Colorectal cancer screening in high-risk groups is increasing, although current smokers fall behind
Aminat O. Oluyemi MD,
Department of Medicine, Division of Gastroenterology and Hepatology, Penn State University Milton S. Hershey Medical Center, Hershey, Pennsylvania
Corresponding author: Aminat Oluyemi, MD, Department of Medicine, Division of Gastroenterology and Hepatology, Penn State University Milton S. Hershey Medical Center, 500 University Drive, Mail Code HU33, PO Box 850, Hershey, PA 17033; Fax: (717) 531 0061; firstname.lastname@example.org
There is limited information about colorectal cancer (CRC) screening trends in high-risk groups, including the black, obese, diabetic, and smoking populations. For this study, the authors evaluated national CRC screening trends in these high-risk groups to provide insights into whether screening resources are being appropriately used.
This was a nationally representative, population-based study using the Behavioral Risk Factor Surveillance System from the Centers for Disease Control. Data analysis was performed using bivariate analyses with weighted logistic regression.
In the general population, CRC screening increased significantly from 59% to 65% during the years 2006 to 2010. The screening prevalence in non-Hispanic blacks was 58% in 2006 and 65% in 2010. Among obese individuals, the prevalence of up-to-date CRC screening increased significantly from 59% in 2006 to 66% in 2010. Screening prevalence in individuals with diabetes was 63% in 2006 and 69% in 2010. The CRC screening prevalence in current smokers was 45% in 2006 and 50% in 2010. The odds of CRC screening in the non-Hispanic black population, the obese population, and the diabetic population were higher than in non-Hispanic whites, normal weight individuals, and the population without diabetes, respectively. Current smokers had significantly lower odds of CRC screening than never-smokers in the years studied (2006: odds ratio [OR], 0.71; 95% confidence interval [CI], 0.66-0.76; 2008: OR, 0.67; 95% CI, 0.63-0.71; 2010: OR, 0.69; 95% CI, 0.66-0.73).
According to the Centers for Disease Control and Prevention (CDC), colorectal cancer (CRC) is the second leading cause of cancer-related deaths for men and women in the United States. CRC is a preventable disease, in that most CRCs arise from a precancerous lesion, the adenoma, which is amendable to endoscopic polypectomy. The preventable nature of the disease has been a driving force in the race for CRC screening in individuals with the appropriate risk factors.
It has been demonstrated that CRC screening significantly decreases CRC-related mortality, justifying the recommended screening guidelines set forth by the US Preventive Services Task Force (USPSTF).[2, 3] An increased risk of CRC has been associated with advancing age, a family history of CRC, inflammatory bowel disease, and uncommon genetic syndromes. In recent years, additional risk factors for CRC have been identified, including black race, obesity, diabetes mellitus, and smoking (Table 1).[5-9, 12] The incidence of CRC is highest among African Americans compared with other racial/ethnic groups (Table 1). It has been reported that the risk of CRC Is 19% greater in obese patients, 30% greater in individuals with diabetes, and 20% greater in smokers (Table 1).[6, 11, 12] Consequently, in individuals with these risk factors, adherence to recommended CRC screening guidelines is a pertinent public health concern.
Table 1. Known Risk of Colorectal Carcinoma for Each of the Studied High-Risk Groups
The effectiveness of CRC screening is based on the detection and removal of premalignant adenomas and the detection of early stage cancer that is amendable to curative surgical resection. The initial screening guidelines by the USPSTF in 1997 recommended CRC screening with fecal occult blood testing (FOBT), flexible sigmoidoscopy, or a combination of both. In 2002, the revised USPSTF recommendations added the option of colonoscopy for screening average-risk individuals. The most recent USPSTF CRC screening recommendations were published in 2008 and, in general, remained unchanged since the 2002 publication.
CRC screening has been endorsed by many professional societies, which has increased awareness of the benefits of screening in the public and the medical community. Recently, increased CRC screening rates have paralleled a decrease in the incidence and mortality of CRC.[4, 14, 15] Recent data from the National Health Interview Survey (NHIS) suggest that 58.6% of US adults aged >50 years are receiving appropriate CRC screening, representing significant increases in CRC screening over the past decade. However, these screening rates are still well below the 2020 target benchmark of 80% set by the CRC Control Program.
Less is known about trends in CRC screening in specific high-risk groups, including blacks, obese individuals, diabetics, and smokers. Disparities in other preventive health screening for these groups have been noted, raising concern that they may also be less likely to be screened for CRC. For example, blacks have lower rates of receipt of nearly all preventive services, including adult immunizations and screening for breast and colon cancer. Reported lower rates of breast and colon cancer screening in blacks appear to have been influenced by socioeconomic factors. In addition, overweight and obese women are less likely to receive appropriate cervical and breast cancer screening compared with normal weight women.[19-21] Likewise, smoking has been associated with decreased likelihood of breast and cervical cancer screening. If it is demonstrated that these high-risk populations are less likely to have up-to-date CRC screening, then specific interventions need to be developed to promote screening in these groups.
