Background
The healthcare and cancer care systems are no exception to the racial disparities that pervade United States society. Cancer care systems are deeply impacted by interpersonal, organizational, and structural and systemic racism, and this has been shown to impact patient outcomes.1 This is particularly apparent in the rates of guideline-adherent cancer care stratified by race.2 Myriad independent studies have found that adherence to NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) improves care delivery and outcomes for patients with cancer, including improved rates of survival for patients with colon, ovarian, gastric, nasopharyngeal, and pancreatic cancers; decreased locoregional recurrence of melanoma; and improved pain control.3–8 Despite this evidence, racial disparities in accessing guideline-adherent cancer care persist. In particular, a wide body of research has found that cancer care for Black patients is dramatically more likely to diverge from recommended treatment guidelines compared with care for White patients, with some studies finding this to be true even when controlling for insurance status.2,9,10 Despite some progress in recent years, Black Americans still have the highest death rate and lowest survival rate of any racial or ethnic group for most cancers.11
The root causes of racial disparities in access to guideline-adherent cancer care, and relatedly cancer outcomes, are complex, multifactorial, and not rooted in biology. Contributing factors to racial disparities in care delivery include implicit and explicit bias in care delivery, lack of representation of people of color in the oncology workforce, and homogenous research populations that are not representative of the larger community. Systemic barriers include lack of insurance and underinsurance, costs of cancer treatment and associated costs of care, disparate access to clinical trials, and social determinants of health, including exposure to environmental hazards, access to housing, childcare, and economic injustices.12 Although it is well established that racial disparities in cancer risk, screening and care access, quality, and outcomes exist, there are limited studies exploring the experiences of patients and physicians in relation to bias in care delivery.13 Studies have evaluated oncologists’ rates of implicit bias and found that oncologists scoring higher on implicit bias tests had shorter interactions with Black patients and were viewed as less patient-centered and supportive by Black patients.14,15 Additionally, some studies have evaluated the role of bias and stereotyping in referral to clinical trials and have concluded that racial bias and assumptions do play a role in low rates of clinical trial referral.16
To better understand patient experiences of care quality and bias in care, the American Cancer Society Cancer Action Network (ACS CAN), National Comprehensive Cancer Network (NCCN), and the National Minority Quality Forum (NMQF) convened the Elevating Cancer Equity initiative. This initiative aims to both better understand the causes and identify and promote actionable solutions through the combined experience and expertise of racial and ethnic minority patients, caregivers, and national experts. National surveys of (1) patients with cancer and caregivers of patients with cancer, and (2) oncologists were conducted with the primary aim of understanding the varying perspectives of bias in the receipt of guideline-concordant cancer care across race and ethnicity groups. Results of these surveys are published herein, and were also used to inform a multistakeholder expert working group charged with developing actionable policy and practice change recommendations, which will be detailed in a subsequent report.
Methods
Despite a substantial body of research demonstrating the existence of disparities in cancer care access, quality, and outcomes, there is a need for additional research to understand the individual patient and caregiver experience of bias and discrimination in care. To address this, ACS CAN, NCCN, and NMQF contracted with a national survey research firm, Public Opinion Strategies, to launch 2 electronic surveys: 1 examining patient and caregiver experiences with cancer care, and 1 examining oncologist perspectives on bias and discrimination in cancer care. Participants were identified using opt-in online panels, and both surveys were distributed in December 2020. As is standard practice in online surveys, participants were compensated for their time to complete the survey. This project was determined to be exempt from Institutional Review Board (IRB) review by Pearl IRB, an independent IRB accredited by the Association for the Accreditation of Human Research Protection Programs.
A core national sample of 600 patients with cancer and caregivers of patients with cancer completed surveys, including 300 patients and 300 caregivers. The demographics of the core sample participants are reflective of the national population, which was accomplished by setting quotas by region, state, gender, age, and ethnicity using data from the American Cancer Society,17 U.S. Census Bureau Adult Population Estimates,18 and AARP and National Alliance for Caregiving19 (Table 1). Additionally, oversampling of African American/Black (AA/B) and Hispanic/Latino (H/L) patients/caregivers was used to ensure appropriate sample sizes for characterization of data by ethnicity. Oversampling resulted in an additional 157 AA/B and 152 H/L participants, for a total of 227 AA/B and 238 H/L patients or caregivers. The core national survey plus oversampling resulted in a total of 909 interviews. Overall, 27,872 patients and caregivers were invited to participate; 28% qualified by meeting the inclusion criteria, and of those qualified, 11.6% completed the survey. The oversampled interviews are reported as part of the national core data but weighted along with the existing AA/B and H/L interviews from the core to the appropriate percentage representative of patients and caregivers, respectively. The confidence interval at the 95% confidence level for a survey with a sample size of 600 is ±4.56%; for a sample size of 227 it is ±7.42%, and for a sample size of 238 it is ±7.24%.
