Mediators of racial and ethnic inequities in clinical trial participation among patients with cancer, 2011-2023.

In the United States, less than 5% of patients with cancer participate in clinical trials, and racially and ethnically minoritized populations, especially Black and Latinx patients, are less likely to participate ( 1-3 ). These persistent racial and ethnic inequities in trial participation may contribute to disparate access to investigational drugs and, consequently, divergent cancer outcomes and limited real-world generalizability of trial findings ( 4 , 5 ). To address these concerns, the US Congress passed the Food and Drug Omnibus Reform Act of 2022 ( 6 ). Effective in 2024, the act requires drug sponsors to submit diversity plans to the US Food and Drug Administration. These plans must outline approaches for ensuring that clinical trials better reflect the demographics of the intended patient populations, with a focus on including minoritized racial and ethnic groups ( 7 , 8 ).

Efforts to implement evidence-based diversity plans, however, may be hampered by substantial knowledge gaps. Past studies on trial inequities often relied on data from published trials, yet up to 60% of these trials do not report participant race and ethnicity ( 1 , 2 , 9 ), and they often focused on specific health systems, excluding community-based practices, where limited resources may exacerbate inequities ( 10 , 11 ). Moreover, these studies largely overlook the potential role of social determinants of health (SDOH)—the structures, systems, and conditions that shape our lives ( 12 )—in explaining racial and ethnic inequities in clinical trial participation. The few studies evaluating SDOH factors and their influence on trial participation generally relied on relatively small or outdated surveys ( 13 , 14 ), used large geographic measures as proxies for SDOH (eg, county income) ( 15 , 16 ), and focused on socioeconomic status measures ( 17 , 18 ), while neglecting to consider participation barriers such as residential segregation (a form of structural racism) and employing adjusted regression models that failed to account for the potentially mediating effect of SDOH on racial and ethnic inequities ( 19 , 20 ). Identifying factors that mediate, or underlie, racial and ethnic inequities in clinical trial participation is crucial to guide policy, institutional, and clinical initiatives to reduce these inequities.

Here, we aimed to quantify racial and ethnic inequities in clinical trial participation among patients with 22 common cancers and evaluate which area-level SDOH factors mediate observed inequities using a contemporary nationwide electronic health record (EHR) database.

Methods

Study design

This retrospective study used data from Flatiron Health, a longitudinal, real-world database comprising deidentified patient-level structured and unstructured EHR data curated through technology-enabled abstraction ( 21 , 22 ). The deidentified data originated from approximately 280 US cancer clinics (approximately 800 sites of care). Patients were required to have clinical activity in the EHR between January 2011 and December 2022, with follow-up through December 2023 (inclusion criteria are outlined in Supplementary Table 1 , available online). Individual-level clinical and health care factors were ascertained from the EHR. To measure area-level SDOH factors, the EHR dataset was linked to data from the American Community Survey ( 23 ), the US Department of Agriculture ( 24 ), and the US Health Resources and Services Administration ( 25 ) using geocoded patient addresses ( 26 ).

Measures

Clinical trial participation was defined as documented receipt of a clinical study drug (per structured and abstracted EHR data) during the follow-up period. Patients were followed from index diagnosis date (diagnosis of initial, advanced, or metastatic disease, depending on cancer type) ( Supplementary Table 2 , available online) until clinical trial participation. If trial participation was not observed, patients were censored at their last confirmed activity (recording of vital information, medication administration, a noncanceled drug order, or a reported laboratory test), death [per EHR, obituary, and Social Security Death Index data ( 27 )], or end of the study period.

Race and ethnicity were categorized as non-Latinx White (hereinafter referred to as “White”), non-Latinx Black (“Black”), Latinx, or other or unknown (groups with insufficient data for reliable reporting, such as Asian, Pacific Islander, and Indigenous persons, or persons with missing information). The collection of race and ethnicity information varies by clinical site (and clinical workflows) and the limitations of the EHR software used. Although the records likely reflect self-reported assignments by patients, it is unknown whether these values are self-reported, staff reported, or clinician reported.

Clinical factors included age (at index diagnosis), index diagnosis year, sex, disease stage (at initial diagnosis), Eastern Cooperative Oncology Group (ECOG) performance status (maximum value within 30 days of index diagnosis), and cancer type. Patient health care factors included insurance (last insurer during the 2 years before index diagnosis to 1 year following diagnosis) and practice type (academic or community-based cancer center where care was received). Area-level SDOH factors (described in detail in Supplementary Table 3 , available online) included state Census region, county or tract medically underserved area designation ( 25 ), Census tract urbanicity ( 24 ), predominant race and ethnicity (proxy measure of residential segregation) ( 28 ), percentage of persons who are noncitizen (ie, not US citizens), percentage of persons with limited English proficiency, socioeconomic status ( 29 , 30 ), percentage of households with no vehicle ownership, and percentage of households experiencing overcrowding.

