Racism Is a Modifiable Risk Factor: Relationships Among Race, Ethnicity, and Colorectal Cancer Outcomes

Abstract

Colorectal cancer (CRC) is the second overall leading cause of cancer incidence and cancer-related mortality in the United States.1Siegel R.L. Miller K.D. Fuchs H.E. Jemal A. Cancer statistics, 2021.CA Cancer J Clin. 2021; 71: 7-33Google Scholar The burden of CRC varies across racial and ethnic groups, with differences between Black people and White people especially well documented. Compared to White people, Black people have a higher incidence of CRC,2DeSantis C.E. Miller K.D. Goding Sauer A. et al.Cancer statistics for African Americans, 2019.CA Cancer J Clin. 2019; 69: 211-233Google Scholar tend to be diagnosed at a younger age,2DeSantis C.E. Miller K.D. Goding Sauer A. et al.Cancer statistics for African Americans, 2019.CA Cancer J Clin. 2019; 69: 211-233Google Scholar have shorter stage-specific survival,3Rutter C.M. Johnson E.A. Feuer E.J. et al.Secular trends in colon and rectal cancer relative survival.J Natl Cancer Inst. 2013; 105: 1806-1813Google Scholar and have higher overall CRC mortality.4Dieleman J.L. Chen C. Crosby S.W. et al.US health care spending by race and ethnicity, 2002–2016.JAMA. 2021; 326: 649-659Google Scholar These racial disparities are often interpreted, either explicitly or implicitly, as evidence of elevated prevalence of CRC risk factors among Black people that are due to genetic predisposition or general health behaviors, such as diet and exercise. This interpretation of the data ignores the contribution of racial disparities in access to high-quality care and other resources affecting health behaviors that are amenable to more immediate policy action and overstates the ability of statistical approaches to account for these differences. As a result, policy-relevant modifiable characteristics that are key to addressing inequities are either minimized or overlooked. In this commentary, we discuss the importance of how we define race and think about characteristics related to race, including structural racism, when interpreting statistical models used to explore racial disparities in CRC.Despite the worse CRC outcomes experienced by Black people, there is very limited evidence that Black people have a greater prevalence of biological risk factors than White people. Estimation of racial differences in CRC outcomes is challenging because every outcome can be affected by screening. Screening prevents CRC by removing precursor lesions, so populations with higher screening levels have lower CRC incidence. Patients who have previously undergone colonoscopy tend to have fewer precursor lesions at subsequent examinations, and both precursor lesions and asymptomatic CRCs that are detected tend to be less advanced. Previously screened patients may also tend to have relatively more proximal lesions if tests are better able to detect distal lesions. In addition, by selectively preventing CRC after age 50 years, screening can shift the distribution of age at CRC detection to younger ages. Thus, differences in CRC screening participation and follow-up of abnormal results can affect the differences we observe in CRC outcomes for Black people and White people.Black patients and White patients who have similar screening histories, especially patients who are screening-naive, have similar prevalence of precursor lesions,5Rutter C.M. Knudsen A.B. Lin J.S. Bouskill K.E. Black and white differences in colorectal cancer screening and screening outcomes: a narrative review.Cancer Epidemiol Biomarkers Prev. 2021; 30: 3-12Google Scholar,6Stemboroski L. Samuel J. Alkaddour A. et al.Characteristics of serrated adenomas in non-Hispanic whites and African Americans undergoing screening colonoscopy.Cureus. 2021; 13: 7-12Google Scholar a crucial argument against differences in underlying risk, including risk attributable to behavioral factors such as smoking, diet, exercise, and obesity. Still, there are concerns that Black patients may have more aggressive disease based on findings that they are more likely than White patients to be diagnosed with interval CRC.7Fedewa S.A. Flanders W.D. Ward K.C. et al.Racial and ethnic disparities in interval colorectal cancer incidence a population-based cohort study.Ann Intern Med. 2017; 166: 857-866Google Scholar Higher incidence of early-onset CRC in Black patients, diagnosed before age 50 years, had also raised concerns about more aggressive disease, although the incidence of early-onset CRC is now similar in Black people and White people because of increases in incidence among White people.8Murphy C.C. Wallace K. Sandler R.S. Baron J.A. Racial disparities in incidence of young-onset colorectal cancer and patient survival.Gastroenterology. 