Racial disparities in colon cancer

Primary care endoscopy as a tool to increase screening rates among minority patients


  • Presented at Exploring Models to Eliminate Cancer Disparities Among African American and Latino Populations: Research and Community Solutions, Atlanta, GA, April 21–22, 2005.


Colon cancer is a condition whose far-reaching effects have been well documented nationally and within the state of South Carolina. Fortunately, the disease is amenable to both primary and secondary prevention through screening colonoscopy. Despite the conceptual simplicity of recommending colonoscopy, barriers exist to universal (or even widespread) screening. Currently the infrastructure necessary to achieve screening goals set by the American Cancer Society (ACS), the American College of Gastroenterology (ACG), and the South Carolina Department of Health and Environmental Control (DHEC) has not been established. At current rates of training gastroenterologists, the medical community will not be able to come close to achieving widespread screening. Given the discrepancy between the public health benefit of achieving the goals and the deaths that have occurred because of the resource shortfall, we propose alternative measures to screen the at-risk population for consideration. This need is most acute in the black community, in which where screening rates tend to be lower and polyps have been found to progress more quickly than among white populations. In South Carolina, one model has used primary care physicians as the labor force to provide routine screening colonoscopy for their own patients. This model makes screening much more accessible to minority patients, as the wait is shorter and the cost typically lower. In combination with a faith-based partnership with minority religious organizations, this model has begun to make needed inroads toward addressing the disparities associated with colon cancer. Cancer 2007. © 2006 American Cancer Society.

Few diseases offer as much opportunity for both primary and secondary prevention as colon cancer. Nevertheless, the American Cancer Society (ACS) estimates that there will be 148,610 new cases and 55,170 deaths attributed to colorectal cancer nationwide in 2006, of which 2370 new cases and 880 deaths will occur in South Carolina. 1–3 According to the National Cancer Institute, colorectal cancer—cancer that affects the colon, the rectum, or both—is the second leading cause of cancer-related death and the third most common cause of overall premature death in the United States.2 In South Carolina, colorectal cancer is the fourth most commonly diagnosed form of cancer and the second leading cause of death due to cancer, behind lung cancer.2

Contrary to public opinion, colon cancer kills men and women with nearly equal frequency. 2 Although polyps are slightly more common in men, they are more deadly in women, giving rise to virtually equivalent mortality rates.2 Polyps are also an equal-opportunity condition racially, affecting black people and white people with equal frequency.2 The tragedy we discuss in this paper is the alarming mortality disparity: African Americans are 1.67 times more likely to die within 5 years after surgical treatment. 7 The difference lies only partially and to a small extent in the stage of the lesion at detection, which is typically much later in African Americans due to a lack of screening.4 A careful analysis of the data has prompted the American College of Gastroenterology (ACG) to revise its recommendations in March 2005 for African Americans to begin screening colonoscopies at age 45.5 Twice as many cancers are discovered prior to age 50 in this higher-risk population (10.6% vs. 5.5%), evincing a general pattern of increased indolence with age.6 In this paper, we propose to increase access to care and hence screening rates through a two-pronged attack: 1) to increase capacity by training primary care physicians to perform screening colonoscopy and 2) by implementing a faith-based initiative, to address community awareness and increase community support for screening efforts.3

Descriptive Epidemiology

Variations in cancer incidence and mortality rates within the United States can provide etiologic clues and suggestions for applying resources for prevention efforts, such as screening. As a state with an outstanding cancer registry and some of the largest disparities in the United States, South Carolina provides an excellent example for purposes of illustration. There were 2087 incident cases of colorectal cancer in the state in 2001 and 817 deaths in 2003. 7 Of these deaths, 399 (49%) were among men and 418 (51%) were among women. These statistics clearly demonstrate that colorectal cancer is a fatal cancer and that mortality affects men and women about evenly, but for both sexes, incidence rates of colorectal cancer are greater among blacks than among whites. In South Carolina, the incidence rate for colorectal cancer in black males is 71.9 per 100,000, compared with 66.4 per 100,000 in white males. The incidence rate for black females is 52.0 per 100,000/year compared with 44.8 per 100,000/year in white females (Fig. 1).3, 7

Figure 1.