Statistical analysis

We estimated the proportion of patients who participated in clinical trials at any time during follow-up. To account for differences in follow-up and competing risks, we also estimated the cumulative incidence of participation in clinical trials within a year (365 days) of index diagnosis using cause-specific time-to-event methods ( 31 ). Differences by race and ethnicity were evaluated using Χ 2 tests for categorical variables and log-rank tests for Kaplan-Meier survivor functions.

Cox proportional hazards models adjusted for clinical factors were used to examine the association between race and ethnicity and clinical trial participation, among all cancers and stratified by cancer type. We also estimated participation to prevalence ratios to characterize the representation of racial and ethnic groups in cancer trials ( 20 ). The participation to prevalence ratio was calculated by dividing the racial and ethnic distribution of clinical trial participants (at any time during follow-up) by the distribution in the entire cohort. We considered racial and ethnic groups with participation to prevalence ratios below 0.95 as underrepresented in clinical trials.

We conducted mediation analysis to examine which factors explain the association of race and ethnicity on clinical trial participation. Informed by the National Academy of Medicine (formerly the Institute of Medicine) definition of health care disparities , or differences not justified by clinical characteristics or patient preferences ( 32 , 33 ), we developed a conceptual framework (drafted as a directed acyclic graph) ( 34 ) that posits that patient health care factors and area-level SDOH may mediate the association between race and ethnicity and trial participation ( Supplementary Figure 1 , available online). Under the proposed framework, it would be inappropriate to adjust for SDOH factors as confounders ( 34 ). Instead, SDOH factors, which are on the causal pathway between race and ethnicity and trial participation, should be considered as potential mediators. In addition to conceptual considerations, factors may mediate this relationship if the following conditions are met: 1) Potential mediators must be associated with the outcome (clinical trial participation) after adjusting for the exposure (race and ethnicity) and 2) potential mediators must be associated with the exposure (race and ethnicity) ( 35 ). If neither condition was met, it was not considered a mediator nor included in the final model. For more information, see Supplementary Table 4 (available online), which presents which SDOH factors were considered mediators based on these conditions. Based on the National Academy of Medicine health care disparities definition, clinical factors were considered confounders a priori.

To inform the interpretation of the mediation analysis, we first conducted unadjusted Cox regressions, evaluating the associations of clinical and SDOH factors on clinical trial participation. Then, we used the multiple mediation analysis method proposed by Yu et al., implemented using a generalized method with multiple additive regression trees ( 35 , 36 ). We chose this approach because it can measure multiple mediating effects, implement time-to-event models, and better account for multicollinearity than linear methods ( 35 , 37 ). We reported the total, direct, and indirect effects of race and ethnicity on clinical trial participation through hazard ratios (HRs) and relative effect percentages, using 100 bootstrap iterations to accommodate uncertainty ( 37 ).

We conducted statistical analyses using R, version 4.0.2 (R Foundation for Statistical Computing, Vienna, Austria) from January 2023 to February 2024. P values were 2 sided, P less than .05 was considered statistically significant, and estimates were presented with 95% confidence intervals (CIs).

Institutional review board (IRB) approval of the study protocol (WCG Clinical, approval No. 420180044), which included a waiver of informed consent, was obtained before study conduct.

Results

Study cohort

This study included 250 105 patients (64.7% White, 8.9% Black, 5.2% Latinx, and 21.2% other or unknown racial or ethnic group) ( Table 1 ). Black and Latinx patients were younger than White patients but had similar rates of de novo metastatic disease. Latinx patients had better performance status scores than Black and White patients but were more likely to have missing values than other groups. Black and Latinx patients were more likely to be uninsured or Medicaid insured than White patients. Most Black patients (91.3%) received care at a community-based cancer center compared with approximately 80% of Latinx and White patients. Black and Latinx patients were more likely to live in economically or socially marginalized areas (ie, disproportionately minoritized, limited English proficiency, and low vehicle ownership) than White patients.

Racial and ethnic inequities in clinical trial participation

In this cohort, 5.2% of patients participated in a clinical trial, and 3.9% began a trial within a year of diagnosis. Black (3.7%) and Latinx patients (4.4%) had lower participation rates than White patients (6.3%) throughout their cancer care trajectory ( Figure 1 ). Similarly, Black (2.7%) and Latinx patients (3.0%) had lower participation rates within the first year of diagnosis than White patients (4.7%) ( Supplementary Figure 2 , available online). Although the proportion of patients participating in a trial within a year of diagnosis remained relatively stable from 2011 (4.4%) to 2022 (3.7%), racial and ethnic inequities persisted ( Figure 2 ). Notably, among patients diagnosed from 2019 to 2022, fewer Black (2.3%) and Latinx (3.2%) patients participated in a trial within a year of diagnosis compared to their White counterparts (4.2%).

