Abstract
Importance Measures of health care disparities across racial and socioeconomic groups are needed to guide research and policy, particularly for hospitals, because they are the major locus of health care.
Objective To quantify socioeconomic differences between a hospital’s patients and the surrounding area, and to assess associations with hospital characteristics.
Design A cross-sectional observational study using Medicare fee-for-service claims and the 2019 American Community Survey (ACS) 5-year zip code data. Quantile regressions identified hospital characteristics associated with lowest and highest inclusivity.
Setting Inpatient admissions to non-specialty, non-federal hospitals (N = 3,426) in calendar years 2018 and 2019.
Participants Medicare beneficiaries enrolled in Parts A & B, and over the age of 65 years
Main Outcome and Measures Inclusivity was defined as the differences between education level, income, and proportion of racial groups, between a hospital’s community area and its patient zip codes (obtained from US Census data). The community area was calculated based on a hospital’s patient counts from contributing zip codes and by finding this travel time radius from each hospital campus. Percentiles of the inclusivity scores and their composite were compared across hospital characteristics using quantile regressions.
Results We included 10,221,387 patients with a mean (± standard deviation) age of 77.4 (8.0) years and 55.3% were female. The median travel time radius was 48.2 minutes (IQR: 34.7 to 75.0). Weighted median incomes for hospitalized patients’ zip codes ranged from $27,060 less than their community area to $31,505 more. Black people had the widest weighted percentage differences across hospitals: one hospital had a Black patient proportion 41.6% greater than its community area, while another had a proportion 37.2% less than its community area. Safety net status was the characteristic consistently associated with a higher median inclusivity score (0.4 points greater than non-safety net hospitals, CI: 0.31 to 0.45; p < 0.001). Metropolitan area hospitals had both higher 90th and lower 10th percentile scores compared to non-metropolitan, as did teaching hospitals compared to non-teaching.
Conclusions and Relevance US hospitals’ measured inclusivity varies widely, with patterns of greater segregation in metropolitan areas. The results for safety net hospitals reflect their essential role in the US healthcare system.
Question What are the variation patterns of socioeconomic and racial characteristics of hospital inpatient populations in relation to the demographics of their surrounding communities?
Findings Using Medicare claims data, we calculated the differences in race, income and educational attainment between hospitalized patients and a defined surrounding community area. Quantile regressions showed that the most and the least inclusivity existed in larger, teaching, and metropolitan area hospitals. Safety net hospitals consistently maintained higher inclusivity across both urban and rural settings.
Meaning Systematic patterns of racial and class segregation exist among hospitalized Medicare patients.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
This project was self-funded by the Lown Institute.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
This study was approved by the New England IRB, and a waiver of consent was granted as there was no risk to patients and we did not report identifiable data.
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Footnotes
We updated the period of data included in the study from 2018 to 2018-2019. We made amendments to some of our methods for calculating the community area boundaries, including incorporating travel time from the hospital. The methods section has been updated to reflect this. Subsequently, the results section (including the tables and included figures in the appendix) have been altered. The main findings are similar. In summary, this version of the manuscript has been revised to update the methods for our community area boundary calculations, the period of data included in the study, and the subsequent new results. The main findings regarding hospital characteristics and inclusivity are the same.
Data Availability
Medicare data was accessed after approvals and agreement with CMS. Individual hospital results will be available on https://lownhospitalsindex.org/. Other data can be accessed upon reasonable request to The Lown Institute, please contact info{at}lowninstitute.org.
