Evaluating Racial Concordance in a Telephonic Care Management Program Among Black Patients

ABSTRACT

Objective: To examine whether racial concordance between telephonic care managers and Medicare Advantage beneficiaries engaged in a care management program influenced the likelihood of fulfilling a set of identified health care needs.

Study Design: Retrospective study of a real-world telephonic care management program among Medicare Advantage beneficiaries.

Methods: This study involved Medicare Advantage beneficiaries identified as having at least 1 of 23 gaps in care quality at baseline who were randomly assigned to a telephonic care management program between June 2020 and March 2021. We examined participating Black beneficiaries and assigned racial concordance based on engaging with a Black (race-concordant) or White (race-discordant) care manager. The primary outcome was a measure of whether the gap in care was closed at 90 days. We used logistic regression models adjusted for beneficiary characteristics to examine the impact of racial concordance on binary measures of gap closure at the individual and gap levels.

Results: Among the study population of 12,636 Black race beneficiaries, 1291 (10.2%) had a race-concordant care manager and 11,345 (89.8%) had a race-discordant care manager. In adjusted models, beneficiary–care manager racial concordance did not impact closure of gaps in care when examined at the beneficiary level (OR, 0.98; 95% CI, 0.90-1.08) or the gap level (OR, 0.99; 95% CI, 0.88-1.12).

Conclusions: In a real-world telephonic care management program aiming to resolve gaps in care quality, beneficiary–care manager racial concordance did not impact the rate of resolving gaps in care for Black beneficiaries.

Am J Manag Care. 2025;31(11):In Press

Takeaway Points

This study examined whether racial concordance impacted program outcomes among Black Medicare Advantage beneficiaries engaged in a telephonic care management program.

• Black beneficiaries with either a Black or White care manager fulfilled health care needs at the same rate—around 45%.
• Although patient-provider race concordance has been linked with favorable outcomes for Black patients, these benefits were not found in a telephonic care management program that did not have race-related selection.

Racial concordance between a patient and provider has been associated with more favorable health care quality, including medication adherence and preventive care among Black patients under some circumstances.^1-3 However, the contexts and outcomes for which racial concordance provides benefits are not well understood, as overall findings are mixed.⁴ Care management practices delivered in managed care commonly use telephone calls to facilitate care quality and patient engagement.⁵ This setting provides an opportunity to examine the role of race concordance when characteristics of the interaction may affect outcomes but awareness of race may not be explicit.

Racially concordant provider relationships offer an opportunity for shared identity and culturally appropriate care with reduced fears of discrimination and stigma for Black patients, which may promote improved health care experience through greater trust, more favorable communication, and greater satisfaction.^6-10 Despite efforts to reduce disparities, historical and pervading discrimination and structural racism in health care systems continue to be reflected in the poorer health care quality and avoidable adverse outcomes experienced by Black patients.^11-16 Effective strategies to ensure equal treatment and eliminate structural barriers to equitable health and health care quality for Black and other minoritized patients are needed.^17,18 Solutions such as improving diversity in the health care system, including increasing representation of Black health care providers, which may promote more culturally competent care, have been proposed.^19-21

Given the need for effective strategies to reduce racial inequities in health care quality, this study investigated the impact of racial concordance between telephonic care managers (CMs) and Medicare Advantage (MA) beneficiaries. We evaluated whether racial concordance influenced the likelihood of fulfilling a set of identified health care needs among Black beneficiaries in the context of a telephonic care management program initiated in response to the COVID-19 pandemic and designed to assist beneficiaries who were not receiving the appropriate primary and preventive care.

METHODS

Setting and Participants

This retrospective study utilized existing data from a completed quality improvement project in which MA beneficiaries with at least 1 identified gap in care were randomly assigned to receive outreach from a telephonic CM or to a control group with no outreach. Beneficiaries with at least 1 of 23 gaps in care were randomly assigned between June 2020 and March 2021. Randomization was stratified according to a claims-based operational model of 4 categories of risk of inpatient admission. CMs made audio calls to beneficiaries assigned to the outreach condition in states where they held licensure from a queue in the administrative system that prompted calls to the next available CM, providing a scenario where race concordance was plausibly exogenous.

For this study of beneficiary-CM race concordance, we included Black beneficiaries who received outreach from a Black or White CM. Beneficiaries were excluded if they had evidence of COVID-19 infection in the last 30 days, were participating in another outreach program, or were in an MA plan contractually excluded from research. The Humana Healthcare Research Human Subject Protection Office reviewed this retrospective study, which used a limited data set, and determined that it did not meet the criteria of human subject research; thus, it was exempt from institutional review board review.