After adjusting for clinical factors, Black (HR = 0.55, 95% CI = 0.52 to 0.60) and Latinx patients (HR = 0.63, 95% CI = 0.58 to 0.69) were less likely to participate in clinical trials at any point during follow-up than White patients ( Table 2 ). Statistically significant racial and ethnic inequities in clinical trial participation were observed in 19 of 22 cancer types examined. For example, in the 3 largest patient subgroups—advanced non-small cell lung cancer, metastatic colorectal cancer, and metastatic breast cancer—Black and Latinx patients were 30% to 50% less likely to participate in clinical trials than White patients. Consequently, Black and Latinx patients were adequately represented in trials for only 4 and 5 cancer types examined (participation to prevalence ratio ≥0.95), respectively, while White patients were adequately represented for all 22 cancers ( Supplementary Figure 3 , available online).

Mediation analysis examining SDOH factors underlying racial and ethnic inequities in clinical trial participation

In unadjusted models ( Supplementary Table 5 , available online), clinical factors such as older age, more recent diagnosis year, female sex, earlier disease stage at diagnosis, and higher ECOG performance status were associated with lower participation rates. Lacking insurance, relying on public insurance, and receiving treatment at community-based cancer centers was also associated with lower participation rates. Living in economically or socially marginalized areas was consistently associated with lower participation rates. For example, patients in predominately Black and Latinx neighborhoods had lower participation rates than did patients in White neighborhoods. Similarly, patients in neighborhoods whose residents disproportionately lacked vehicles or English proficiency had lower participation rates than patients living elsewhere.

In the mediation analysis ( Table 3 ), the 40% lower rate of clinical trial participation among Black versus White patients (total effect: HR = 0.60, 95% CI = 0.55 to 0.64) was substantially explained (proportion mediated: 62.2%, 95% CI = 49.5% to 74.8%) by patient health care factors (19.3%, 95% CI = 15.8% to 22.7%) and area-level SDOH (42.8%, 95% CI = 32.1% to 53.6%). Practice type (15.7%, 95% CI = 13.1% to 18.3%) and area-level SDOH, including racial and ethnic composition (29.9%, 95% CI = 21.2% to 38.6%) and low vehicle ownership rates (11.6%, 95% CI = 7.0% to 16.1%), were the most important mediators of the Black-White inequities. The 34% lower rate of clinical trial participation among Latinx versus White patients (total effect: HR = 0.66, 95% CI = 0.61 to 0.72) was also substantially explained (proportion mediated: 65.1%, 95% CI = 49.8% to 80.3%) by patient health care factors (5.7%, 95% CI = 3.0% to 8.4%) and area-level SDOH (59.2%, 95% CI = 45.3% to 73.2%). Area-level SDOH, especially racial and ethnic composition (39.8%, 95% CI = 28.3% to 51.3%), limited English proficiency (11.6%, 95% CI = 2.8% to 20.4%), and low vehicle ownership (9.6%, 95% CI = 5.8% to 13.5%), were the most important mediators of Latinx-White inequities.

Discussion

In this up-to-date evaluation using a nationwide EHR-derived database of patients receiving care for 22 common cancers, there were statistically significant racial and ethnic inequities in clinical trial participation in 19 cancers. Although just 5.2% real-world patients participated in trials, Black and Latinx patients were 40% to 50% less likely to participate than White patients. In a mediation analysis designed to elucidate which factors underlie racial and ethnic inequities in trial participation, we found that area-level SDOH, especially measures of structural racism and practical obstacles, explained 50% to 60% of these inequities. These findings highlight the potential for clinical trials to perpetuate racial and ethnic inequities in access to investigational drugs and subsequent cancer outcomes, demonstrating the urgent need to prioritize upstream causes of racial and ethnic inequities, including SDOH, to facilitate equitable trial participation.

Previous research has identified structural (trial availability), clinical (trial eligibility), and patient and physician barriers to trial participation ( 38 ), but this study uniquely quantified the role of SDOH in explaining racial and ethnic inequities in participation. Neighborhood racial and ethnic composition (a proxy for segregation) emerged as the most substantial mediator, explaining one-third of inequities between Black, Latinx, and White patients. In the context of clinical trials, segregation may be capturing participation barriers that this study could not directly assess. For example, persistent underinvestment in health care systems that serve segregated neighborhoods of color ( 39 , 40 ) can hinder patient access to facilities offering investigational cancer therapies, worsening barriers to trial participation for Black and Latinx patients.

Despite the oft-cited notion that patients of color decline clinical trial participation due to lower trust in the health care system ( 41 ), clinicians themselves may perpetuate this inequity by failing to offer trials to these patients, particularly in segregated neighborhoods, potentially driven by their own biases and discriminatory practices (although not explicitly examined in this study) ( 42 ). Notably, recent research suggests that although Black patients are less likely to be offered participation in a trial than White patients ( 43 ), they are just as likely to participate when offered the chance ( 38 ).

