Sociocultural Factors, Access to Healthcare, and Lifestyle: Multifactorial Indicators in Association with Colorectal Cancer Risk

Colorectal cancer incidence causes a large disease burden where an estimated 151,030 individuals in the United States will be diagnosed in 2022. The colorectal cancer burden especially impacts Black Americans who have the highest colorectal cancer incidence of any racial group in the United States (1). The causes of the colorectal cancer racial disparity are not completely understood; theorized causes include a combination of differences in socioeconomic status (SES), in screening and access to healthcare, and in the prevalence of healthy behaviors.

Epidemiologic studies show that several health behaviors and lifestyle factors are related to decreased colorectal cancer risk, including nonsmoking, maintaining a healthy weight, moderate alcohol intake, physical activity, and healthy diets (1). These health behaviors are less prevalent among Blacks Americans than non-Hispanic White Americans (1). However, much of the epidemiologic research that provides support for the associations between lifestyle factors and colorectal cancer risk has been conducted in cohorts where most individuals are of high socioeconomic position and are non-Hispanic White (2). In addition, previous epidemiologic studies provide evidence that lifestyle factors may have weaker associations with health outcomes in Blacks and populations of low socioeconomic position (3, 4). The Southern Community Cohort Study (SCCS) provides an opportunity to investigate associations in a cohort comprised of individuals of low-SES and who are Black Americans. An effective strategy to reduce colorectal cancer racial disparities includes identifying risk factors most important to colorectal cancer risk in high risk populations.

Herein, we characterize the associations between sociocultural factors, access to healthcare, and lifestyle factors with colorectal cancer risk. In addition, we evaluate whether the associations between race and colorectal cancer risk are influenced by lifestyle, SES and access to care.

The study data arise from the prospective SCCS (5, 6). The SCCS enrolled over 85,000 adult participants from 2002 to 2009 in 12 states in southeastern United States. Participants were English-speaking, and age 40 to 79 at enrollment. The majority of participants (86%) were enrolled at Community Health Centers. The remaining 14% of cohort participants were enrolled using an identical mailed questionnaire sent to stratified random samples of residents in the same 12 states. At study enrollment, all participants completed questionnaires to obtain information on demographics, socioeconomics, cancer screening participation, medical history, and lifestyle factors, such as height, weight, smoking history, alcohol intake, diet during the previous year, sitting time, and physical activity. The SCCS was approved by Institutional Review Boards at Vanderbilt University Medical Center (Madison, WI) and Meharry Medical College (Nashville, TN). All participants provided written informed consent. The study was done with ethical standards consistent with the Belmont Report.

In 2021, SCCS staff performed linkages to state cancer registries and the National Death Index to acquire information on incident colon and rectal cancers as defined by International Classification of Diseases-Oncology codes C180–189, C199, and C209 (N = 690) through December 31, 2017.

We were interested in establishing the associations between sociocultural factors, access to healthcare, and lifestyle with colorectal cancer risk in the SCCS. All exposures were assessed at the baseline interview. Sociocultural factors of interest were race (Black, White, or other), sex (male or female), household income, educational attainment, and neighborhood deprivation index. The neighborhood deprivation index variable represents SES measured at the census tract-level to summarize: ownership and type of housing, income measures, household makeup, unemployment, high school graduation rates, occupations, and car ownership (7, 8). Access to health care was operationalized by variables for ever undergoing colorectal cancer screening and health insurance status. We evaluated lifestyle factors by investigating the relations between colorectal cancer incidence and adhering to the American Cancer Society (ACS) Guideline for Diet and Physical Activity for Cancer Prevention (9, 10). Associations with cancer incidence were assessed for sedentary time, body mass index (BMI), physical activity, an ACS dietary score, alcohol consumption, and smoking status. BMI was calculated using the values for weight and height provided at enrollment. Participants were considered as having met current physical activity recommendations via sports and exercise if they reported ≥150 minutes per week of moderate activity, ≥75 minutes per week of vigorous activity or ≥150 minutes per week of moderate and vigorous activity combined. The ACS dietary score consisted of three component parts for meeting guidelines for: consuming ≥4 cups of fruits and vegetables, choosing at least 50% of grains as whole grains, and limiting consumption of processed and red meat (Supplementary Table S1). Sedentary time was defined as the number of hours per day the participant reported sitting. Dietary intake was assessed using a Food Frequency Questionnaire, developed and validated specifically for the diet in the Southeastern United States (11, 12). We classified non- and moderate alcohol drinkers as having met the cancer prevention guideline, as defined by the United States Department of Agriculture (USDA) Dietary Guidelines for Americans for moderate drinking as alcohol intake reported as >0 but ≤1 drink/day for women or ≤2 drinks per day for men, and heavy drinking as >1 drink per day for women and >2 drinks per day for men (13). Never smokers met the cancer prevention guideline. Former smokers were defined as participants who had ever smoked and did not report cigarette smoking at enrollment interview.