2019; 156: 958-965Google Scholar If there are racial differences in aggressiveness, these could be partially driven by biological factors, including undiscovered differences in genetic factors. Heritable genetic factors account for only 5% of CRC in White patients. A similar fraction of CRC in Black patients is likely attributable to heritable genetic factors, although we do not have a reliable estimate, in part because Black patients have lower rates of referral for genetic testing.9Muller C. Lee S.M. Barge W. et al.Low referral rate for genetic testing in racially and ethnically diverse patients despite universal colorectal cancer screening.Clin Gastroenterol Hepatol. 2018; 16: 1911-1918Google Scholar,10Augustus G.J. Ellis N.A. Colorectal cancer disparity in African Americans: risk factors and carcinogenic mechanisms.Am J Pathol. 2018; 188: 291-303Google Scholar Genome-wide association studies have identified genetic variants that increase risk for sporadic CRC and have reported racial differences in some risk variants.10Augustus G.J. Ellis N.A. Colorectal cancer disparity in African Americans: risk factors and carcinogenic mechanisms.Am J Pathol. 2018; 188: 291-303Google Scholar Polygenic risk scores hold promise for understanding variability in underlying risk and can identify people at high risk who could benefit from earlier screening, but genotyping studies used to develop risk scores have focused exclusively on participants of European descent.11Archambault A.N. Su Y.-R. Jeon J. et al.Cumulative burden of colorectal cancer–associated genetic variants is more strongly associated with early-onset vs late-onset cancer.Gastroenterology. 2020; 158: 1274-1286Google Scholar Another concern is that Black patients are more likely to develop proximal adenomas and cancers, but the evidence of these associations is mixed.5Rutter C.M. Knudsen A.B. Lin J.S. Bouskill K.E. Black and white differences in colorectal cancer screening and screening outcomes: a narrative review.Cancer Epidemiol Biomarkers Prev. 2021; 30: 3-12Google Scholar A recent study showing that Black patients have accelerated epigenetic aging in the proximal colon, but White patients do not, suggests a plausible mechanism for increased risk of proximal lesions.12Acuna-Villaorduna A.R. Lin J. Kim M. Goel S. Racial/ethnic disparities in early-onset colorectal cancer: implications for a racial/ethnic-specific screening strategy.Cancer Med. 2021; 10: 2080-2087Google Scholar If there are racial differences in aggressiveness and location of disease, these could be partially driven by biological factors, including genetic factors and downstream biological consequences of systemic racism and social injustice.13Davidson K.W. Mangione C.M. Barry M.J. et al.Actions to transform US Preventive Services Task Force methods to mitigate systemic racism in clinical preventive services.JAMA. 2021; 326: 2405-2411Google Scholar A fundamental issue is that we need better information about which (or any) aspects of the CRC disease process vary across racial groups and the extent and mechanisms of variation.Evidence that differences in the CRC disease process can explain observed racial disparities in CRC burden is either null or inconclusive, but there is clear evidence of racial disparities in the timeliness and quality of CRC care, from early detection to treatment. For many years, Black patients have been less likely to be screened for CRC than White patients,14Rutter CM, de Lima PN, Lee JK, Ozik J. Too good to be true? Evaluation of colonoscopy sensitivity assumptions used in policy models [published online January 28, 2022]. Cancer Epidemiol Biomarkers Prev https://doi.org/10.1158/1055-9965.EPI-21-1001.Google Scholar and when screened, they may be less likely to receive endoscopic tests at high-quality facilities.15Laiyemo A.O. Doubeni C. Pinsky P.F. et al.Race and colorectal cancer disparities: health-care utilization vs different cancer susceptibilities.J Natl Cancer Inst. 2010; 102: 538-546Google Scholar Quality of care matters—colonoscopy is less effective at reducing CRC mortality when patients are examined by providers with lower adenoma detection rates.16Corley D.A. Jensen C.D. Marks A.R. et al.Adenoma detection rate and risk of colorectal cancer and death.N Engl J Med. 2014; 370: 1298-1306Google Scholar Once diagnosed, Black patients experience delays in treatment initiation17Bui A. Yang L. Myint A. May F.P. Race, ethnicity, and socioeconomic status are associated with prolonged time to treatment after a diagnosis of colorectal cancer: a large population-based study.Gastroenterology. 