Institutional racism (or the role institutional systems have in race-based oppression, including racism within a specific institution) also influences inequities in trial participation ( 44 ). For example, within institutions known for high-quality cancer care, racial and ethnic inequities persist ( 45 ), with patients of color receiving lower-quality care and worse access to novel agents ( 12 , 46 , 47 ). Trial sponsors, as institutions themselves, may also contribute to these inequities by favoring trial partnerships with well-resourced cancer care facilities ( 38 ), even though those facilities are less likely to serve patients of color ( 48 , 49 ). Finally, health care systems that serve segregated neighborhoods of color may not develop the infrastructure necessary to conduct trials or provide social support to encourage equitable participation (eg, navigation services) because of their own resource constraints ( 39 , 40 ). This may worsen inequities, as living in segregated neighborhoods of color can exacerbate other health care barriers, such as limited access to the resources necessary to manage the financial, time, and logistical demands of trial participation ( 50 ).

Although sustained efforts are necessary to assess and address the mechanisms through which segregation and other forms of structural racism contribute to racial and ethnic inequities in trial participation, this study sheds light on specific barriers that drug sponsors should address to fulfill their Food and Drug Administration mandate of implementing strategies to enhance racial and ethnic diversity in trial participation ( 7 ). For instance, differences in practice type account for approximately 10% of the Black-White inequity in trial participation. In our cohort, most Black patients received care at community-based cancer centers, yet it is widely recognized that trials disproportionately recruit from academic centers ( 38 ). Therefore, increasing recruitment from community-based centers could provide Black patients with more opportunities to participate in clinical trials.

Area-level vehicle ownership, a proxy for individual-level transportation barriers, is also an important mediator, explaining approximately 10% of the trial participation inequities observed across racial and ethnic groups. Distance to clinical sites is a commonly cited concern among potential trial recruits ( 19 , 51 ). To mitigate these concerns, sponsors could consider providing transportation vouchers or using protocols that require less frequent visits. Decentralized trials, where some or all requisite visits are conducted by telemedicine, may also improve participation ( 52 , 53 ), but the implementation of decentralized trials alone may not improve existing inequities because Black and Latinx patients have lower telemedicine access than White patients ( 54 ).

Area-level limited English proficiency, a proxy for patient-level English proficiency, explained more than 10% of the Latinx-White inequity in trial participation, suggesting that trial sponsors should consider strategies to encourage participation among patients with limited English proficiency. These strategies could include hiring multilingual staff, translating trial consent forms, and ensuring that interpreters are readily available. Although EHR portals are increasingly used for trial recruitment, patients with limited English proficiency are less likely to log in ( 55 ). Therefore, sponsors should also ensure that these portals are accessible to patients with limited English proficiency.

Despite legislative efforts, including the National Institutes of Health Revitalization Act of 1993 (encouraging the inclusion of minoritized racial and ethnic populations) ( 56 ) and laws mandating insurance coverage for trial expenses ( 57 ), our study found persistent racial and ethnic inequities in clinical trial participation between 2011 and 2023. This finding underscores the critical need for the Food and Drug Administration’s ongoing commitment to increasing racial and ethnic diversity in trials ( 7 ). Without detailed information about the specific barriers underrepresented groups face, however, these efforts may be less effective. Therefore, policymakers should consider legislation mandating that drug sponsors collect granular race and ethnicity and SDOH information to enable ongoing assessments and timely interventions to address participation inequities.

Lower clinical trial participation among minoritized racial and ethnic populations can exacerbate existing cancer inequities because these trials often provide novel therapies that extend or improve quality of life ( 42 , 58 ), particularly for advanced or metastatic cancers, where other options are limited. This lack of diversity raises concerns about the external validity of trial findings, potentially leading to adverse health outcomes for underrepresented groups. Our study found that Black and Latinx patients were adequately represented in less than one-third of the 22 cancer types examined, while White patients were adequately represented in all. Since minoritized racial and ethnic populations experience distinct risk factors for adverse cancer outcomes (eg, compounding comorbidities, tumor aggressiveness, treatment disparities) ( 12 ), clinical trials conducted among disproportionately White populations may not be scientifically valid (because of issues of transportability and generalizability) ( 5 ) to the broader, more diverse population of patients with cancer in the United States.

In this US cohort of patients with 22 common cancers, Black and Latinx patients were 40% to 50% less likely to participate in cancer clinical trials than White patients. Area-level SDOH factors, especially manifestations of structural racism (eg, segregation) and practical obstacles (eg, lower vehicle ownership and English language proficiency), explained a substantial proportion of racial and ethnic inequities in trial participation. These findings can inform targeted efforts to increase racial and ethnic diversity in cancer trials. Without focused policy and operational efforts to increase clinical trial participation among racially and ethnically diverse patient populations, trials may perpetuate inequities in investigational drug access and generate findings that are not generalizable to real-world patients with cancer.