Measures

Outcomes. The quality improvement project included 23 measures of care quality meant to identify beneficiaries who may not be receiving appropriate primary and preventive care (eAppendix [available at ajmc.com]). Measures included indicators of chronic disease management, preventive care, and medication fulfillment and adherence. Medical and prescription claims were used to identify beneficiaries with gaps in care for any measure. Gaps in care were measured at baseline and were remeasured 90 days post randomization to assess whether the gap in care was resolved. Indicators for the presence of a gap in care at baseline and for gap closure at 90-day follow-up were created. Our primary outcome variable was the indicator for whether the gap was closed at follow-up.

As a secondary outcome, we included the number of calls with a CM as a measure of program engagement. Beneficiaries may have had more than 1 call to help address or follow up on needs with the same or a different CM. We used call records to calculate the number of CM calls each beneficiary had during the program.

Race concordance. Using administrative plan data, we identified Black beneficiaries using the CMS beneficiary race code. CM race was not a factor in the original administration of the care management quality improvement program. To complete this study, we used business records to identify CMs’ self-reported race and included those who identified as Black or White.

CMs were assigned to call beneficiaries identified as having a gap in care. The initial call included a standardized assessment of beneficiaries’ needs. The beneficiary profile and assessment for the program did not include information about race, although race was accessible in the administrative record. We used administrative call records to attribute a CM to each beneficiary based on completion of the initial screening call, with the assumption that the first call was likely to influence response to the program. Using this attribution and the race data, we categorized beneficiary-CM dyads as race concordant (Black beneficiary, Black CM) and race discordant (Black beneficiary, White CM).

Covariates. We used administrative plan data to extract covariates that included age at random assignment, sex, original reason for Medicare entitlement, Part D low-income subsidy (LIS) status, Medicare-Medicaid dual eligibility status, and level of risk of inpatient admission (according to percentiles of an internally derived and validated predictive model: top 1%, 2% to 5%, 6% to 10%, and bottom 90%). We assigned geographic region based on beneficiary state and used zip code to assign rural status when the Rural-Urban Commuting Area code was 4 or greater. Age was categorized as younger than 65, 65 to 74, 75 to 84, and 85 years and older.

Statistical Analysis

We first described the demographic characteristics and the number of gaps in care at baseline between Black beneficiaries attributed to a Black CM or a White CM. We next constructed logistic regression models adjusted for beneficiary characteristics to evaluate the impact of race concordance on 2 binary measures of gap closure. Our independent variable of interest was racial concordance, coded as 1 if racially concordant (Black beneficiary–Black CM) or 0 if racially discordant (Black beneficiary–White CM). We conducted the analysis at the beneficiary level, with a binary outcome measure indicating whether each beneficiary had any gap closed (1=gap closed), and at the gap level, which included all the gaps in care with an indicator of closure for each gap. Although CMs may have multiple attributed beneficiaries, due to the low intraclass correlation from the gap-level model (0.02), we determined that models without CM clustering were appropriate. For each model, we calculated ORs of gap closure associated with race concordance (compared with discordance). For ease of interpretation, we reported predicted gap closure rates for race-concordant and race-discordant dyads. To examine the impact of race concordance on program engagement, we completed an adjusted negative binomial regression model at the beneficiary level with count of calls as
the outcome.

We conducted several sensitivity analyses. To explore whether the role of race concordance may differentially impact outcomes, we estimated predicted gap closure rates in a subgroup analysis. Because some beneficiaries interacted with more than 1 CM and the potential influence of subsequent CMs is unknown, we completed the primary analysis among the subset of beneficiaries interacting only with their attributed CM. Also, given state-specific policies that may impact access to care, we examined the appropriateness of a state random effect.

RESULTS

There were 33,721 eligible Black beneficiaries randomly assigned to receive outreach from a CM. Of those, 12,636 engaged with a Black or White telephonic CM, with 1291 (10.2%) having a race-concordant CM and 11,345 (89.8%) having a race-discordant CM (Table). Most beneficiaries (n = 8466) interacted with a single CM (68% of race-concordant and 67% of race-discordant dyads). Overall, 80% of beneficiaries interacted with CMs of the same race as their attributed CM throughout the program (73% of race-concordant and 81% of race-discordant dyads).

Among the final study population of 12,636, the mean (SD) age was 70.5 (9.8) years, 7399 (59%) were women, 5794 (46%) had disability as the original reason for Medicare entitlement, 2403 (19%) lived in rural areas, and 6012 (48%) either had LIS or were dually eligible. Characteristics were similar between beneficiaries with race-concordant and race-discordant CMs, with standardized differences of less than 0.2. At baseline, the mean number of gaps for Black beneficiaries with race-concordant CMs was 2.00 and with race-discordant CMs was 1.97.