Characteristics of Survey Participants and National Demographics
A national survey was completed by 208 practicing oncologist participants. Overall, 1,435 oncologists were invited to participate, 43% of whom met the criteria for inclusion; of those qualified, 33.7% completed the survey. Oncologist participants were queried on patient population served: 49% reported serving a patient population that was <35% non-White, and 43% reported serving a patient population that was ≥35% non-White, with the median percentage of non-White patients served being 30%. To ensure the survey sample was representative of practicing oncologists nationally, quotas were set by region, state, gender, and ethnicity using demographic data from ASCO20 and the Kaiser Family Foundation21 (Table 2). Oncologists were queried on their typical care practices, perceptions of bias in cancer care at large, and feelings on how bias may or may not impact their own delivery of care. The confidence interval at the 95% confidence level for a survey sample size of 208 is ±7.75%.
Characteristics of Oncologist Survey Participants and National Demographics
Results
Patients and Caregivers
Respondent Characteristics
Patient respondents were 73% White (W) and 27% non-White, and caregiver respondents were 71% W and 29% non-White (Table 1). Patients currently in active cancer treatment comprised 53% of respondents, with these numbers being higher among AA/B and H/L respondents (65% and 73%, respectively). The remainder of patient/survivor respondents had all received cancer treatment within the past 5 years, or had just been diagnosed and had not begun treatment. A total of 24 different cancer types were represented among patients, with the most common types being breast (27%), prostate (25%), lung (19%), ovarian/uterine (12%), cervical (9%), and colorectal (7%). Caregivers were currently caring for or in the past 2 years cared for a loved one with cancer. Cancer types most prevalent among those being cared for by caregiver respondents included breast (23%), lung (18%), prostate (17%), colorectal (11%), multiple myeloma (6%), and bladder (6%).
Perceptions of Bias and Discrimination in Cancer Care
AA/B or H/L patients were substantially more likely to report that the healthcare system treats people unfairly based on their racial or ethnic background, and that racial bias occurs often or very often when a patient and doctor are of different racial/ethnic backgrounds (supplemental eAppendix 1, available with this article at JNCCN.org). There were also dramatic racial differences between W, AA/B, and H/L respondents in their response to whether they agree non-White patients with cancer are more likely than W patients to receive poor quality care (W, 42%; AA/B, 73%; H/L, 69%). Additionally, older people (age ≥65 years) were less likely to believe racial bias and discrimination in the healthcare system and disparities in care quality occurred often, whereas younger people (age 18–34 years) were more likely to believe it does occur often. AA/B or H/L patients and caregivers were also more likely to believe that non-White patients receive poorer communication and worse medical care, and have worse outcomes. However, a majority of respondents across all racial subgroups believed W and non-White patients receive the same quality of care and communication (supplemental eAppendix 1).
Experiences of Cancer Care
Participants were asked whether they or their loved one had experienced a series of 14 care practices considered “best practice” for shared decision-making, patient education, and engagement (Table 3). There were slight variations in responses across race. However, younger patients and patients treated at cancer centers were more likely to report their physician doing all 14 “best practices” than older patients (age 18–34 years, 26%; 35–44 years, 23%; 45–54 years, 20%; 55–64 years, 14%; ≥65 years, 15%) and patients treated in an office or hospital setting (office, 14%; hospital, 16%; cancer center, 25%).