Strengths and limitations

To our knowledge, this study is the first to use mediation analysis to quantify the extent to which SDOH factors underlie racial and ethnic inequities in clinical trial participation among patients with cancer. We used an up-to-date nationwide, real-world dataset, including patients cared for in academic and community practices for 22 common cancers,, allowing for more generalizable evaluations of existing inequities.

Despite these strengths, our study has limitations. First, our definition of clinical trial participation, which relied on EHR-documented receipt of systemic clinical study drugs and did not capture other trials (eg, non-treatment, surgical, or radiotherapy trials), may underestimate participation and attenuate observable inequities. Second, missing data and unmeasured confounders may have affected the estimation of inequities. For example, 20% of this cohort was categorized as an other or unknown race or ethnicity, which will likely attenuate observed inequities because patients of color are more likely to have missing data ( 59 ). We were also unable to examine inequities among certain racial and ethnic groups (ie, Asian, Pacific Islander, or Indigenous populations) nor disaggregate highly heterogeneous groups (eg, Latinx) because of limited sample sizes. Third, most SDOH factors were measured at the area-level, which may differ from associations between corresponding individual-level SDOH factors and cancer outcomes. Fourth, the EHR data primarily capture advanced-stage cancers; therefore, these findings may not be generalizable to early-stage disease or other cancer diagnoses. Finally, although this study demonstrates the feasibility of identifying mediators of racial and ethnic inequities in trial participation, it does not determine all underlying factors.

Supplementary Material

Acknowledgements

Results from this study were presented, in part, at the 15th American Association for Cancer Research Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved in Philadelphia, Pennsylvania (September 2022); the American Society of Clinical Oncology Annual Meeting in Chicago, Illinois (June 2023); and the International Conference on Pharmacoepidemiology & Therapeutic Risk Management (August 2023). Flatiron Health, Inc, was involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; and decision to submit the manuscript for publication. The parent company (Roche Group) was not involved in the design or any aspect of the conduct of the study. We thank Darren Johnson, PhD, for providing editorial support.

Contributor Information

Jenny S Guadamuz, Flatiron Health, New York, NY, USA; Division of Health Policy and Management, School of Public Health, University of California, Berkeley, Berkeley, CA, USA.

Xiaoliang Wang, Flatiron Health, New York, NY, USA.

Ivy Altomare, Flatiron Health, New York, NY, USA.

Wendy Camelo Castillo, Department of Practice, Sciences, and Health Outcomes Research, University of Maryland School of Pharmacy, Baltimore, MD, USA.

Somnath Sarkar, Flatiron Health, New York, NY, USA.

Ronac Mamtani, Division of Hematology and Oncology, Department of Medicine, University of Pennsylvania, Philadelphia, PA, USA.

Gregory S Calip, Flatiron Health, New York, NY, USA; Now with AbbVie, North Chicago, IL, USA; Program on Medicines and Public Health, Alfred E. Mann School of Pharmacy, University of Southern California, Los Angeles, CA, USA.

Data availability

The data that support the findings of this study were originated by and are the property of Flatiron Health, Inc, which has restrictions prohibiting the authors from making the dataset publicly available. Requests for data sharing by license or by permission for the specific purpose of replicating results in this manuscript can be submitted to publicationsdataaccess@flatiron.com .

Data access, responsibility, and analysis

J. S. Guadamuz had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Author contributions

Jenny S. Guadamuz, MSPH, PhD (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Software; Supervision; Validation; Visualization; Writing—original draft; Writing—review & editing; Resources); Xiaoliang Wang, MPH, PhD (Conceptualization; Formal analysis; Methodology; Validation; Writing—review & editing); Ivy Altomare, MD (Conceptualization; Writing—review & editing); Wendy Camelo Castillo, MD, MSc, PhD (Writing—review & editing); Somnath Sarkar, PhD (Writing—review & editing); Ronac Mamtani, MD, MSCE (Conceptualization; Methodology; Writing—review & editing); Gregory S. Calip, PharmD, MPH, PhD (Conceptualization; Methodology; Writing—review & editing).

Funding

This study was sponsored by Flatiron Health, Inc, which is an independent member of the Roche Group.

Conflicts of interest

Drs Guadamuz, Wang, Altomare, and Calip report current or previous employment with Flatiron Health, Inc, which is an independent member of the Roche group, and stock ownership in Roche. Dr Guadamuz has current grants from the Robert Wood Johnson Foundation. Dr Calip reports current employment with AbbVie.