We created two healthy lifestyle scores to indicate the number of guidelines adhered to from ACS Guidelines on Nutrition and Physical Activity for Cancer Prevention and Nonsmoking. The first healthy lifestyle score was created by counting and summing (0–5) the number of ACS Guidelines on Nutrition and Physical Activity for Cancer Prevention (9, 10) the participant met upon entry into the cohort by assigning one point for each of: BMI in the “healthy” category, meeting physical activity guidelines, being a never smoker, being a non- or moderate alcohol drinker, and meeting ≥1 diet quality expectations. We created a second healthy lifestyle score that did not include BMI, because unlike the other components of the score, BMI is not a health behavior. We did not include sedentary time in the healthy lifestyle scores because sitting time was not associated with colorectal cancer in this cohort, and the optimal amount of sedentary time per day is currently undefined.

This study includes 75,182 participants who met the following inclusion criteria: ≥2 years of follow-up and; no diagnosis of cancer (except nonmelanoma skin cancer) before baseline interview. Missing covariate data (0.9%–2.8% of participants) were set to sex- and race-specific medians or modes.

Frequency distributions of participant characteristics were tabulated by colorectal cancer incidence, and variables of interest. Cox models were used to estimate HR for colorectal cancer incidence associated with sociocultural factors, access to healthcare, lifestyle factors, and two healthy lifestyle scores. Age was used as the time scale. Entry time in the Cox models was defined as age at enrollment and exit time as age at colorectal cancer diagnosis, age at death, loss to follow-up, or December 31, 2017, whichever came first. We evaluated the proportional hazards assumption graphically, and considered it met.

Statistical models included the following variables as potential confounders, measured at baseline interview: race (Black, White, other), sex (male, female), enrollment source (community health center [CHC], non-CHC), colorectal cancer screening (ever, never participated in colonoscopy or sigmoidoscopy) health insurance status (yes, no), household income (<$15,000; $15,000‐-24,999; ≥$25,000), education (<high school, high school, >high school), neighborhood deprivation index (quintiles based on the distribution of neighborhood deprivation index value of all the census tracts in the 12 states that encompass the SCCS recruitment area), BMI (<18.5, 18.5–24.9, 25.0–29.9, ≥30.0 kg/m²), physical activity (meets, does not meet guideline), sedentary time (quartiles), ACS diet quality variable (0–3), smoking status (never, former, current), alcohol intake (women: none, 0< drink/day ≤1 drink/day, >1 drink/day; men: none, 0< drinks/day ≤2 drinks/day, >2 drinks/day), and family history of colorectal cancer diagnosis in a first- degree relative (yes, no, unknown).

We also calculated HRs for colorectal cancer incidence by variables of interest stratified by sex, race, ever participation in colorectal cancer screening and anatomic site. Possible interactions between variables of interest were assessed by likelihood ratio tests to compare main effects models with and without the addition of cross-product terms. Statistical analyses were performed using SAS statistical software (version 9.4; SAS Institute Inc) in 2022.

Data available for qualified investigators and can be requested via: southerncommunitystudy.org/

The prevalence of exposures for SES, access to healthcare, and health behaviors varied by race and case status (Supplementary Tables S1 and S2). In general and by case status, Black participants had lower household income, less educational attainment, and more often lived in areas with lower area-level SES. Black participants also were less likely to report participation in colorectal cancer screening.