Colorectal cancer age-adjusted incidence rates per 100,000 in South Carolina by race and gender, 1998–2003. Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:214.

If the disease expresses a similar natural history and screening rates are equivalent by race, one would expect mortality to parallel incidence. Unfortunately, as with most other cancers, a disparity exists between the mortality rates for African Americans and whites, with the rates in African Americans tending to amplify incidence rate disparities. By analyzing South Carolina data from 1998 to 2003, we can see that black males have a 46% higher mortality rate than that of white men, and African American women have a 31% higher mortality rate (Fig. 2). 7 In black men, the mortality rate is 34.5 per 100,000/year and 23.6 per 100,000/year among white men. Likewise, the annual mortality rate among black women is higher than that of white women: 21.1 per 100,00 and 16.1 per 100,000 respectively.8

Figure 2.

Colorectal cancer age-adjusted mortality rates per 100,000 in South Carolina by race and gender, 1998–2003. Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:214.

Overall, South Carolinians are diagnosed with colorectal cancer at late stages (59% for whites and 61% for blacks). 7 Figures 3–6 show the colorectal cancer incidence and mortality in the 46 counties in the state relative to their same-race counterparts nationally. These diagrams indicate by race which counties have significantly higher rates of incidence and mortality than the national averages. All said, incidence rates for both white people and black people are significantly higher in many South Carolina counties than the national average (Figs. 3 and 4), particularly among white men.7 Few counties, however, have mortality rates that are significantly higher than average U.S. mortality rates (Figs. 5 and 6).7

Figure 3.

Colorectal cancer age-adjusted incidence rates per 100,000 among white men in South Carolina by county (1998–2003) compared with average U.S. rates among white men (2001). Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:215.

Figure 4.

Colorectal cancer age-adjusted incidence rates per 100,000 among black men in South Carolina by county (1998–2003) compared with average U.S. rates among black men (2001). Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:216.

Figure 5.

Colorectal cancer age-adjusted mortality rates per 100,000 among white men in South Carolina by county (1998–2003) compared with average U.S. rates among white men (2002). Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:216.

Figure 6.

Colorectal cancer age-adjusted mortality rates per 100,000 among black men in South Carolina by county (1998–2003) compared with average U.S. rates among black men (2002). Reprinted with permission from Daguise VG, Burch, JB, Horner M-J, et al. Colorectal cancer disparities in South Carolina: Descriptive epidemiology, screening, special programs, and future direction. J S C Med Assoc. 2006;102:215.

Research shows that greater than 90% of these deaths could have been prevented had the patients undergone routine colonoscopic screening. 9 Although colonoscopy does not detect every lesion, it is able to detect smaller lesions that other tests may neglect. In one study, 58% of patients with negative fecal occult blood tests (FOBT) were found on examination with colonoscopy to have neoplastic lesions.10 Thirty-eight percent of the lesions would have been missed with flexible sigmoidoscopy without any findings of distal lesions.10 It has been demonstrated that fecal DNA tests miss lesions as well. Researchers compared the results of colonoscopy with that of fecal DNA tests performed on stool samples submitted before endoscopy. Fecal DNA tests positively identified colon cancer in 51.6% of patients diagnosed with cancer by colonoscopy; 12.9%, however, were diagnosed with colon cancer by colonoscopy but had negative fecal DNA tests.11 Although CT colonoscopy does not involve endoscopy, the patient still must take the laxative preparation, and the detection rate for lesions 10 mm or larger is 59%, compared with 98% with optical colonoscopy.12 If a lesion is found, CT colonoscopy cannot remove it, and so the patient will have to undergo optical colonoscopy regardless.12 Although indirect tests have a place, the new gold standard is colonoscopy.9

Nonetheless, a 2000 study commissioned by the Governor's Taskforce for Colorectal Cancer Prevention of 4 rural, impoverished South Carolina counties determined that only 8% of the appropriate target population had undergone screening colonoscopy. 13 One reason is undoubtedly the low ratio of gastroenterologists to the private population. As of 2000, South Carolina had only 38 board-certified gastroenterologists to serve its 1.5 million residents over 50 years of age.13 Primary care physicians are the most effective group in the medical community who can address the screening problem, as they form a larger workforce.14 It should be noted that in some areas, nurses and nurse practitioners have been tapped as a labor source to augment screening capacity.15–17 In South Carolina, there is 1 primary care physician for every 500 patients over 50, versus a ratio of 1: 37,500 for gastroenterologists.13, 18