References

1. Duma N, Vera Aguilera J, Paludo J, et al. Representation of minorities and women in oncology clinical trials: review of the past 14 years. J Oncol Pract. 2018;14(1):e1-e10. doi: 10.1200/JOP.2017.025288

2. Loree JM, Anand S, Dasari A, et al. Disparity of race reporting and representation in clinical trials leading to cancer drug approvals from 2008 to 2018. JAMA Oncol. 2019;5(10):e191870. doi: 10.1001/jamaoncol.2019.1870

3. Pittell H, Calip GS, Pierre A, et al. Racial and ethnic inequities in us oncology clinical trial participation from 2017 to 2022. JAMA Netw Open. 2023;6(7):e2322515. doi: 10.1001/jamanetworkopen.2023.22515

4. Royce TJ, Zhao Y, Ryals CA.. Improving diversity in clinical trials by using real-world data to define eligibility criteria. JAMA Oncol. 2023;9(4):455-456. doi: 10.1001/jamaoncol.2022.7170

5. Calip GS, Altomare IP, Guadamuz JS.. Evaluating external validity of oncology biosimilar safety studies. JAMA Netw Open. 2023;6(4):e235776. doi: 10.1001/jamanetworkopen.2023.5776

6. Consolidated Appropriations Act, 2023 (Section 3603), 117th Cong, 2nd Sess (2022). https://www.congress.gov/117/plaws/publ328/PLAW-117publ328.pdf

7. United States Food and Drug Administration. FDA Takes Important Steps to Increase Racial and Ethnic Diversity in Clinical Trials . FDA. https://www.fda.gov/news-events/press-announcements/fda-takes-important-steps-increase-racial-and-ethnic-diversity-clinical-trials . Published 2022. Accessed December 1, 2023.

8. Baumann J. Diversity in Clinical Trials at FDA Gets a Boost From New Law . Bloomberg Law. https://news.bloomberglaw.com/pharma-and-life-sciences/diversity-in-clinical-trials-at-fda-gets-a-boost-from-new-law . Published January 19, 2023. Accessed December 1, 2023.

9. Turner BE, Steinberg JR, Weeks BT, Rodriguez F, Cullen MR.. Race/ethnicity reporting and representation in US clinical trials: a cohort study. Lancet Reg Health—Am. 2022;11:100252. doi: 10.1016/j.lana.2022.100252

10. Perni S, Moy B, Nipp RD.. Disparities in phase 1 cancer clinical trial enrollment. Cancer. 2021;127(23):4464-4469. doi: 10.1002/cncr.33853

11. Riner AN, Girma S, Vudatha V, et al. Eligibility criteria perpetuate disparities in enrollment and participation of black patients in pancreatic cancer clinical trials. J Clin Oncol. 2022;40(20):2193-2202. doi: 10.1200/JClinOncol.21.02492

12. American Association for Cancer Research. AACR Cancer Disparities Progress Report 2022 . https://cancerprogressreport.aacr.org/wp-content/uploads/sites/2/2022/06/AACR_CDPR_2022.pdf . Published June 7, 2022.

13. Unger JM, Hershman DL, Albain KS, et al. Patient income level and cancer clinical trial participation. J Clin Oncol 2013;31(5):536-542. doi: 10.1200/JClinOncol.2012.45.4553

14. Unger JM, Gralow JR, Albain KS, Ramsey SD, Hershman DL.. Patient income level and cancer clinical trial participation: a prospective survey study. JAMA Oncol. 2016;2(1):137-139. doi: 10.1001/jamaoncol.2015.3924

15. Wang WJ, Ramsey SD, Bennette CS, Bansal A.. Racial disparities in access to prostate cancer clinical trials: a county-level analysis. JNCI Cancer Spectr. 2022;6(1):pkab093. doi: 10.1093/jncics/pkab093

16. Abbas A, Diaz A, Obeng-Gyasi S, et al. Disparity in clinical trial participation among patients with gastrointestinal cancer. J Am Coll Surg. 2022;234(4):589-598. doi: 10.1097/XCS.0000000000000129

17. Gross CP, Filardo G, Mayne ST, Krumholz HM.. The impact of socioeconomic status and race on trial participation for older women with breast cancer. Cancer. 2005;103(3):483-491. doi: 10.1002/cncr.20792

18. Fayanju OM, Ren Y, Thomas SM, et al. A case-control study examining disparities in clinical trial participation among breast surgical oncology patients. JNCI Cancer Spectr. 2020;4(2):pkz103. doi: 10.1093/jncics/pkz103

19. Eskander MF, Gil L, Beal EW, et al. Access denied: inequities in clinical trial enrollment for pancreatic cancer. Ann Surg Oncol. 2022;29(2):1271-1277. doi: 10.1245/s10434-021-10868-4

20. Khadraoui W, Meade CE, Backes FJ, Felix AS.. Racial and ethnic disparities in clinical trial enrollment among women with gynecologic cancer. JAMA Netw Open. 2023;6(12):e2346494. doi: 10.1001/jamanetworkopen.2023.46494

21. Birnbaum B, Nussbaum N, Seidl-Rathkopf K, et al. Model-assisted cohort selection with bias analysis for generating large-scale cohorts from the EHR for oncology research. 10.48550/arXiv.2001.09765. January 13, 2020, preprint: not peer reviewed.