The variable for Black race was strongly related to colorectal cancer and the association was not diminished by accounting for family history, BMI, health behaviors, SES, or access to healthcare (Table 1; Fig. 1). Specifically, the association between Black race and colorectal cancer risk in minimally-adjusted models was 1.35 (95% CI, 1.13–1.62), whereas the HR in fully-adjusted models was 1.33 (95% CI, 1.10–1.62). Because Black participants were less likely to report ever undergoing colorectal cancer screening, we examined whether the association between race and colorectal cancer risk was consistent when stratified by screening status. Among participants that had never been screened and age-eligible for colorectal cancer screening at enrollment (age ≥50 at enrollment), we observed a strong and consistent association between Black race and increased colorectal cancer risk (Table 1). However, among participants that were age-eligible for colorectal cancer screening at enrollment and who reported ever having colorectal cancer screening, the association between Black race and colorectal cancer risk was attenuated and 95% CIs crossed unity (fully-adjusted HR for Black race = 1.16; 95% CI, 0.78–1.73).

In this cohort where most participants are of low SES, household income and neighborhood socioeconomic environment were not strongly associated with colorectal cancer risk (Table 2). Participants with greater than a high school education were at lower risk of colorectal cancer when compared with participants with less than high school attainment (HR = 0.81; 95% CI, 0.67–0.98). Gender was not associated with colorectal cancer risk.

As previously reported, healthcare access and use were associated with lower colorectal cancer risk in the SCCS (Table 2; Fig. 1; ref. 14). Ever undergoing colorectal cancer screening via colonoscopy or sigmoidoscopy was the strongest factor associated with reduced colorectal cancer incidence in the study. Participants that had health insurance at enrollment were also at decreased risk of colorectal cancer.

The majority of cohort members did not meet the ACS Guidelines on Nutrition and Physical Activity for Cancer Prevention and Nonsmoking for BMI and physical activity (Table 3; Supplementary Tables S1 and S2). Participants that were subsequently diagnosed with incident colorectal cancer had the following characteristics at enrollment: 21.6% had a BMI between 18.5 and 24.9 kg/m², 16.6% met the guideline for physical activity set forth in the Physical Activity Guidelines for Americans, 56.6% sat for 8.5 or fewer hours per day (a measure of sedentary time), and 51.5% of cases did not drink alcohol. When stratified by race, Black cases were more likely to be overweight or obese than non-Hispanic White cases (77.6% vs. 72.0%), and more likely to be classified as heavy drinkers (17.1% vs. 11.9%).

We created a Diet Quality Score based on the ACS Guidelines on Nutrition for Cancer Prevention, and found no association with colorectal cancer (Table 3). In addition, the component parts of the diet quality score (consuming ≥ 4 cups of fruits and vegetables, choosing at least 50% of grains as whole grains, and limiting consumption of processed and red meat) were not associated with risk (Supplementary Table S3).

Whereas, the associations between individual health behaviors and colorectal cancer risk were null, a healthy lifestyle score that included smoking status, alcohol, physical activity, and the ACS diet score was associated with lower colorectal cancer risk in models adjusted for sex and race. Specifically, participants that adhered to three or four guidelines for healthy lifestyle had a 19% decreased colorectal cancer risk compared with participants that adhered to ≤1 guideline (HR -= 0.81; 95%CI: 0.66,0.99). The association was consistent after adjustment for SES and access to healthcare, although HRs were no longer significant (Table 4). When adherence to BMI weight guidelines were added to the healthy lifestyle score the association with colorectal cancer risk was slightly attenuated but the association remained that meeting more guidelines was associated with a nonsignificant decreased risk of colorectal cancer (Table 4). The associations with the healthy lifestyle variable and colorectal cancer risk did not vary by sex (P_interaction = 0.07), or race (P_interaction = 0.37), although associations were less apparent in analyses restricted to Black participants (Supplementary Table S4).

Along with colorectal cancer screening, race, health insurance coverage, and attaining a high school education or more, adhering to a healthy lifestyle was a consistent risk reduction factor for decreased colorectal cancer risk (Fig. 1). We evaluated whether the exposures most strongly associated with risk in our study population had differing strengths of the association by anatomic site. We found consistent inverse associations with colon and rectal cancers by colorectal cancer screening, race, health insurance coverage, education, and adhering to a healthy lifestyle (Supplementary Table S5).