2021; 160: 1394-1396Google Scholar and are less likely to receive curative treatment.18Tramontano A.C. Chen Y. Watson T.R. et al.Racial/ethnic disparities in colorectal cancer treatment utilization and phase-specific costs, 2000–2014.PLoS One. 2020; 15e0231599Google Scholar These differences are driven, in part, by unequal access to high-quality care. Studies that examine patients with equal access to care find that Black patients have similar or greater screening participation than White patients19May F.P. Yano E.M. Provenzale D. et al.Race, poverty, and mental health drive colorectal cancer screening disparities in the Veterans Health Administration.Med Care. 2019; 57: 773-780Google Scholar,20Changoor N.R. Pak L.M. Nguyen L.L. et al.Effect of an equal-access military health system on racial disparities in colorectal cancer screening.Cancer. 2018; 124: 3724-3732Google Scholar and similar time to treatment after diagnosis.21Eaglehouse Y.L. Georg M.W. Shriver C.D. Zhu K. Racial comparisons in timeliness of colon cancer treatment in an equal-access health system.J Natl Cancer Inst. 2020; 112: 410-417Google Scholar In addition, when Black patients and White patients have the same access to care, they have similar clinical outcomes.22Berry J. Caplan L. Davis S. et al.A Black-White comparison of the quality of stage-specific colon cancer treatment.Cancer. 2010; 116: 713-722Google ScholarEvidence is built based on statistical models that estimate the effect of race on CRC outcomes and relies on our interpretation of these estimates. Although it is common practice to control for race in regression models of CRC incidence and outcomes, interpreting race coefficients requires care and precision that are typically lacking.23Kaufman J.S. Race: ritual, regression, and reality.Epidemiology. 2014; 25: 485-487Google Scholar Race is a social and power construct that captures the historical context that produced the stratified society that we live in today and drives contemporary characteristics of different racial groups, which include phenotypes (especially skin color), culture, wealth, income, education, access to health care, neighborhood characteristics related to healthy food access and walkability, and experiences of structural and interpersonal racism. When included in regression models, race variables thus become an imperfect proxy for a myriad of between-group differences. Estimated race coefficients capture a conglomeration of unmeasured and uncertain influences that vary depending on the other factors in the model. For example, racial differences in average age at CRC diagnosis may reflect differences in the population age structure, which are influenced by differences in all-cause mortality; thus, age-specific incidence rates provide a better comparison of risk of early-onset CRC. Prior screening, which influences CRC outcomes, is often included in models that include race as a predictor to explore the magnitude of racial disparities while controlling for prior screening. In this sense, the estimated effect of race has meaning only in the context of the other factors included in the model. These other factors will encompass only variables that have been observed and measured in the specific study at hand, which often differ widely among studies. However, the implied differences in interpretation are rarely acknowledged. Interpretation of the estimated race coefficient also depends on the specific population under study, the ascertainment and parameterization of racial categories, and on how heterogenous the sample is with respect to factors related to race and CRC outcomes, such as access to care. Indeed, analyzing homogeneous samples that include only people who all have equal access to care and similar socioeconomic status is a fundamental approach used to separate the effects of race and access to care on CRC outcomes. In summary, when interpreting analyses that examine the effect of race, it is critical to think about what race is measuring; the mechanisms through which race might affect outcomes; and which characteristics we need to control for to make the desired (or claimed) comparisons, including unmeasured or unobserved characteristics. Even if we could hypothetically measure and adjust for all meaningful differences among racial groups, we might conclude that “the adjustment is a kind of fiction”24Kaufman J.S. Dissecting disparities.Med Decis Making. 