Oncologist Care Practices
When patients and caregivers were asked about negative care experiences, differences across race were observed. Overall, AA/B and H/L patients and caregivers were more likely to report at least one negative care experience (Table 4 and Figure 1). More specifically, AA/B and H/L respondents were more likely than W respondents to report that they felt their care team did not answer questions in an easy-to-understand manner (W, 29%; AA/B, 44%; H/L, 46%) and that they did not feel comfortable discussing their questions with their care team (W, 25%; AA/B, 34%; H/L, 35%). Additionally, AA/B and H/L patients and caregivers were also more likely to report that they felt their care team made overall assumptions about them (W, 11%; AA/B, 38%; H/L, 30%) or, more precisely, assumptions about their financial circumstances (W, 23%; AA/B, 37%; H/L, 34%). Finally, AA/B and H/L patients were more likely to report that they felt they received poorer quality of care based on their race or ethnicity (W, 9%; AA/B, 31%; H/L, 29%).
Patient/Caregiver Self-Report of Negative Experiences With Oncologist or Cancer Care Team
Patient/Caregiver report of experiencing negative situations with oncologist/cancer care team.
Citation: Journal of the National Comprehensive Cancer Network 20, 10; 10.6004/jnccn.2022.7051
The survey also attempted to better understand patient and caregiver care experiences by asking open-ended questions about their biggest barriers and challenges with their care team and what they would have liked to have been done differently. Common themes for barriers and challenges across all racial groups included communication, though AA/B and H/L patients were more likely to report communication challenges. Additionally, AA/B participants were more likely to report challenges regarding mental health support, and H/L participants were more likely to report challenges regarding travel, language barriers, and difficulty finding the right treatment (supplemental eAppendix 2).
Oncologists
Respondent Characteristics
Due to the lack of diversity among oncologists,22 Public Opinion Strategies was unable to oversample for diverse representation among oncologists. Respondents were representative of the oncology field, with demographics reflecting 68% men and 32% women, spread representatively across the United States. Respondents were 58% W and 42% non-White, but due to a lack of racial representation among oncologists, we were unable to oversample for adequate representation of AA/B and H/L oncologists. Respondent specialty areas were 53% medical oncology, 44% hematology/oncology, 2% radiation oncology, and 1% surgical oncology. In examining office setting, 30% of respondents practiced in a physician’s office or single-specialty group practice, 17% in a multispecialty group practice, 37% in a hospital/health system practice, and 16% in a cancer treatment center.
Perceptions of Bias in Care
A slight majority of oncologists (54%) reported that the healthcare system treats people unfairly based on their racial or ethnic background often or very often, with differences across gender and race. Male oncologists were more likely to believe this than female oncologists, and younger oncologists (age <40 years) were more likely to believe this than older oncologists (age ≥40 years). Additionally, geographic differences exist, with oncologists practicing in the Northeast and Western United States more likely to believe this to be true than those practicing in the Midwest and Southern United States, and those practicing in urban areas are more likely to believe this to be true than those practicing in suburban and rural areas. An even smaller majority (51%) of oncologists believe racial bias occurs when a patient and doctor are of different racial or ethnic backgrounds, again with men more likely than women and younger oncologists more likely than older oncologists to believe this to be true.
A majority of oncologists (62%) agreed that non-White patients are more likely to receive poor quality care than W patients, and also reported that non-White patients have worse health outcomes (62%) than W patients. However, when oncologists were asked whether they believed all patients received the same, better, or worse communication from their oncologist, only 38% believed that non-White patients with cancer received worse communication than W patients with cancer. It is possible that this disparity between beliefs about worse outcomes versus worse care and communication may be due to a belief that social determinants and other extenuating factors play a large role in care outcomes.
Oncologists were also asked how bias may play a role in their own care delivery. In response to the question, “Thinking about you and your patients, is it possible that you have had unintentional bias in treating your patients of different racial or ethnic backgrounds?”, 44% of all oncologists reported that this was possible and 56% reported it was not possible. Men were more likely than women to report it as possible and oncologists aged <40 years were more likely to report it as possible than those aged ≥40 years. There were also some variations among practice setting, number of years in practice, practice size, and urban versus rural settings. A key tenet of antiracism efforts is the acceptance that everyone has existing biases and a willingness to engage in self-reflection to advance practice improvement.23 With 54% of W oncologists reporting that it was not possible that they ever unintentionally had bias in the provision of care, it is important to consider whether this is because they truly have never acted with bias, or whether they are not yet reflecting upon the potential for bias in their practice. This finding may indicate a need for additional antibias and antiracist training in medical education.