Economic and Social Issues

There are only a few cancer sites (cervix, oral cavity, and colorectum) for which early detection via screening is an effective primary prevention. 19–21 At such sites, preneoplasmic lesions can be detected and removed. Breast cancer screening via mammography, although achieving an 80% screening compliance in those over 50,22 reduces mortality by only 30%.23 For colorectal cancer, neoplastic lesions also can be detected and treated at very early stages, thus making screening one of the most cost-effective prevention procedures.24 Not only does screening for neoplastic growth within the colon and rectum yield highly successful prevention rates, it is one of the most cost-effective prevention procedures in all of medicine.24 The cost-effectiveness ratio compares the cost spent per one life-year saved from cancer death to the number of prevented cases.24 Incremental cost ratios, which are calculated by dividing the difference in cost by the difference in effectiveness of the tests, provide a means to compare the efficacy of similar screening modalities.24 Colonoscopy has an incremental cost-effectiveness ratio of $6600 per life year gained as a result of the procedure.24 This procedure is much less expensive than screening for breast cancer ($22,000 per life year gained), cervical cancer ($250,000 per life year gained), or even heart transplantation ($160,000 per life year gained).24 Compared with citizens of other industrialized nations, Americans spend nearly twice the percentage of the gross national product on healthcare, but the life expectancy continues to remain among the lowest.24 Clearly, alternative strategies must be employed to combat this highly preventable illness.

The cost of colonoscopy ($1500-$5000) has been a tremendous barrier to widespread adoption of this lifesaving procedure. Nevertheless, an increasing number of third-party payers (including Medicare) are covering colonoscopy for screening even the average-risk population. 25 In South Carolina, the House and Senate passed a concurrent resolution, H4756, encouraging the State Health Plan and all health insurers to cover colon cancer screening by colonoscopy, but fell short of requiring such coverage.26 Many insurance plans have begun to cover screening colonoscopy because its cost-effectiveness has been overwhelmingly demonstrated.25 Although FOBT is purported to reduce colorectal cancer deaths by 40%,27 the calculated cost per year of life saved is nearly $82,000.24 The benefit deriving from colonoscopy every 10 years is much larger: only $28,000 for each year of life saved.24 Many social barriers still exist, however. While mammograms and prostate digital rectal exams are slightly demeaning, exposing oneself to colonoscopy is even more daunting. Such barriers are not insurmountable, though, as evidenced by the psychological hurdles that Pap smears have overcome.

Increasing Capacity

If the manpower and infrastructure necessary for this increased capacity of screening colonoscopy were not in place, our efforts to overcome social obstacles would be to no avail. If we are to educate the public more thoroughly on the tremendous benefits of colonoscopy for colorectal cancer screening, we must first ask ourselves: are we prepared to provide these services? Unfortunately, the answer appears to be an overwhelming ‘no.’ Delays for referral appointments for routine screening colonoscopies to gastroenterologists are measured in months. 28 To screen the 1.5 million people in South Carolina's average-risk population would require 150,000 screening colonoscopies yearly, with the most minimal necessary follow-up screenings of 10 year intervals. At the currently observed adenomatous polyp incidence rate, 200,000 repeat colonoscopies would need to be performed annually,29 but current capacity is only 30% of this number. To meet the demand, each gastroenterologist would have to perform 5000 colonoscopies a year—25 each workday.

Fortunately, a vast manpower resource lies practically untapped—primary-care physicians already trained in flexible sigmoidoscopy (in one study, 52% of primary care physicians reported being trained in flexible sigmoidoscopy). 30 The lone difference between a flexible sigmoidoscope and a colonoscope is the length—the instruments are identical in every other respect. Because these physicians have received endoscopy training, they already possess many of the technical skills necessary to command a colonoscope.8 As a result, training programs for these physicians are essentially review sessions rather than de novo training. Although it may be daunting to find time to leave a practice often already strained in finances and staff, the benefits of increasing screenings will reap benefits in terms of quality of care for the practice.31


Physicians were recruited for training through a fairly aggressive public relations campaign. Dr. Lloyd canvassed primary care physicians throughout the greater Columbia area, offering training to them. The physicians utilizing the facility are those physicians who agreed to participate in the training program. Patient populations are screened for healthy, asymptomatic patients, while other patients with health problems are still referred for gastroenterologists for diagnostic workup.