2008; 28: 9-11Google Scholar—a fiction because it is built on comparison of people in different race/ethnicity categories as if all of the factors adjusted for were similar across racial groups in spite of observed differences.Studies that demonstrate racial differences in CRC outcomes rarely adjust for differential access to wealth, income, education, high-quality health care, and healthy neighborhood environments, which are key drivers of racial differences in outcomes. In summary, when evaluating the role of race in health outcomes, including CRC outcomes, it is critical to do the following:•Clearly describe how race was measured and categorized while adhering to inclusive reporting standards.13Davidson K.W. Mangione C.M. Barry M.J. et al.Actions to transform US Preventive Services Task Force methods to mitigate systemic racism in clinical preventive services.JAMA. 2021; 326: 2405-2411Google Scholar•Recognize that estimated effects of race and ethnicity are highly dependent on other variables that are also included in the model. The degree of adjustment required to draw causal inference may not be possible and, if possible, may not be realistic. Causal models focus on characteristics that can be modified.23Kaufman J.S. Race: ritual, regression, and reality.Epidemiology. 2014; 25: 485-487Google Scholar Racism—structural, interpersonal, and internalized—is a modifiable risk factor.•Identify factors that are amenable to policy intervention, such as access to care and quality of care, that are modifiable drivers of CRC outcome disparities that are rooted in structural racism.•Recognize and acknowledge structural racism and social injustices as key drivers of poorer access to and receipt of high-quality care across all steps in the CRC continuum to enable new solutions to be imagined.•Propose solutions, such as investments in the people, studies, policies, neighborhoods, and processes, that can lead to more accessible and culturally competent care—care that can be delivered by health systems that foster community trust—and that achieve equal participation and timeliness of screening, treatment, clinical trials enrollment, and survivorship care.In the past 2 decades, CRC incidence has decreased as screening has increased, and racial disparities have generally narrowed.25May F.P. Glenn B.A. Crespi C.M. et al.Decreasing black-white disparities in colorectal cancer incidence and stage at presentation in the United States.Cancer Epidemiol Biomarkers Prev. 2017; 26: 762-768Google Scholar Nevertheless, disparities persist. Solutions to racial inequities in CRC outcomes require that we name structural racism and study its role as a causal factor. Colorectal cancer (CRC) is the second overall leading cause of cancer incidence and cancer-related mortality in the United States.1Siegel R.L. Miller K.D. Fuchs H.E. Jemal A. Cancer statistics, 2021.CA Cancer J Clin. 2021; 71: 7-33Google Scholar The burden of CRC varies across racial and ethnic groups, with differences between Black people and White people especially well documented. Compared to White people, Black people have a higher incidence of CRC,2DeSantis C.E. Miller K.D. Goding Sauer A. et al.Cancer statistics for African Americans, 2019.CA Cancer J Clin. 2019; 69: 211-233Google Scholar tend to be diagnosed at a younger age,2DeSantis C.E. Miller K.D. Goding Sauer A. et al.Cancer statistics for African Americans, 2019.CA Cancer J Clin. 2019; 69: 211-233Google Scholar have shorter stage-specific survival,3Rutter C.M. Johnson E.A. Feuer E.J. et al.Secular trends in colon and rectal cancer relative survival.J Natl Cancer Inst. 2013; 105: 1806-1813Google Scholar and have higher overall CRC mortality.4Dieleman J.L. Chen C. Crosby S.W. et al.US health care spending by race and ethnicity, 2002–2016.JAMA. 2021; 326: 649-659Google Scholar These racial disparities are often interpreted, either explicitly or implicitly, as evidence of elevated prevalence of CRC risk factors among Black people that are due to genetic predisposition or general health behaviors, such as diet and exercise. This interpretation of the data ignores the contribution of racial disparities in access to high-quality care and other resources affecting health behaviors that are amenable to more immediate policy action and overstates the ability of statistical approaches to account for these differences. As a result, policy-relevant modifiable characteristics that are key to addressing inequities are either minimized or overlooked. In this commentary, we discuss the importance of how we define race and think about characteristics related to race, including structural racism, when interpreting statistical models used to explore racial disparities in CRC. Despite the worse CRC outcomes experienced by Black people, there is very limited evidence that Black people have a greater prevalence of biological risk factors than White people. Estimation of racial differences in CRC outcomes is challenging