Discussion
These survey results highlight a number of areas for improvement to address patient experiences of racial bias, discrimination, and disparities in cancer care. The dearth of AA/B and H/L oncologists raises the urgent need for appropriate racial representation within oncology. Policymakers, health system leadership, and educational institutions must work together to incentivize and promote appropriate racial representation in the oncology professional workforce.
Among oncologist respondents, 56% reported that it was not possible they ever had unintentional bias in their care delivery for their patients. Individual biases are essentially universal and influence our behavior both consciously and unconsciously. Additionally, multiple studies have found that healthcare professionals have similar rates of bias to the general population.24 Our survey’s finding may indicate a lack of self-reflection and awareness among some oncologists, indicating a need for comprehensive implicit bias training among oncologists. This also points to the need for the incorporation of antiracism education within medical school curricula and continuing education activities.
These survey results also highlight several key areas for quality improvement. AA/B and H/L patients were significantly more likely to report at least one negative care experience than W patients. In particular, AA/B or H/L patients were significantly more likely to report challenges with communication from their provider. This may indicate a need for oncologists and other professional healthcare staff to examine the language used as well as the cultural appropriateness of their messaging. Additionally, across all races, both physicians and patients reported low rates of clinical trials being discussed as an option. Health systems and practices should move toward asking all patients in cancer treatment if they would be interested in learning more about clinical trials. The survey results indicate numerous areas for improvement to address bias in oncology care delivery, patient communication, and workforce diversity.
This study includes several limitations, primarily in the areas of appropriate racial representation and the potential pitfalls of self-reporting. In particular, as noted earlier, we were unable to oversample for appropriate racial representation among oncologists due to a lack of diversity within the field and a limited timeline for data collection. As a result, we are unable to determine how W and AA/B or H/L oncologist respondents may differ in their perceptions of bias and discrimination in cancer care. Additionally, we did not query patient and caregiver respondents on the race of their oncologist, and therefore we are unable to extrapolate how appropriate racial representation in oncology may impact bias in care delivery. Similarly, among the patients and caregivers surveyed, we were unable to oversample for Asian, Pacific Islander, and Indigenous Americans, and therefore the data presented are primarily focused on the differences among patients and caregivers who are W, AA/B, and H/L. Finally, we recognize the selection bias and limitations in generalizability when using opt-in online panels of participants who have agreed to participate in survey work. We also acknowledge the limitations of self-reported data which may present a number of challenges, including the potential for attribution and social desirability bias. This survey did not explore systemic challenges and focused more on individual experiences with the healthcare system. Structural racism and inequities within health systems have significant impacts on health and cancer outcomes and will be addressed in a subsequent article.
Conclusions
Despite a significant body of research establishing disparities in cancer care across race and ethnicity, there is a dearth of research examining both patient experiences with and oncologist perceptions of disparities in cancer care. This survey queried patients, caregivers, and oncologists about their care experiences and perceptions and found several areas for improvement. These areas include a need for improved racial representation in the oncology professional workforce, improved implicit bias training, improved clinical trial recruitment efforts, and improved communication strategies across all racial groups. Further research into best practices and actionable strategies to address these challenges is warranted.
Acknowledgments
The authors would like to thank Gary Puckrein, PhD, CEO of the National Minority Quality Forum, and Lisa Lacasse, MBA, President of the American Cancer Society Cancer Action Network, for their dedicated leadership and thoughtful guidance on the Elevating Cancer Equity initiative and resulting survey. The authors appreciate and thank K. Robin Yabroff, PhD, MBA, for helpful and insightful feedback on the draft manuscript.
References
- 1.↑
Esnaola NF, Ford ME. Racial differences and disparities in cancer care and outcomes: where’s the rub? Surg Oncol Clin N Am 2012;21:417–437, viii.
- 2.↑
Fang P, He W, Gomez D, et al. Racial disparities in guideline-concordant cancer care and mortality in the United States. Adv Radiat Oncol 2018;3:221–229.
- 3.↑
Erickson Foster H, Velasco JM, Hieken TJ. Adverse outcomes associated with noncompliance with melanoma treatment guidelines. Ann Surg Oncol 2008;15:2395–2402.
- 4.↑
Visser BC, Ma Y, Zak Y, et al. Failure to comply with NCCN guidelines for the management of pancreatic cancer compromises outcomes. HPB (Oxford) 2012;14:539–547.
