154
Non-urgent ED Use by Young Adults under Health Reform (Impact of ACA on Emergency levels of ED visits by Young Adults)
Background: The ACA that signed into law in March 2010 allows young adults age 19-25 to be covered under their parents’ insurance plans. Better access to basic health care is believed to be associated with improved health outcomes and fewer unnecessary ED visits. On the other hands, ED services would be less expensive for patients with insurance coverage, which may lead to more ED visits with even mild conditions. A few studies have examined the impact on the number of ED visits by young adults, and the results are mixed (Amuedo-Dorantes and Yaya 2013, Mulcahy et al. 2013). Nearly no study has examined whether the ED visits vary in terms of emergency levels before and after a major policy change. If a rising trend in ED visits appeared, it could due to an increase in non-avoidable and urgent visits or a rise in preventable or non-urgent visits. Our results are expected to have important implications on issues of overcrowding ED and increasing healthcare costs.
Methods: In Tennessee, there were two major policy changes: a change in state law in 2008 allowing young adults up to 24 years of age to get coverage as dependents and the passage of ACA in 2010 allowing young adults up to 26 years of age to be covered under their parents’ policies. We will use a difference-in-different approach to compare the emergency levels of ED visits in 2007, 2009 and 2011 (a year before any policy change, year after state policy change, year after ACA policy change). Although the sample is not nationally representative, it has detailed information at not only patient level but also hospital and county levels. Since Tennessee had state law that expanded dependents’ insurance coverage (to unmarried and financially dependent children up to age 24) prior to the ACA, the model includes treatment effects for two treated groups at two points of time (young adults aged 19 to 25 covered by private insurance plans as the treated group under ACA after 2010 and young adults aged 19 to 23, unmarried, financially dependent, covered by private insurance plans under the state law after 2008). The control group includes adults with age 26 to 31, who are not affected by the ACA or the state law. Our observation unit is ED visit. The dependent variable is the non-urgent probability of an ED visit (a number between 0 and 1), determined by the NYUED Algorithm. We also include explanatory variables for patient personal characteristics, admission characteristics, hospital characteristics and external access-to-care environment. Our data will come from three sources: (1) Tennessee outpatient discharge data set, (2) Tennessee Joint Annual Report of Hospitals, and (3) The Area Resource File.