Patient-Centered Medical Homes aim to improve care through expanded services, such as case management, care coordination, and health promotion. These models are designed to reduce the need for acute care and promote more efficient use of the health care system. However, evaluations of these programs have shown mixed evidence of improved outcomes among participants. One potential explanation is a misalignment between the services as prescribed by these models and what patients perceive as high quality care. Many of these services overlap with those emphasized in pay-for-performance models. Yet the quality of care received can not be fully captured by clinical outcomes alone, because these outcomes are also influenced by patient-level factors. According to the Institute of Medicine, the continuity of the patient-provider relationship is one of the key aims of primary care. Continuity is also listed as a core attribute of high-quality primary care by the National Academy of Medicine. Continuity has been linked to improvements in health outcomes and can serve as an indicator of the quality of the care delivered from the patient’s point of view. Returning to the same provider reflects a patient's trust in the care received and a willingness to continue that relationship. While most existing studies focus on continuity at the patient level, provider-level continuity can offer important insights into provider performance and care delivery patterns. Understanding continuity at the provider level can help identify systemic and organizational factors that support or hinder sustained relationships with patients. However, the majority of provider-level continuity studies have been conducted outside the United States or have focused on populations covered by Medicare or employer-sponsored insurance. To our knowledge, this is the first study to examine provider-level continuity of care among a low-income population in the United States. This dissertation uses data from a cohort of Maryland Medicaid enrollees from 2015-2019, focusing on individuals who sought care from a primary care provider during 2015 and who were enrolled for at least one month during each year. The first aim of this dissertation examines how variation in average continuity (measured using the continuity metric Usual Provider of Care [UPC]) at the provider-level is associated with provider characteristics. The second aim reviews the relationship between provider-level continuity of care and hospital utilization among a provider’s patients. The third aim examines the relationship between provider-level continuity of care and emergency department (ED) use within their patient panels. The results of our study suggest that provider characteristics have a statistically significant relationship to the distribution of the UPC among their patient panel. Larger providers, female providers, more experienced providers, and providers outside of the Baltimore Metro region all have higher logUPC scores, holding all other variables constant. Furthermore, the results of Article 2 show that the regression estimates that a 1% decrease in the logUPC of a provider’s patient panel is associated with a .01% and .05% decrease in their panel’s average ambulatory care sensitive conditions (ACSCs) and inpatient stays, respectively. Providers who are female and/or serve fewer people with very-high comorbidities have patient panels with lower rates of ACSCs and inpatient stays. Article 3 demonstrates that continuity as measured by logUPC has less demonstrable impact on ED utilization rates. While the models show a statistically significant negative relationship between logUPC and rates of ED use, they also show a positive relationship with ED visits classified as non-emergent and/or preventable. Expanding our understanding of patient-provider continuity among Medicaid enrollees can help supplement the literature on continuity for a population with lower incomes, one that is more likely to be comprised of minorities and to have chronic conditions.