The objective of this study was to evaluate the national prevalence of CRC screening among recently identified high-risk groups compared with the general population. Quantifying this CRC screening prevalence will provide insight into whether screening resources are being used appropriately in these populations.
MATERIALS AND METHODS
The Behavioral Risk Factor Surveillance System (BRFSS) is a standardized telephone survey of US adults that is conducted annually by state health agencies and is facilitated by the CDC. The BRFSS collects data on health behaviors related to the leading causes of death in the United States. Data from this survey are used throughout the United States for making policy, allocating resources, and monitoring the prevalence of disease and screening practices. It uses a stratified design to sample adults living in homes with telephones. The BRFSS survey contains a core group of questions used by all states and optional questions that may be included at the individual state's discretion. Data for the current study were obtained from answers to the core group of questions and included all 50 states and other US territories. To study trends in colorectal screening prevalence, we analyzed data from the 2006, 2008, and 2010 surveys. In light of accepted CRC guidelines for average-risk individuals, we limited the sample to adults ages 50 to 75 years for whom colorectal screening is recommended. Answers of participants ages 50 to 75 years were included if information regarding CRC screening status (yearly FOBT or colonoscopy/sigmoidoscopy within 10 years) were available.
Definition of Dependent Variables
Up-to-date CRC screening prevalence was the outcome variable for this study. The CRC module contained questions pertaining to cancer screening by FOBT, sigmoidoscopy, and colonoscopy. A participant was defined as having received up-to-date CRC screening if they reported FOBT in the past year or having a sigmoidoscopy or colonoscopy within the past 10 years. Although CRC screening with sigmoidoscopy is recommendation every 5 years, the 2006 survey only allowed the participant to indicate whether a sigmoidoscopy had been performed in the past 10 years. Thus, for analytic purposes, either a colonoscopy or a sigmoidoscopy within 10 years was defined as up-to-date CRC screening in all survey years.
Definition of Independent Variables
Non-Hispanic black race, obesity, diabetes, and tobacco smoking were the main independent variables evaluated in this analysis. Race/ethnicity was categorized as non-Hispanic white, non-Hispanic black, Hispanic, Asian, and other. Body mass index (BMI) weight categories were defined using self-reported height and weight as follows: <18.5 kg/m2 (underweight), 18.5 to 24.9 kg/m2 (normal weight), 25 to 29.9 kg/m2 (overweight), and ≥30 kg/m2 (obese). Diabetes was assessed by the question, “Have you ever been told by a physician that you have diabetes?” Participants were defined as having diabetes if they reported having nonpregnancy-related diabetes. Tobacco smoking status was categorized as never smoked, former smoker (smoked at least 100 cigarettes in your life, but not currently smoking), or active smoker (any current smoking).
Other socioeconomic variables were included as control variables. These variables included age, sex, education (did not graduate high school, high school graduate, some college, and college graduate), annual income (<$25,000, $25,000-$49,999, and ≥$50,000), and any health insurance.
Bivariate analyses with weighted logistic regression were performed to determine associations between the independent variables or other covariates and up-to-date CRC screening. Multivariable logistic regression was used to adjust for all other independent variables and covariates. Model-adjusted odds ratios (ORs) were used to quantify the magnitude and direction of the significant associations. All analyses used weighting procedures in SAS version 9.3 (SAS Institute Inc., Cary, NC) to account for the complex sampling stratification and clustering used by the BRFSS.
There were 155,020 participants, 197,969 participants, and 229,202 participants in the 2006, 2008, and 2010 surveys, respectively, who met the inclusion criteria and were analyzed (Table 2). The overall weighted prevalence of CRC screening increased over time from 59%, 63%, and 65% in 2006, 2008, and 2010, respectively (2006 vs 2010, P < .001) (Table 2). A comparison of up-to-date CRC screening prevalence by survey year is provided in Table 3. CRC screening prevalence also is listed with respect to other sociodemographic variables (Table 2).
Table 2. Prevalence and Adjusted Odds Ratio of Colorectal Cancer Screening for All Independent Variables and Covariates
2006, N = 155,020
2008, N = 197,969
2010, N = 229,202
OR (95% CI)
OR (95% CI)
OR (95% CI)
Abbreviations: BMI, body mass index; CI, confidence interval; OR, adjusted odds ratio.
Table 3. Prevalence of Colorectal Cancer Screening in the Study Populations
CRC Screening Rate per Studies Year, %
2006 vs 2008
2006 vs 2010
2008 vs 2010
Abbreviations: CRC, colorectal carcinoma.