A database focusing on monitoring quality parameters was developed and has been accumulating data prospectively since August 2001 through a program designed for and commissioned by South Carolina Medical Endoscopy Center (SCMEC) using Visual Basic® modules. This software records procedure start times, maximum progression, and end time, as well as any findings, including polyps. From the raw data, the software calculates the cecal intubation rate and the polyp detection rate. Complications are reported by the primary care doctors, as these patients are their own patients and they would have to treat any complications.

Quality assurance is maintained by a committee peer review of every case. Impartial observers are consulted on a periodical basis to supervise cases and perform chart audits. On the basis of recommendations of the observers, changes are made for quality improvements.


Primary Care Physician Training

At SCMEC in Columbia, South Carolina, we have begun training and providing support facilities so that this able population of physicians may perform colonoscopy screening for their own patients. After performing a minimum of 50 supervised procedures, each physician may apply for privileges to perform colonoscopy independently. An expert (a board-certified gastroenterologist, a board-certified colorectal surgeon, or Dr. Lloyd), however, remains on site at the physician's disposal for guidance, a second opinion, or simply another set of eyes. Thus far, 50 physicians have completed the program and have performed over 12,000 procedures. Using three quantitative measures of quality for the exams—completion rate, polyp detection rate, and complication rate—the primary care physicians have demonstrated a high level of competency. Many articles address the completion percentage—the national average is 92%. 32 Our physicians achieve a cecal intubation rate of 98.3%. Reported national rates for polyp detection are 26%, while our primary care physicians average 52.3%.33 Of the 52.3%, 36.0% possess adenomatous or ‘pre-cancerous’ polyps. The third measure, the rate of serious complications due to perforation, is historically predicted to be between 1 per 500 and 1 per 1000.34 Fortunately, though, these primary care physicians have experienced only one fourth of this rate. Admittedly, the primary care physicians screen their healthy asymptomatic patients and continue to refer those patients who pose higher risks or require diagnostic evaluation of symptoms. Using these measures of quality, the tests performed by primary care physicians appear to exhibit a level of quality that meets or exceeds national norms for colonoscopy practice. The key component of our research, therefore, has been the ability to structure a program (physician selection, training, and continued endoscopies) demonstrating as high quality as their colleagues, many of who are specialists. The SCMEC model provides a unique and successfully demonstrated opportunity for primary care physicians to obtain skills to provide this scare resource in an environment conducive to excellent quality.

By conducting screening colonoscopy procedures themselves, primary care physicians are able to take a more aggressive stance against colon cancer, and they take confidence in knowing that they provide a quality of care to their patients equal or superior than their sub-specialty colleagues. The patients may appreciate that they did not have to be referred elsewhere. Currently, the procedures are conducted in areas with relatively good access to care (compared with what occurs in large segments of the South Carolina population). 35–38 As capacity builds, our intention is to increase outreach into more severely underserved populations and in very rural areas. This is in clear recognition that, as with most medical procedures, there are racial and socioeconomic disparities in terms of access to medical care procedures.42 Many of the physicians at SCMEC operate practices that serve lower-income communities. By virtue of socio-economic status, the screening rates of these practices traditionally remained low while patients were being referred out for screening colonoscopy. The primary care physicians reported anecdotally that under the referral system, they had estimated that approximately 80% of their patients were screened. After beginning their own screenings, however, they found that the true rate was in the range of 20%. Now, after performing their own screenings, those rates have jumped to a true 80%. In other areas of the country, similar situations have been reported.31 Colonoscopy has been established as the preferred technique for colon cancer screening over sigmoidoscopy10–12, 39; however, when primary care physicians ceased performing their own sigmoidoscopies and began referring all their patients to gastroenterologists for colonoscopy, only half of these patients were ever screened.31 This is a startling statistic that calls into question another crucial step in the process of providing preventive services to eligible patients. Many primary care physicians erroneously assume their patients are being screened. In South Carolina, despite increased screenings, the colon cancer death rate has increased over the past 3 years,3 which may be attributed to relying on the referral system for colon cancer screenings. By screening their own patients, primary care physicians have taken a more active role in colon cancer prevention because they have a greater investment in the process.

