OBJECTIVES:

In the late 1930s, “redlining” practices, which considered the racial composition of an area, were used to indicate the level of security for real-estate investments in 239 United States cities. Our objective was to compare healthcare utilization between individuals residing in historically redlined and non-redlined areas to better understand the association between this discriminatory practice and health resource utilization in the U.S.

METHODS:

We mapped Medicare FFS Beneficiaries to historical Home Owner’s Loan Corporation (HOLC) categories of lending risk using their 5-digit zip code of residence in the 100% Research Identifiable Files for calendar year 2019. Zip codes often contain more than one HOLC category and therefore a member’s claim data and enrollment are used more than once. We used standardized counting rules from Milliman’s Health Cost Guidelines^TM–Grouper software to organize a summary of medical claims and analyzed differences in average risk score, allowed costs (spend), and utilization by HOLC categories of “hazardous (previously redlined) and “best” (previously non-redlined). Results were normalized for differences in population demographics (age, gender, and risk scores) where appropriate.

RESULTS:

We identified 6.6 million Medicare FFS beneficiaries with an average risk score of 1.9 and average spend of $971 per member per month (PMPM) reside in previously redlined areas and 4.0 million beneficiaries with an average risk score of 1.2 and average spend of $877 PMPM in previously non-redlined areas. We observed the greatest differences in healthcare utilization as follows: outpatient dialysis utilization was 52%, inpatient substance abuse disorder utilization 37%, and psychiatric intensive outpatient service utilization 36%, higher for the previously redlined areas as compared to the previously non-redlined areas.

CONCLUSIONS:

In addition to the well-documented long-lasting impacts of the redlining practices of the 1930s, including the ongoing racial wealth gap, our findings demonstrate disparities in current health resource utilization in historically redlined areas as well.