- 5.↑
Bristow RE, Powell MA, Al-Hammadi N, et al. Disparities in ovarian cancer care quality and survival according to race and socioeconomic status. J Natl Cancer Inst 2013;105:823–832.
- 6.↑
Bristow RE, Chang J, Ziogas A, et al. High-volume ovarian cancer care: survival impact and disparities in access for advanced-stage disease. Gynecol Oncol 2014;132:403–410.
- 7.↑
Mearis M, Shega JW, Knoebel RW. Does adherence to National Comprehensive Cancer Network guidelines improve pain-related outcomes? An evaluation of inpatient cancer pain management at an academic medical center. J Pain Symptom Manage 2014;48:451–458.
- 8.↑
Schwam ZG, Sosa JA, Roman S, et al. Receipt of care discordant with practice guidelines is associated with compromised overall survival in nasopharyngeal carcinoma. Clin Oncol (R Coll Radiol) 2016;28:402–409.
- 9.↑
Ubbaonu C, Chang J, Ziogas A, et al. Disparities in the receipt of National Comprehensive Cancer Network (NCCN) guideline adherent care in triple-negative breast cancer (TNBC) by race/ethnicity, socioeconomic status, and insurance type [abstract]. J Clin Oncol 2020;38(Suppl):Abstract 1080.
- 10.↑
Blom EF, Ten Haaf K, Arenberg DA, et al. Disparities in receiving guideline-concordant treatment for lung cancer in the United States. Ann Am Thorac Soc 2020;17:186–194.
- 11.↑
DeSantis CE, Miller KD, Goding Sauer A, et al. Cancer statistics for African Americans, 2019. CA Cancer J Clin 2019;69:211–233.
- 12.↑
Winkfield KM, Regnante JM, Miller-Sonet E, et al. Development of an actionable framework to address cancer care disparities in medically underserved populations in the United States: expert roundtable recommendations. JCO Oncol Pract 2021;17:e278–293.
- 13.↑
Wang X, Huang Y, Li L, et al. Assessment of performance of the Gail model for predicting breast cancer risk: a systematic review and meta-analysis with trial sequential analysis. Breast Cancer Res 2018;20:18.
- 14.↑
Penner LA, Dovidio JF, Gonzalez R, et al. The effects of oncologist implicit racial bias in racially discordant oncology interactions. J Clin Oncol 2016;34:2874–2880.
- 15.↑
Smedley BD, Stith AY, Nelson AR, et al. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: The National Academies Press; 2003.
- 16.↑
Niranjan SJ, Martin MY, Fouad MN, et al. Bias and stereotyping among research and clinical professionals: perspectives on minority recruitment for oncology clinical trials. Cancer 2020;126:1958–1968.
- 17.↑
American Cancer Society. Cancer facts & figures 2020. Accessed August 8, 2022. Available at: https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2020.html
- 18.↑
AARP and National Alliance for Caregiving. Caregiving in the United States 2020. Washington, DC: AARP. May 2020. Accessed August 8, 2022. Available at: https://doi.org/10.26419/ppi.00103.001
- 19.↑
United States Census Bureau. ACS demographic and housing estimates. Accessed August 8, 2022. Available at: https://data.census.gov/cedsci/table?q=age&g=0100000US%240400000&tid=ACSDP5Y2020.DP05
- 21.↑
Professionally active specialist physicians by field. KFF website. Published February 7, 2022. Accessed September 13, 2022. Available at: https://www.kff.org/other/state-indicator/physicians-by-specialty-area/
- 22.↑
American Society of Clinical Oncology. Facts & figures: diversity in oncology. Accessed July 28, 2021. Available at: https://www.asco.org/news-initiatives/current-initiatives/diversity-oncology-initiative/facts-figures
- 23.↑
Singh AA. Handout from the Racial Healing Handbook: practical activities to help you challenge privilege, confront systemic racism, and engage in collective healing. Accessed August 8, 2022. Available at: https://nmaahc.si.edu/sites/default/files/downloads/resources/racialhealinghandbook_p87to94.pdf
- 24.↑
FitzGerald C, Hurst S. Implicit bias in healthcare professionals: a systematic review. BMC Med Ethics 2017;18:19.
