Disparities in Emergency Department Visits Among Collocated Racial/Ethnic Medicare Enrollees

Elsevier, Annals of Emergency Medicine, Volume 73, March 2019
Authors: 
Hanchate A.D., Dyer K.S., Paasche-Orlow M.K., Banerjee S., Baker W.E., Lin M. et al.
Study objective: We estimate emergency department (ED) use differences across Medicare enrollees of different race/ethnicity who are residing in the same zip codes. Methods: In this retrospective cohort study, we stratified all Medicare fee-for-service beneficiaries aged 66 years and older (2006 to 2012) by residence zip code and identified zip codes with racial/ethnic diversity, defined as containing at least 1 enrollee from each of 3 racial/ethnic groups: Hispanics, (non-Hispanic) blacks, and (non-Hispanic) whites. Our primary study population consisted of a stratified random sample of approximately equal number of each racial/ethnic group from each zip code with racial/ethnic diversity (N=1,563,631). We identified ED visits, comorbidities, primary-care-treatable status, and patient disposition. We characterized socioeconomic status by zip code poverty rate. The main outcome measure was the ratio of ED visit rate (number of visits/100 person-years) between each minority group and whites. Results: Of 38,423 zip codes nationally, 41% met the racial/ethnic diversity criterion; these zip codes contained 85% of the Medicare fee-for-service beneficiaries. Among enrollees from zip codes with racial/ethnic diversity, the ED visit rate among whites was 45.4 (95% confidence interval 45.1 to 45.6), and the ED visit rate ratio was 1.34 (95% confidence interval 1.33 to 1.36) among blacks and 1.23 (95% confidence interval 1.22 to 1.24) among Hispanics. ED visit rate ratios for both minority groups were greater than 1.00 among all subgroups by age, comorbidity, zip code poverty rate, urban/rural area, and primary-care-treatable and disposition status. Conclusion: Among Medicare enrollees, blacks and Hispanics had higher ED use rates than whites overall and among subgroups by demographics and socioeconomic status.