22. Ma X, Long L, Moon S, Adamson B, Baxi S. Comparison of Population Characteristics in Real-World Clinical Oncology Databases in the US: Flatiron Health, SEER, and NPCR Ma X, Long L, Moon S, Adamson, Blythe JS, Baxi SS. medRxiv. Published 2020. 10.1101/2020.03.16.20037143. Accessed November 2, 2023.

23. US Census Bureau. American Community Survey (ACS). Census.gov . https://www.census.gov/programs-surveys/acs . Published December 1, 2023. Accessed February 10, 2024.

24. United States Department of Agriculture. Rural-Urban Commuting Area Codes . https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/ . Published 2023. Accessed November 11, 2023.

25. Health Resources and Services Administration. Find Shortage Areas. https://data.hrsa.gov/tools/shortage-area . Published 2023. Accessed November 2, 2023.

26. Brokamp C. DeGAUSS: decentralized geomarker assessment for multi-site studies. JOSS. 2018;3(30):812. doi: 10.21105/joss.00812

27. Zhang Q, Gossai A, Monroe S, Nussbaum NC, Parrinello CM.. Validation analysis of a composite real-world mortality endpoint for patients with cancer in the United States. Health Serv Res. 2021;56(6):1281-1287. doi: 10.1111/1475-6773.13669

28. White K, Borrell L.. Racial/ethnic residential segregation: framing the context of health risk and health disparities. Health Place. 2011;17(2):438-448. doi: 10.1016/j.healthplace.2010.12.002

29. Yu M, Tatalovich Z, Gibson JT, Cronin KA.. Using a composite index of socioeconomic status to investigate health disparities while protecting the confidentiality of cancer registry data. Cancer Causes Control CCC. 2014;25(1):81-92. doi: 10.1007/s10552-013-0310-1

30. Guadamuz JS, Wang X, Ryals CA, et al. Socioeconomic status and inequities in treatment initiation and survival among patients with cancer, 2011-2022. JNCI Cancer Spectr. 2023;7(5):pkad058. doi: 10.1093/jncics/pkad058

31. Gooley TA, Leisenring W, Crowley J, Storer BE.. Estimation of failure probabilities in the presence of competing risks: New representations of old estimators. Statist Med. 1999;18(6):695-706. doi: 10.1002/(sici)1097-0258(19990330)18:6<695::aid-sim60>3.0.co;2-o

32. Lê Cook B, McGuire TG, Zaslavsky AM.. Measuring racial/ethnic disparities in health care: Methods and practical issues. Health Serv Res. 2012;47(3 Pt 2):1232-1254. doi: 10.1111/j.1475-6773.2012.01387.x

33. Institute of Medicine (US) Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. Unequal Treatment. Washington, DC: National Academies Press (US; ); 2003. doi: 10.17226/12875.

34. Howe CJ, Bailey ZD, Raifman JR, Jackson JW.. Recommendations for using causal diagrams to study racial health disparities. Am J Epidemiol. 2022;191(12):1981-1989. doi: 10.1093/aje/kwac140

35. Yu Q, Wu X, Li B, Scribner RA.. Multiple mediation analysis with survival outcomes: With an application to explore racial disparity in breast cancer survival. Stat Med. 2019;38(3):398-412. doi: 10.1002/sim.7977

36. Yu Q, Li B.. Package ‘mma’: Multiple Mediation Analysis. 2022. https://cran.r-project.org/web/packages/mma/mma.pdf . Accessed October 7, 2024.

37. Leapman MS, Dinan M, Pasha S, et al. Mediators of racial disparity in the use of prostate magnetic resonance imaging among patients with prostate cancer. JAMA Oncol. 2022;8(5):687-696. doi: 10.1001/jamaoncol.2021.8116

38. Unger JM, Vaidya R, Hershman DL, Minasian LM, Fleury ME.. Systematic review and meta-analysis of the magnitude of structural, clinical, and physician and patient barriers to cancer clinical trial participation. JNCI J Natl Cancer Inst. 2019;111(3):245-255. doi: 10.1093/jnci/djy221

39. 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

40. Lin SC, Hammond G, Esposito M, Majewski C, Foraker RE, Joynt Maddox KE.. Segregated patterns of hospital care delivery and health outcomes. JAMA Health Forum. 2023;4(11):e234172. doi: 10.1001/jamahealthforum.2023.4172

41. Nipp RD, Hong K, Paskett ED.. Overcoming barriers to clinical trial enrollment. Am Soc Clin Oncol Educ Book. 2019;(39:):105-114. doi: 10.1200/EDBK_243729

42. Nong P, Raj M, Creary M, Kardia SLR, Platt JE.. Patient-reported experiences of discrimination in the US health care system. JAMA Netw Open. 2020;3(12):e2029650. doi: 10.1001/jamanetworkopen.2020.29650

43. Metastatic Breast Cancer Alliance. Black Experience of Clinical Trials and Opportunities for Meaningful Engagement Research Report . https://www.mbcalliance.org/wp-content/uploads/BECOME-Final-Report-FULL.pdf . Published 2022. Accessed January 16, 2024.

