Patient Selection and Impacts of Plan Choice on Specialty, Urgent Care Center and Emergency Department Visits and Total Expenditures when Families Choose between a Direct Primary Care Option and a Traditional Option for Employee Medical Benefits

Abstract

Introduction Two of the most difficult and seemingly intractable problems in US health care are high expenditures and the state of primary care. Direct primary care (DPC) is a newer approach to primary care delivery in which patients or employers pay directly for unlimited access to a broad, defined set of primary care services; insurance is not used for primary care. DPC may improve the delivery of primary care in ways that reduce overall utilization and spending and benefit patients and physicians.Purpose This dissertation aimed to determine factors associated with plan selection and quantify effects of plan selection on health care utilization and expenditures when an employer adds a DPC option to its medical benefit offering.Data and Methods Data was from a large employer group in which a DPC option was added to a typical PPO option (the “Standard” plan). The two plans differed only in elements related to primary care delivery and enrollees who chose DPC were required to switch to a DPC physician for primary care.The selection aim used logistic regression at the family level to examine the plan selected when DPC was first offered. The utilization and expenditures aims used two period multiple linear regression difference in differences at the member level, examined 18 months before and 39 months after DPC was first offered. Expenditures excluded pharmaceutical costs and included DPC fees.Results Plan choice was statistically significantly associated with enrollee age, enrollee race/ethnicity, presence of chronic conditions in the family (CCs) and presence of a usual source of care in the family (USC). DPC enrollees were more likely to be younger, white or other race/ethnicity, without CCs and without a USC.DPC statistically significantly increased specialist visits. Some results indicated that DPC may have lowered urgent care center (UCC) and emergency department (ED) visits but were inconclusive in the main analyses. DPC increased expenditures and this result was statistically significant when members with the top 1% of expenditures were excluded but not when they were included.CCs and USC were consistently the most statistically significant control variables and had the largest effect size compared to the size of the difference in differences result. Member age was also statistically significant for utilization and expenditure aims.Conclusions In contrast to studies of DPC and similar primary care models, this study did not show DPC reduced ED visits or expenditures. Selection results align with prior studies. The relevance of age, CCs and USC for plan choice, utilization and expenditures aligns with prior research. Future DPC research should examine plan selection, utilization and expenditures over time by year to allow employees more time to consider the new DPC option and to allow changes to doctor-patient relationship and health over time to emerge. Future research of specialist visits should consider that new patients likely experience different referral patterns than existing patients, which would impact specialist visits for DPC but not Standard enrollees. The level of DPC fees should be examined for its effect on expenditures. More research is needed to determine whether and when DPC can reduce utilization and spending.