because every outcome can be affected by screening. Screening prevents CRC by removing precursor lesions, so populations with higher screening levels have lower CRC incidence. Patients who have previously undergone colonoscopy tend to have fewer precursor lesions at subsequent examinations, and both precursor lesions and asymptomatic CRCs that are detected tend to be less advanced. Previously screened patients may also tend to have relatively more proximal lesions if tests are better able to detect distal lesions. In addition, by selectively preventing CRC after age 50 years, screening can shift the distribution of age at CRC detection to younger ages. Thus, differences in CRC screening participation and follow-up of abnormal results can affect the differences we observe in CRC outcomes for Black people and White people. Black patients and White patients who have similar screening histories, especially patients who are screening-naive, have similar prevalence of precursor lesions,5Rutter C.M. Knudsen A.B. Lin J.S. Bouskill K.E. Black and white differences in colorectal cancer screening and screening outcomes: a narrative review.Cancer Epidemiol Biomarkers Prev. 2021; 30: 3-12Google Scholar,6Stemboroski L. Samuel J. Alkaddour A. et al.Characteristics of serrated adenomas in non-Hispanic whites and African Americans undergoing screening colonoscopy.Cureus. 2021; 13: 7-12Google Scholar a crucial argument against differences in underlying risk, including risk attributable to behavioral factors such as smoking, diet, exercise, and obesity. Still, there are concerns that Black patients may have more aggressive disease based on findings that they are more likely than White patients to be diagnosed with interval CRC.7Fedewa S.A. Flanders W.D. Ward K.C. et al.Racial and ethnic disparities in interval colorectal cancer incidence a population-based cohort study.Ann Intern Med. 2017; 166: 857-866Google Scholar Higher incidence of early-onset CRC in Black patients, diagnosed before age 50 years, had also raised concerns about more aggressive disease, although the incidence of early-onset CRC is now similar in Black people and White people because of increases in incidence among White people.8Murphy C.C. Wallace K. Sandler R.S. Baron J.A. Racial disparities in incidence of young-onset colorectal cancer and patient survival.Gastroenterology. 2019; 156: 958-965Google Scholar If there are racial differences in aggressiveness, these could be partially driven by biological factors, including undiscovered differences in genetic factors. Heritable genetic factors account for only 5% of CRC in White patients. A similar fraction of CRC in Black patients is likely attributable to heritable genetic factors, although we do not have a reliable estimate, in part because Black patients have lower rates of referral for genetic testing.9Muller C. Lee S.M. Barge W. et al.Low referral rate for genetic testing in racially and ethnically diverse patients despite universal colorectal cancer screening.Clin Gastroenterol Hepatol. 2018; 16: 1911-1918Google Scholar,10Augustus G.J. Ellis N.A. Colorectal cancer disparity in African Americans: risk factors and carcinogenic mechanisms.Am J Pathol. 2018; 188: 291-303Google Scholar Genome-wide association studies have identified genetic variants that increase risk for sporadic CRC and have reported racial differences in some risk variants.10Augustus G.J. Ellis N.A. Colorectal cancer disparity in African Americans: risk factors and carcinogenic mechanisms.Am J Pathol. 2018; 188: 291-303Google Scholar Polygenic risk scores hold promise for understanding variability in underlying risk and can identify people at high risk who could benefit from earlier screening, but genotyping studies used to develop risk scores have focused exclusively on participants of European descent.11Archambault A.N. Su Y.-R. Jeon J. et al.Cumulative burden of colorectal cancer–associated genetic variants is more strongly associated with early-onset vs late-onset cancer.Gastroenterology. 