Although we are making headway toward evening the gap in screening rates at this one facility, there continue to be racial disparities across the country. The severe shortage of manpower to provide screenings that would help equalize these racial differences is further exemplified in national training statistics. While slightly more than 1000 physicians are in fellowship training in gastroenterology in the United States, only 41 of them are African American. 40 The SCMEC program in South Carolina has already trained 15 African-American primary care physicians whose practices serve primarily African-American patient populations. These physicians provide 2000 screening colonoscopies each year, in a state that in 2002 only performed 45,000 colonoscopies statewide.7

Community Outreach

In an effort to expand the program's outreach to other African-American patients (those not seeing doctors utilizing the facility), SCMEC began a faith-based initiative to create awareness and educate potential minority participants through their churches. After several less formal forms, this program is now administered in partnership with the Carolina Community-Based Health Support Network (CCBHSN), which has a relationship with the Baptist Education and Missionary Convention of South Carolina. Under the program, individual churches submit a grant application to CCBHSN. Once the application has been processed, a church member serves as coordinator for that church and supervises the education and outreach program at that church. Fortunately, the church's grant includes provisions for indigent patients, preventing the cost of the exam from precluding anyone from screening due to financial reasons.

In addition to financial barriers, the program helps to ease some of the social stigmas associated with the test. Many people are embarrassed or scared to undergo the test, but the church program introduces a communal element to the procedure. Having friends and family members going through the same experience helps to assuage the anxiety and makes it seem less daunting. Several times a church group elected to ride together on the church van, further illustrating the notion of a shared experience. In addition to the comfort of community, another major motivating factor for some has been the detection of a dangerous lesion in one of their fellow parishioners. Once church members see how closely colon cancer can affect them, they hold greater inspiration and resolve to get their own test. This partnership with churches in minority communities has proven to be an effective tool for increasing minority colon cancer screening rates. To date, SCMEC has been able to screen over 30 churches, or more than 1000 patients. This program has allowed another 1000 minority patients to receive screening (92% of whom had not been screened previously). Since November 2005, of 424 patients, 126 patients (30%) had at least 1 adenomatous polyp. Of the adenomatous polyps, 14 (3% of all patients) of these were life threatening (villous, high grade dysplasia, or cancer). Fifty percent of the 14 could be treated endoscopically. The other 7 patients were treated surgically, 6 of whom had adequate insurance. A local colorectal surgeon graciously treated the 1 patient without insurance coverage. Only 1 patient was found to have metastatic disease. In summary, out of more than 1000 patients, only 1 has required indigent care follow-up. These statistics underscore the efficacy of a faith-based initiative to have a positive impact on such a serious public health issue as colon cancer. Linking a larger provider base with the concept of a faith-based approach (especially in South Carolina, where the vast majority of citizens are regular church attenders) 41 has huge implications for improving the public's health.


Clearly, colon cancer is a condition that has unnecessarily wreaked much havoc in the lives of many individuals. As discussed earlier, current infrastructure is inadequate to provide screening for all patients who are considered at-risk. Widespread population-based screening is warranted based on cost-benefit analyses alone. Taking into account the human suffering makes the argument for doing this even stronger. The primary care endoscopy model has been demonstrated to increase patient screening rates and hence has the capacity to increase prevention by detecting lesions before they become problematic. Under this model, including minority primary care physicians who have access to minority patients could help tremendously to increase minority screening rates. Other creative programs, for example, partnerships with organizations associated with established institutions within minority communities, such as churches, will help to increase screening rates and to decrease disproportionate death rates. Until research develops a greater genomic and proteomic understanding of colon polyps and colon cancer to predict how the disease might affect an individual, regular screenings will be necessary. As long as there are disparate risk levels for certain population segments, alternative screening strategies such as primary care endoscopy and church programs will be necessary to maintain comparable levels of care for all population groups.