44. Braveman PA, Arkin E, Proctor D, Kauh T, Holm N.. Systemic and structural racism: definitions, examples, health damages, and approaches to dismantling. Health Aff (Millwood). 2022;41(2):171-178. doi: 10.1377/hlthaff.2021.01394

45. Mahal BA, Chen YW, Muralidhar V, et al. National sociodemographic disparities in the treatment of high-risk prostate cancer: do academic cancer centers perform better than community cancer centers? Cancer. 2016;122(21):3371-3377. doi: 10.1002/cncr.30205

46. Ma TM, Agarwal N, Mahal B, et al. Racial and ethnic disparities in use of novel hormonal therapy agents in patients with prostate cancer. JAMA Netw Open. 2023;6(12):e2345906. doi: 10.1001/jamanetworkopen.2023.45906

47. Fiala MA, Wildes TM, Vij R.. Racial disparities in the utilization of novel agents for frontline treatment of multiple myeloma. Clin Lymphoma Myeloma Leuk. 2020;20(10):647-651. doi: 10.1016/j.clml.2020.04.018

48. Alio AP, Wharton MJ, Fiscella K.. Structural racism and inequities in access to Medicaid-funded quality cancer care in the United States. JAMA Netw Open. 2022;5(7):e2222220. doi: 10.1001/jamanetworkopen.2022.22220

49. Himmelstein G, Ganz PA.. Distribution of cancer care resources across us hospitals by patient race and ethnicity. JAMA Oncol. 2024;10(1):134-137. doi: 10.1001/jamaoncol.2023.4952

50. Oyer RA, Hurley P, Boehmer L, et al. Increasing racial and ethnic diversity in cancer clinical trials: an American Society of Clinical Oncology and Association of Community Cancer Centers Joint Research Statement. J Clin Oncol Off J Am Soc Clin Oncol. 2022;40(19):2163-2171. doi: 10.1200/JClinOncol.22.00754

51. Hoos WA, James PM, Rahib L, Talley AW, Fleshman JM, Matrisian LM.. Pancreatic cancer clinical trials and accrual in the United States. J Clin Oncol Off J Am Soc Clin Oncol. 2013;31(27):3432-3438. doi: 10.1200/JClinOncol.2013.49.4823

52. Goodson N, Wicks P, Morgan J, Hashem L, Callinan S, Reites J.. Opportunities and counterintuitive challenges for decentralized clinical trials to broaden participant inclusion. Npj Digit Med. 2022;5(1):58-56. doi: 10.1038/s41746-022-00603-y

53. Dahne J, Hawk LW.. Health equity and decentralized trials. JAMA. 2023;329(23):2013-2014. doi: 10.1001/jama.2023.6982

54. Guadamuz JS, Wang X, Royce TJ, Calip GS.. Sociodemographic Inequities in Telemedicine Use Among US Patients Initiating Treatment in Community Cancer Centers During the Ongoing COVID-19 Pandemic, 2020-2022. J Clin Oncol Oncol Pract 2023;19(12):1206-1214., Published online September 25, OP.23.00144. doi: 10.1200/OP.23.00144

55. Roy M, Purington N, Liu M, Blayney DW, Kurian AW, Schapira L.. Limited English proficiency and disparities in health care engagement among patients with breast cancer. J Clin Oncol Oncol Pract. 2021;17(12):e1837-e1845. doi: 10.1200/OP.20.01093

56. National Institutes of Medicine. NIH Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research . https://grants.nih.gov/policy/inclusion/women-and-minorities/guidelines.htm . Published 2017. Accessed December 10, 2023.

57. American Society of Clinical Oncology. Insurance Coverage of Clinical Trials . ASCO. https://old-prod.asco.org/research-data/clinical-trials/insurance-coverage-clinical-trials . Published 2023. Accessed December 10, 2023.

58. Unger JM, Barlow WE, Martin DP, et al. Comparison of survival outcomes among cancer patients treated in and out of clinical trials. JNCI J Natl Cancer Inst. 2014;106(3):dju002. doi: 10.1093/jnci/dju002

59. Branham DK, Finegold K, Chen L, et al. Trends in missing race and ethnicity information after imputation in healthcare.gov marketplace enrollment data, 2015-2021. JAMA Netw Open. 2022;5(6):e2216715. doi: 10.1001/jamanetworkopen.2022.16715

Associated Data

Supplementary Materials

Data Availability Statement

The data that support the findings of this study were originated by and are the property of Flatiron Health, Inc, which has restrictions prohibiting the authors from making the dataset publicly available. Requests for data sharing by license or by permission for the specific purpose of replicating results in this manuscript can be submitted to publicationsdataaccess@flatiron.com .