2020; 158: 1274-1286Google Scholar Another concern is that Black patients are more likely to develop proximal adenomas and cancers, but the evidence of these associations is mixed.5Rutter C.M. Knudsen A.B. Lin J.S. Bouskill K.E. Black and white differences in colorectal cancer screening and screening outcomes: a narrative review.Cancer Epidemiol Biomarkers Prev. 2021; 30: 3-12Google Scholar A recent study showing that Black patients have accelerated epigenetic aging in the proximal colon, but White patients do not, suggests a plausible mechanism for increased risk of proximal lesions.12Acuna-Villaorduna A.R. Lin J. Kim M. Goel S. Racial/ethnic disparities in early-onset colorectal cancer: implications for a racial/ethnic-specific screening strategy.Cancer Med. 2021; 10: 2080-2087Google Scholar If there are racial differences in aggressiveness and location of disease, these could be partially driven by biological factors, including genetic factors and downstream biological consequences of systemic racism and social injustice.13Davidson K.W. Mangione C.M. Barry M.J. et al.Actions to transform US Preventive Services Task Force methods to mitigate systemic racism in clinical preventive services.JAMA. 2021; 326: 2405-2411Google Scholar A fundamental issue is that we need better information about which (or any) aspects of the CRC disease process vary across racial groups and the extent and mechanisms of variation. Evidence that differences in the CRC disease process can explain observed racial disparities in CRC burden is either null or inconclusive, but there is clear evidence of racial disparities in the timeliness and quality of CRC care, from early detection to treatment. For many years, Black patients have been less likely to be screened for CRC than White patients,14Rutter CM, de Lima PN, Lee JK, Ozik J. Too good to be true? Evaluation of colonoscopy sensitivity assumptions used in policy models [published online January 28, 2022]. Cancer Epidemiol Biomarkers Prev https://doi.org/10.1158/1055-9965.EPI-21-1001.Google Scholar and when screened, they may be less likely to receive endoscopic tests at high-quality facilities.15Laiyemo A.O. Doubeni C. Pinsky P.F. et al.Race and colorectal cancer disparities: health-care utilization vs different cancer susceptibilities.J Natl Cancer Inst. 2010; 102: 538-546Google Scholar Quality of care matters—colonoscopy is less effective at reducing CRC mortality when patients are examined by providers with lower adenoma detection rates.16Corley D.A. Jensen C.D. Marks A.R. et al.Adenoma detection rate and risk of colorectal cancer and death.N Engl J Med. 2014; 370: 1298-1306Google Scholar Once diagnosed, Black patients experience delays in treatment initiation17Bui A. Yang L. Myint A. May F.P. Race, ethnicity, and socioeconomic status are associated with prolonged time to treatment after a diagnosis of colorectal cancer: a large population-based study.Gastroenterology. 2021; 160: 1394-1396Google Scholar and are less likely to receive curative treatment.18Tramontano A.C. Chen Y. Watson T.R. et al.Racial/ethnic disparities in colorectal cancer treatment utilization and phase-specific costs, 2000–2014.PLoS One. 2020; 15e0231599Google Scholar These differences are driven, in part, by unequal access to high-quality care. Studies that examine patients with equal access to care find that Black patients have similar or greater screening participation than White patients19May F.P. Yano E.M. Provenzale D. et al.Race, poverty, and mental health drive colorectal cancer screening disparities in the Veterans Health Administration.Med Care. 2019; 57: 773-780Google Scholar,20Changoor N.R. Pak L.M. Nguyen L.L. et al.Effect of an equal-access military health system on racial disparities in colorectal cancer screening.Cancer. 2018; 124: 3724-3732Google Scholar and similar time to treatment after diagnosis.21Eaglehouse Y.L. Georg M.W. Shriver C.D. Zhu K. Racial comparisons in timeliness of colon cancer treatment in an equal-access health system.J Natl Cancer Inst. 2020; 112: 410-417Google Scholar In addition, when Black patients and White patients have the same access to care, they have similar clinical outcomes.22Berry J. Caplan L. Davis S. et al.A Black-White comparison of the quality of stage-specific colon cancer treatment.Cancer. 2010; 116: 713-722Google Scholar Evidence is built based on statistical models that estimate the effect of race on CRC outcomes and relies on our interpretation of these estimates. Although it is common practice to control for race in regression models of CRC incidence and outcomes, interpreting race coefficients requires care and precision that are typically lacking.23Kaufman J.S. Race: ritual, regression, and reality.Epidemiology. 