As displayed in the Figure, in adjusted models, beneficiary-CM racial concordance did not impact closure of gaps in care when examined at the beneficiary level (OR, 0.98; 95% CI, 0.90-1.08) or the gap level (OR, 0.98; 95% CI, 0.88-1.12). The predicted proportion of all gaps closed at 90 days overall was 30.3% among race-concordant dyads and 30.7% among race-discordant dyads. The predicted number of beneficiaries with at least 1 gap closed was 45.8% among race-concordant dyads and 45.9% among race-discordant dyads after controlling for covariates.

Beneficiary-CM race concordance did not impact engagement with the program, as measured by the number of telephone calls (incidence rate ratio, 0.96; 95% CI, 0.91-1.01). The adjusted mean number of calls was 2.31 among race-concordant dyads and 2.41 among race-discordant dyads.

Sensitivity Analyses

When analyses were restricted to the 8466 beneficiaries who interacted only with their attributed CM, the results remained unchanged. We did not find a differential effect among beneficiary subgroups based on age, sex, rurality, and income. The covariance parameter from a state-level random effects model showed nonsignificant state variation (estimate, 0.002; P = .24) (eAppendix).

DISCUSSION

In this study of Black MA beneficiaries with identified health care needs engaged in a telephonic care management program, we found that racial concordance was not associated with the rate of resolving gaps in care quality. We utilized data from a completed quality improvement project that randomly assigned a group of beneficiaries to a care management program. This provided an opportunity to examine the potential impact of racial concordance in a real-world health care setting where the race of the CM was exogenous. In this context of an audio call, where beneficiary-CM relationships were not self-selected and where race was not made explicitly known, racial concordance did not impact quality outcomes.

Across a number of in-person settings, racially concordant patient-provider relationships have been associated with positive outcomes.^1-3 Care management—including telephonic outreach—is one commonly used approach by payers to improve care quality and outcomes for patients and provides an opportunity to improve health equity. For in-person interactions, the positive outcomes associated with race concordance could be related to a combination of explicit similarities or representation observed in those settings, as well as more implicit communication patterns.²² Moreover, although a telephonic setting could lessen the potential for racial discrimination or fears of discrimination, it may not provide the benefits that come through observing shared racial or cultural representation.

Limitations

This study has several limitations. Due to a lack of data, we were unable to account for other CM characteristics that may have influenced the outcome (eg, professional experience). Due to limitations of the CMS race and ethnicity data, the ethnicity of the Black beneficiaries is unknown. Also, the program was conducted during the COVID-19 pandemic, and results may have been impacted by the disruption in care patterns during that time. The fact that no impact was observed with telephonic CMs could reflect the importance of race being observed or known or that this environment (ie, short-term relationship of limited scope) was not adequate to facilitate the influence of communication patterns or other implicit mechanisms, although these explanations are tenuous in part due to a lack of information on whether race was discussed or implied.

CONCLUSIONS

Amid evidence of significant racial inequities in health care quality, it is important that health care systems continue to develop and examine strategies to address structural barriers to health and improve health outcomes for Black and other racially minoritized patients.^18,23 We examined race concordance in telephonic care management as one potential avenue for improvement and found that it did not impact outcomes for Black beneficiaries with health care needs, although more study is warranted.

Acknowledgments

The authors would like to thank Alexjandro Daviano, DN, DO, DrPH, of Humana, for his contribution to the quality improvement study background.

Author Affiliations: Humana Healthcare Research, Inc. (MC, AK, AKS, YL, EB), Louisville, KY; Humana Inc. (BWP), Louisville, KY.

Source of Funding: None.

Author Disclosures: All authors are employed by Humana. Dr Katsikas presented a poster at the AcademyHealth 2024 Annual Research Meeting. Drs Boudreau and Powers own stock in Humana.

Authorship Information: Concept and design (MC, AK, AKS, YL, EB, BWP); acquisition of data (YL); analysis and interpretation of data (MC, AK, AKS, YL, EB, BWP); drafting of the manuscript (MC, YL); critical revision of the manuscript for important intellectual content (MC, AKS, EB, BWP); statistical analysis (AK, AKS); provision of patients or study materials (BWP); administrative, technical, or logistic support (BWP); and supervision (EB, BWP).

Address Correspondence to: Melanie Canterberry, PhD, Humana Healthcare Research, 101 E Main St, Louisville, KY 40202. Email: Mcanterberry1@humana.com.

REFERENCES

1. LaVeist TA, Nuru-Jeter A, Jones KE. The association of doctor-patient race concordance with health services utilization. J Public Health Policy. 2003;24(3-4):312-323.