We also evaluated whether the associations between sociocultural, access to healthcare and lifestyle factors with colorectal cancer risk varied by participation in colorectal cancer screening (Supplementary Table S6), and found no differences in our point estimates by health insurance status, neighborhood SES, or individual-level health behaviors and income. Among participants that had ever been screened for colorectal cancer, men had a higher colorectal cancer risk (HR = 1.67; 95% CI, 1.19–2.34). Among participants eligible for colorectal cancer screening who has never participated in screening, education was no longer associated with decreased colorectal cancer risk (HR for attaining ≥ high school education = 1.01; 95% CI, 0.76–1.32).

We examined the relations between healthcare, sociocultural, and lifestyle factors with colorectal cancer incidence, and found Black race is strongly related to colorectal cancer. Black race is consistently associated with increased colorectal cancer risk in the SCCS, and the association does not vary when statistical adjustments are made for variables for health insurance coverage, SES, or lifestyle. Our results are in line with national data that shows Black Americans have higher colorectal cancer incidence rates in comparison to other racial groups. Nationally, colorectal cancer incidence is 20% higher among Black Americans compared with non-Hispanic Whites and 50% higher than incidence in Asian Pacific Islanders (1).

Higher incidence among Black Americans may partially reflect racial differences in the prevalence of lifestyle factors, such as obesity, although individual health behaviors are not associated with risk in the present study. Black participants in the study, report less participation in colorectal cancer screening, and lower colorectal cancer screening among Black Americans has been documented by the SCCS and others (14–17). Other studies have noted that Black Americans less often access healthcare, evidenced by lower follow-up of colorectal cancer abnormalities found on screening (18). Lower colorectal cancer screening by Black Americans may be related to having fewer financial resources and less flexibility in daily schedule (18–20). Lower colorectal cancer screening rates and less access to healthcare may mediate the association between Black race and colorectal cancer. In support of that assertion, the association between race and colorectal cancer risk was attenuated among participants eligible for colorectal cancer screening who had ever undergone colorectal cancer screening at enrollment (HR for Black race = 1.16; 95% CI, 0.78–1.73). Our results suggest that increasing colorectal cancer screening rates and access to preventative services for Black Americans would lessen the racial disparity in colorectal cancer risk.

Although the majority of the cohort is of low SES, the Black participants have lower household income, less educational attainment, and more often live in areas with lower area-level SES than the White participants. Low SES may influence health outcomes through less access to medical care, social support, and financial resources, including resources to buy and access nutritious foods (21). In addition, unfamiliarity with recommendations for health behaviors may keep individuals from participating in colorectal cancer screening and performing healthy behaviors. In support of that theory, previous studies suggest that the association between education and colorectal cancer risk is reflective of differences in health behaviors and colorectal cancer screening (22, 23). Our study data also supports this theory in that we observed evidence that greater education attainment is associated with a reduction in colorectal cancer in analyses including all participants, and analyses restricted to Black participants or participants that had ever been screening for colorectal cancer. Other determinants that may cause Black Americans to be at increased colorectal cancer risk are increased levels of stress, and the effects of discrimination due to systematic racism, such as lower likelihood of receiving a physician recommendation for colorectal cancer screening in comparison with White patients (24–26).

Whereas individual health behaviors are not strongly associated with colorectal cancer incidence in the SCCS, healthy lifestyle taking into account overall adherence to several health behaviors is associated with lower risk of colorectal cancer. For instance, adhering to three or more health behaviors of nonsmoking, moderate alcohol intake, high diet quality, and physical activity was associated with a 19% (95% CI, 0–34) decrease in colorectal cancer risk. The association does not vary by sex or race. The lower risk associated with healthy lifestyle is of similar magnitude in association as attaining a high school education which is associated with a 19% (95% CI, 2–33) decreased colorectal cancer risk.

Previous studies report mixed findings on the association between adherence to cancer prevention guidelines for healthy lifestyle and colorectal cancer incidence. Two previous studies find weak to null associations between adherence to guidelines set forth by World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) and colorectal cancer among women (27, 28). The WCRF/AICR guideline scores differ from the healthy lifestyle scores in current study in that WCRF/AICR guideline scores also incorporate avoiding adult weight gain as cancer prevention guideline. In two studies that reported analyses specific to Black participants, null associations are also reported (27, 29). A noted limitation of the previous studies is the small sample size of Black cases. In addition, these previous studies, as well as this study report low adherence to cancer prevention guidelines by participants. In other cohorts (28, 30–32), whose members are more often of European ancestry and higher SES, healthy lifestyle has been consistently shown to reduce colorectal cancer risk, and the association was consistent across anatomic site. In addition, data from postmenopausal women enrolled in the Women's Health Initiative Observational Study find a strong inverse association between an ACS cancer prevention guidelines score and lower colorectal cancer risk, where participants that met the most ACS cancer prevention guidelines had a 52% lower risk of colorectal cancer (HR = 0.48; 95% CI, 27–68) compared with participants that met the fewest guidelines (32). The authors did not report race-specific associations. Our data support a role for healthy lifestyle in colorectal cancer prevention, however, the magnitude of association is smaller in our cohort that primarily includes Americans of low SES and who are Black.