2014; 25: 485-487Google Scholar Race is a social and power construct that captures the historical context that produced the stratified society that we live in today and drives contemporary characteristics of different racial groups, which include phenotypes (especially skin color), culture, wealth, income, education, access to health care, neighborhood characteristics related to healthy food access and walkability, and experiences of structural and interpersonal racism. When included in regression models, race variables thus become an imperfect proxy for a myriad of between-group differences. Estimated race coefficients capture a conglomeration of unmeasured and uncertain influences that vary depending on the other factors in the model. For example, racial differences in average age at CRC diagnosis may reflect differences in the population age structure, which are influenced by differences in all-cause mortality; thus, age-specific incidence rates provide a better comparison of risk of early-onset CRC. Prior screening, which influences CRC outcomes, is often included in models that include race as a predictor to explore the magnitude of racial disparities while controlling for prior screening. In this sense, the estimated effect of race has meaning only in the context of the other factors included in the model. These other factors will encompass only variables that have been observed and measured in the specific study at hand, which often differ widely among studies. However, the implied differences in interpretation are rarely acknowledged. Interpretation of the estimated race coefficient also depends on the specific population under study, the ascertainment and parameterization of racial categories, and on how heterogenous the sample is with respect to factors related to race and CRC outcomes, such as access to care. Indeed, analyzing homogeneous samples that include only people who all have equal access to care and similar socioeconomic status is a fundamental approach used to separate the effects of race and access to care on CRC outcomes. In summary, when interpreting analyses that examine the effect of race, it is critical to think about what race is measuring; the mechanisms through which race might affect outcomes; and which characteristics we need to control for to make the desired (or claimed) comparisons, including unmeasured or unobserved characteristics. Even if we could hypothetically measure and adjust for all meaningful differences among racial groups, we might conclude that “the adjustment is a kind of fiction”24Kaufman J.S. Dissecting disparities.Med Decis Making. 2008; 28: 9-11Google Scholar—a fiction because it is built on comparison of people in different race/ethnicity categories as if all of the factors adjusted for were similar across racial groups in spite of observed differences. Studies that demonstrate racial differences in CRC outcomes rarely adjust for differential access to wealth, income, education, high-quality health care, and healthy neighborhood environments, which are key drivers of racial differences in outcomes. In summary, when evaluating the role of race in health outcomes, including CRC outcomes, it is critical to do the following:•Clearly describe how race was measured and categorized while adhering to inclusive reporting standards.13Davidson K.W. Mangione C.M. Barry M.J. et al.Actions to transform US Preventive Services Task Force methods to mitigate systemic racism in clinical preventive services.JAMA. 2021; 326: 2405-2411Google Scholar•Recognize that estimated effects of race and ethnicity are highly dependent on other variables that are also included in the model. The degree of adjustment required to draw causal inference may not be possible and, if possible, may not be realistic. Causal models focus on characteristics that can be modified.23Kaufman J.S. Race: ritual, regression, and reality.Epidemiology. 2014; 25: 485-487Google Scholar Racism—structural, interpersonal, and internalized—is a modifiable risk factor.•Identify factors that are amenable to policy intervention, such as access to care and quality of care, that are modifiable drivers of CRC outcome disparities that are rooted in structural racism.•Recognize and acknowledge structural racism and social injustices as key drivers of poorer access to and receipt of high-quality care across all steps in the CRC continuum to enable new solutions to be imagined.•Propose solutions, such as investments in the people, studies, policies, neighborhoods, and processes, that can lead to more accessible and culturally competent care—care that can be delivered by health systems that foster community trust—and that achieve equal participation and timeliness of screening, treatment, clinical trials enrollment, and survivorship care. In the past 2 decades, CRC incidence has decreased as screening has increased, and racial disparities have generally narrowed.25May F.P. Glenn B.A. Crespi C.M. et al.Decreasing black-white disparities in colorectal cancer incidence and stage at presentation in the United States.Cancer Epidemiol Biomarkers Prev. 2017; 26: 762-768Google Scholar Nevertheless, disparities persist. Solutions to racial inequities in CRC outcomes require that we name structural racism and study its role as a causal factor.