The prevalence of up-to-date CRC screening increased in all racial/ethnic groups across the study years. Screening prevalence in non-Hispanic blacks was 58%, 62%, and 65% in 2006, 2008, and 2010, respectively, compared with 61%, 65%, and 67%, respectively, in non-Hispanic whites (Fig. 1). In the adjusted model, the odds of CRC screening in non-Hispanic blacks were greater compared with the odds in non-Hispanic whites in 2006 (OR, 1.15; 95% confidence interval [CI], 1.05-1.27), in 2008 (OR, 1.19; 95% CI, 1.08-1.30), and in 2010 (OR, 1.23; 95% CI, 1.15-1.32) (Table 2).
Among obese individuals, the prevalence of up-to-date CRC screening increased from 59% in 2006 to 66% in 2010 (P ≤ .001) (Fig. 2, Table 2). In the adjusted analysis, the odds of up-to-date CRC screening did not differ significantly in obese individuals compared with normal weight individuals in 2006 or 2008, but obese individuals had greater odds of up-to-date CRC screening in 2010 compared with normal weight individuals (adjusted OR, 1.16; 95% CI, 1.11-1.23) (Table 2). In overweight individuals, the CRC screening prevalence increased from 60% in 2006 to 65% in 2010 (P ≤ .001) (Fig. 2, Table 2). Compared with normal weight individuals, overweight individuals had a slightly higher odds of screening in 2010 (adjusted OR, 1.06; 95% CI, 1.01-1.12) (Table 2).
Among individuals who had diabetes, CRC screening prevalence was higher in all years compared with the prevalence among those who did not have diabetes (Fig. 3, Table 2). This findings persisted in the adjusted analysis, with 25%, 20%, and 21% greater odds of CRC screening in individuals who had diabetes compared with those who did not have diabetes in 2006, 2008, and 2010, respectively (Table 2).
CRC screening prevalence also differed by smoking status. The highest prevalence of CRC screening was among former smokers, followed by never-smokers, then current smokers. The CRC screening prevalence in current smokers was 45% in 2006 and 50% in 2010 compared with 66% and 71%, respectively, in former smokers and 59% and 66%, respectively, in those who never smoked (Fig. 4, Table 2). In the adjusted analysis, current smokers had significantly lower odds of CRC screening than never-smokers (OR, 0.69; 95% CI, 0.66-0.73) in 2010 (Table 2). In contrast, former smokers were more likely to be screened than never-smokers (OR, 1.16; 95% CI, 1.12-1.22) in 2010 (Table 2).
In all study years, the prevalence of up-to-date screening increased with age. Individuals who had lower levels of education consistently had a lower likelihood of CRC screening compared with college graduates. Similarly, lower levels of income were associated with lower odds of CRC screening compared with higher levels of income. In the study population, individuals with health insurance were much more likely to be screened for CRC. The odds of CRC screening in those who had health insurance increased over time (adjusted OR per studied year: 2.41 in 2006, 2.60 in 2008, and 2.57 in 2010) (Table 2) compared with those who did not have health insurance.
In this nationally representative, population-based study, we documented an increasing trend in CRC screening prevalence in all recently identified high-risk groups compared with the general population. We also identified higher odds of screening associated with black race, obesity, and diabetes. The exception noted in the current study was that there were lower odds of CRC screening among current smokers.
Previous studies have demonstrated that non-Hispanic blacks have a higher incidence of adenomas and CRC even after controlling for possible confounding factors, such as socioeconomic status and health care access.[8, 23] Therefore, effective screening in this population is of particular importance. Our study revealed a reassuring upward trend in screening prevalence over time among non-Hispanic blacks, with higher adjusted odds of CRC screening among blacks compared with whites in all 3 study years. This finding was surprising, because blacks have demonstrated lower rates of other preventive health services. Despite this, the incidence of CRC among non-Hispanic blacks remains higher than that among non-Hispanic whites (Table 1), so efforts are needed to continue increasing the CRC screening prevalence in this high-risk population.
The current literature suggests that overweight and obese individuals are at high risk for colorectal adenoma and CRC. In a meta-analysis by Ben et al, a 5-kg/m2 increase in BMI was associated with a 19% increased risk of colorectal adenoma. Similarly, Calle et al reported that increased body weight was associated with an increased risk of death from CRC. In light of these data, screening for CRC is a major public health concern.
The increased risk of CRC mortality may be caused by lower rates of CRC screening in overweight and obese individuals; however, our current findings demonstrate that, in 2010, both overweight and obese individuals had a greater likelihood of CRC screening than normal weight individuals. Similarly, a gradual increase in screening prevalence over time was observed. Our findings were unexpected, because obesity has been associated with a delay in or avoidance of other types of cancer screening.[25, 26] It was believed that this avoidance was related to the phenomenon of perceived discrimination. This perception has an inverse relation with poor mental and physical outcomes, which play a role in responsible health decisions, including cancer screening.[27, 28] The trend of higher CRC screening prevalence over time among overweight and obese individuals is encouraging, and whether it will result in attenuation of the CRC mortality disparity over time remains to be determined.