Our study has limitations including the use of self-reported health behaviors which are susceptible to measurement error. Due to the prospective cohort study design, misclassification is expected to be nondifferential and, if present, likely will attenuate study results. In addition, we use exposure information collected at baseline and do not have preenrollment data on risk factors that may contribute to risk across the life course, such as diet and body weight. Importantly, our study also has a number of strengths. The SCCS is a large, prospective, cohort study with comprehensive information on sociocultural and lifestyle factors, and complete follow-up to identify incident colorectal cancer cases. The cohort includes underserved at-risk populations seldom included in large numbers in other investigations. The SCCS is uniquely situated to identify exposures that influence colorectal cancer risk in African Americans of very low socioeconomic status, a population with one of the highest colorectal cancer incidence rates in the United States.

Our study provides evidence that among individuals of low SES, there are several factors important to reducing colorectal cancer risk including race, healthy lifestyle, education, colorectal cancer screening, and health insurance coverage. The colorectal cancer risk-lowering benefits of adhering to a healthy lifestyle through health behaviors did not vary by race, sex, or SES. Our findings suggest that colorectal cancer incidence will decrease through focused interventions aimed at increasing uptake and access to colorectal cancer screening, facilitating Americans’ adherence to maintaining a healthy lifestyle, and lessening the social determinants that uniquely harm Black Americans’ health outcomes.

S. Warren Andersen reports grants from NIH/NCI during the conduct of the study. W.J. Blot reports grants from NIH during the conduct of the study. No disclosures were reported by the other authors.

S. Warren Andersen: Conceptualization, formal analysis, funding acquisition, investigation, methodology, writing–original draft, project administration, writing–review and editing. W. Zheng: Resources, funding acquisition, investigation, writing–review and editing. M. Steinwandel: Resources, data curation, writing–review and editing. H.J. Murff: Writing–review and editing. L. Lipworth: Writing–review and editing. W.J. Blot: Conceptualization, resources, funding acquisition, writing–review and editing.

This work was supported by the NCI at the NIH (grant nos. R00 CA207848 and R01 CA255318 to S. Warren Andersen); the University of Wisconsin-Madison, Office of Vice Chancellor for Research and Graduate Education with funding from the Wisconsin Alumni Research Foundation and the University of Wisconsin Carbone Cancer (grant no. P30 CA014520 to Howard H. Bailey which supports S. Warren Andersen). SCCS is supported by the NCI at the NIH (grant nos. R01 CA92447, U01 CA202979 to W.J. Blot), including special allocations from the American Recovery and Reinvestment Act (grant no. 3R01 CA092447‐08S1 to W.J. Blot). Data on SCCS cancer cases used in this publication were provided by the Alabama Statewide Cancer Registry; Kentucky Cancer Registry; Tennessee Department of Health, Office of Cancer Surveillance; Florida Cancer Data System; North Carolina Central Cancer Registry, North Carolina Division of Public Health; Georgia Comprehensive Cancer Registry; Louisiana Tumor Registry; Mississippi Cancer Registry; South Carolina Central Cancer Registry; Virginia Department of Health, Virginia Cancer Registry; Arkansas Department of Health, Cancer Registry. The Arkansas Central Cancer Registry is fully funded by a grant from the National Program of Cancer Registries, Centers for Disease Control and Prevention (CDC). Data on SCCS cancer cases from Mississippi were collected by the Mississippi Cancer Registry which participates in the National Program of Cancer Registries (NPCR) of the CDC. Cancer data for SCCS cancer cases from West Virginia have been provided by the West Virginia Cancer Registry. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the CDC or the Mississippi Cancer Registry. The opinions expressed are those of the authors and do not necessarily represent those of the CDC or the West Virginia Cancer Registry.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.