Medicaid Expansion Resulted In Decreased Odds Of Frequent Emergency Department Use Among Medicaid Beneficiaries And The Uninsured In New York

Research Objective Frequent emergency department (ED) use has been often cited as a marker of fragmented health care provision and is an indicator of poor quality of care. The Affordable Care Act (ACA) attempted to address the rising costs related to the utilization of ED services, among others, by expanding insurance coverage, by fortifying primary care, by supporting the development of integrated health care delivery models, and by introducing alternative quality‐adjusted payment models. The objective of this study is to describe and to compare frequent ED users before and after the Medicaid expansion under the ACA and to estimate the policy’s effect on the likelihood of frequent ED utilization in New York. Study Design We used the 2011 to 2016 Healthcare Cost and Utilization Project (HCUP) State Emergency Department Databases (SEDD) and State Inpatient Databases (SID) for New York. The number of ED visits per patient was calculated using a patient identifier across the HCUP SEDD and SID data. We defined frequent ED use as four or more visits in one year. Multivariate logistic regression was used to estimate the changes in the likelihood of frequent ED use in the post‐Medicaid expansion period. We conducted sensitivity analyses using alternative definitions of frequent use and falsification analyses examining only pre‐expansion years. Population Studied In‐state residents aged 18‐64, who had Medicaid or private coverage or were uninsured. Principal Findings Frequent users represented 7.1% of all ED patients before and 7.3% after the policy implementation and accounted for 26.1% and 27.0% of all ED visits, respectively. Frequent users were mostly covered through Medicaid across all years (69% vs 35%, P < .001) and had higher rates of comorbidities and chronic conditions compared with infrequent users. After controlling for patient and county‐level characteristics, the policy implementation was associated with 4% ( P < .001) and 12% ( P < .001) reductions in the likelihood of being a frequent ED user among Medicaid beneficiaries and the uninsured, respectively, while the likelihood of frequent use increased by 9% ( P < .001) for private insurance enrollees. Beyond the policy, high ED reliance was continuously associated with patients who had multiple comorbidities and chronic conditions, including mental health, substance, and alcohol use, as well as conditions that could be prevented and treated in non‐ED settings. Conclusions After New York expanded Medicaid in 2014, the likelihood of frequent ED use declined among Medicaid beneficiaries and those remaining uninsured. Implications for Policy or Practice The high share of frequent users in Medicaid plans indicates the continuous need to address social determinants of health at community level, particularly for Medicaid patients.