Diabetes is an independent risk factor for CRC in both men and women.[6, 9] Several etiologies for the increased risk of CRC in diabetics have been postulated, including the role of hyperinsulinemia and insulin-like growth factor in supporting the proliferation of colonic cancer cells, thereby leading to CRC. Regardless of the exact pathophysiologic process, the mere presence of a greater cancer risk emphasizes the need for vigilant screening practices. Our findings indicate that individuals who had diabetes were more likely to be up-to-date with CRC screening than their nondiabetic counterparts and had an increasing prevalence of CRC screening over time. This finding parallels the results from a study by Bell et al, who reported that individuals with diabetes in North Carolina were as likely to get screened for CRC as those without diabetes. Higher rates of CRC screening in individuals with diabetes may be attributable to frequent exposure to health care, which provides more opportunities for preventive screening. It is worth noting that the BRFSS identifies diabetes by self-report, so these favorable screening rates are among those recognized to have diabetes. Of course, we are unable to access the CRC screening prevalence among individuals who have undiagnosed diabetes mellitus.
There is a high global burden of disease associated with tobacco use secondary to the increased, accumulated hazards of smoking. Such disease burden is manifested by as much as a 2-fold increased risk of colonic adenoma in addition to an increased risk of CRC incidence and mortality.[32-35] The reported population-attributable risk percentage of smoking as it relates to colon cancer in the United States is 16% to 21% of CRCs among men and 11% of colon cancers and 17% of rectal cancers among women. In light of these findings, effective screening practices for smokers are very important.
Our study demonstrates that, similar to the general population, current smokers had an increasing prevalence of CRC screening over time, but they consistently had a lower likelihood of CRC screening compared with never-smokers in all studied years. The lower screening prevalence among smokers may be caused by decreased overall attention to preventive health care. Although our analysis adjusted for sociodemographic factors, such as income, education, and having health insurance, there may be other socioeconomic barriers that decrease screening in smokers, such as not having a usual source of primary care. Education efforts aimed at informing patients who smoke and primary physicians of low screening rates among smokers may be helpful.
The impact of socioeconomic factors on overall cancer incidence has been well reported in the literature.[37, 38] Individuals with a high school education and those with lower income have higher rates of lung cancer compared with those who have higher income and a college-level education. Similarly, lower income among women has been associated with a statistically significant increased risk of distant-stage breast cancer. The results of our study are consistent with the reports described above. We observed that, in each socioeconomic category, the lowest odds of screening were evident in the younger age group (ages 50-54 years), the least educated population, and those with the lowest income (Table 2). In addition, compared with noninsured individuals, insured individuals had statistically significant higher odds of being screened for colon cancer. This finding, although well reported, highlights target groups for colon cancer screening initiatives.
Overall, our study findings are encouraging, but they suggest that continued aggressive screening efforts are needed to reach our target goal of screening 80% of the eligible population by 2020.17 Reasons for low screening prevalence are likely multifactorial, including cultural factors, lack of screening access, or lack of awareness of risk. Some interventions have been studied, such as using electronic medical record systems to identify those who need screening, contacting those individuals, and scheduling appointments. Others have evaluated the effectiveness of using “patient navigators” to guide patients through the often complicated health care system to obtain CRC screening. Understanding inhibitors of screening and identifying methods that facilitate screening could allow for increased prevention of CRC.
Our study was limited by the constraints of the questions used in the BRFSS. In 2006, unlike in 2008 and 2010, there was an inability to differentiate whether individuals underwent colonoscopy or sigmoidoscopy. Consequently, any endoscopy within 10 years was considered adequate screening during each studied year, so that individuals who had undergone a colonoscopy within the last 10 years would not be excluded. Although this may have overestimated the appropriate observed prevalence of CRC screening, it is not expected that this would differentially bias the results of the subgroups studied. The BRFSS is a telephone survey, and the accuracy of the survey responses depend on the participants' self-report and, thus, is susceptible to recall bias. In addition, the telephone numbers used were land lines, which could result in sampling bias; however, this probably is less of a concern in the age groups that were included in our analysis. In 2011, the BRFSS began to include cell phone numbers in their sample to adjust for the changing communication habits of Americans. Strengths of this study include the large sample size, which was representative of the adult US population, making the results widely generalizable.
The current results demonstrate that the prevalence of CRC screening is trending upward in all high-risk groups. Despite the increases in CRC screening prevalence over time, we observed that current smokers had significantly lower odds of CRC screening compared with the general population. Further studies evaluating the etiology of poor screening practices in current smokers are necessary. Patient and provider education about CRC risk factors and the impact of screening will prove beneficial in achieving our CRC screening goals.