2. Traylor AH, Schmittdiel JA, Uratsu CS, Mangione CM, Subramanian U. Adherence to cardiovascular disease medications: does patient-provider race/ethnicity and language concordance matter? J Gen Intern Med. 2010;25(11):1172-1177. doi:10.1007/s11606-010-1424-8

3. Alsan M, Garrick O, Graziani G. Does diversity matter for health? experimental evidence from Oakland. Am Econ Rev. 2019;109(12):4071-4111. doi:10.1257/aer.20181446

4. Meghani SH, Brooks JM, Gipson-Jones T, Waite R, Whitfield-Harris L, Deatrick JA. Patient–provider race-concordance: does it matter in improving minority patients’ health outcomes? Ethn Health. 2009;14(1):107-130. doi:10.1080/13557850802227031

5. Hong CS, Siegel AL, Ferris TG. Caring for high-need, high-cost patients: what makes for a successful care management program? The Commonwealth Fund. August 7, 2014. Accessed March 17, 2025. https://www.commonwealthfund.org/publications/issue-briefs/2014/aug/caring-high-need-high-cost-patients-what-makes-successful-care

6. Shen MJ, Peterson EB, Costas-Muñiz R, et al. The effects of race and racial concordance on patient-physician communication: a systematic review of the literature. J Racial Ethn Health Disparities. 2018;5(1):117-140. doi:10.1007/s40615-017-0350-4

7. Ku L, Vichare A. The association of racial and ethnic concordance in primary care with patient satisfaction and experience of care. J Gen Intern Med. 2023;38(3):727-732. doi:10.1007/s11606-022-07695-y

8. Brown TT, Hurley VB, Rodriguez HP, et al. Shared decision-making lowers medical expenditures and the effect is amplified in racially-ethnically concordant relationships. Med Care. 2023;61(8):528-535. doi:10.1097/MLR.0000000000001881

9. Loeb S, Ravenell JE, Gomez SL, et al. The effect of racial concordance on patient trust in online videos about prostate cancer: a randomized clinical trial. JAMA Netw Open. 2023;6(7):e2324395. doi:10.1001/jamanetworkopen.2023.24395

10. Cooper LA, Beach MC, Johnson RL, Inui TS. Delving below the surface: understanding how race and ethnicity influence relationships in health care. J Gen Intern Med. 2006;21(suppl 1):S21-S27. doi:10.1111/j.1525-1497.2006.00305.x

11. Egede LE. Race, ethnicity, culture, and disparities in health care. J Gen Intern Med. 2006;21(6):667-669. doi:10.1111/j.1525-1497.2006.0512.x

12. Trivedi AN, Zaslavsky AM, Schneider EC, Ayanian JZ. Relationship between quality of care and racial disparities in Medicare health plans. JAMA. 2006;296(16):1998-2004. doi:10.1001/jama.296.16.1998

13. Fiscella K, Sanders MR. Racial and ethnic disparities in the quality of health care. Annu Rev Public Health. 2016;37:375-394. doi:10.1146/annurev-publhealth-032315-021439

14. National Academies of Sciences, Engineering, and Medicine. Ending Unequal Treatment: Strategies to Achieve Equitable Health Care and Optimal Health for All. The National Academies Press; 2024.

15. Bailey ZD, Feldman JM, Bassett MT. How structural racism works—racist policies as a root cause of US racial health inequities. N Engl J Med. 2021;384(8):768-773. doi:10.1056/NEJMms2025396

16. Yearby R, Clark B, Figueroa JF. Structural racism in historical and modern US health care policy. Health Aff (Millwood). 2022;41(2):187-194. doi:10.1377/hlthaff.2021.01466

17. Chisolm DJ, Dugan JA, Figueroa JF, et al. Improving health equity through health care systems research. Health Serv Res. 2023;58(suppl 3):289-299. doi:10.1111/1475-6773.14192

18. CMS Framework for Health Equity 2022-2032. CMS. Accessed May 20, 2024. https://web.archive.org/web/20240506005818/https://www.cms.gov/files/document/cms-framework-health-equity-2022.pdf

19. Henry TL, Britz JB, Louis JS, et al. Health equity: the only path forward for primary care. Ann Fam Med. 2022;20(2):175-178. doi:10.1370/afm.2789

20. Peek ME. Increasing representation of Black primary care physicians—a critical strategy to advance racial health equity. JAMA Netw Open. 2023;6(4):e236678. doi:10.1001/jamanetworkopen.2023.6678

21. LaVeist TA, Pierre G. Integrating the 3Ds—social determinants, health disparities, and health-care workforce diversity. Public Health Rep. 2014;129(suppl 2):9-14. doi:10.1177/00333549141291S204

22. Cooper LA, Roter DL, Johnson RL, Ford DE, Steinwachs DM, Powe NR. Patient-centered communication, ratings of care, and concordance of patient and physician race. Ann Intern Med. 2003;139(11):907-915. doi:10.7326/0003-4819-139-11-200312020-00009

23. 2023 National Healthcare Quality and Disparities Report. Agency for Healthcare Research and Quality. November 2023. Accessed May 20, 2024. https://www.ahrq.gov/research/findings/nhqrdr/nhqdr23/index.html