Screening preferences for colorectal cancer: a patient demographic study.

Background: Colorectal cancer is a leading cause of cancer-related death. Screening for colorectal cancer is a rational and cost-effective strategy for reducing the incidence of colorectal cancer and related mortality. Despite endorsement by academic and health care organizations, patient awareness and compliance with screening is low, partly because of patient-related barriers to screening.

Methods: A convenience sample of adults attending the internal medicine and family practice clinics of a community teaching hospital was studied. A description of fecal occult blood testing (FOBT), sigmoidoscopy, and colonoscopy procedures was given in a packet along with a questionnaire. The questionnaire focused on screening procedures followed in our hospital (i.e., yearly FOBT and sigmoidoscopy every 5 years or colonoscopy every 10 years for average-risk individuals).

Results: Of the 193 patients who responded, 55% preferred sigmoidoscopy and FOBT, 29% chose colonoscopy, and 16% wanted no screening. Those with knowledge of someone with colon cancer or colon polyps reported a significantly higher preference for screening than those without such knowledge. Catholics were most likely to prefer no screening compared with non-Catholics. Ex-smokers (compared with all others) were more likely to want screening. Catholics were least likely to want colonoscopy. Patients with previous experience of colorectal screening preferred future screening. Those preferring no screening were significantly younger than those who expressed a preference for screening.

Conclusion: The results of this study demonstrate diversity in patient choices for colorectal cancer screening. A focus on people's preferences rather than on the test itself may help develop and target appropriate intervention for prevention of colorectal cancer.

Key Words: colorectal cancer, fecal occult blood testing, patient choice, screening

**********

Colorectal cancer is the second most common cause of cancer-related deaths in the United States. (1) In 2001, an estimated 135,400 new cases of colorectal cancer were diagnosed and 56,700 people died as a result of the disease. (1) Almost 75% of colorectal cancers occur in average-risk individuals, and most arise from benign, adenomatous polyps. (2-4) Adenoma-carcinoma sequence, the major pathway for the development of colorectal cancer, evolves slowly for 10 to 20 years. (4) The biologic behavior of colorectal cancer provides a unique prospect to intervene in the preclinical phase to prevent the progression of cancer. Screening provides an opportunity to detect not only precursor lesions but also early-stage and potentially curable cancer. (5-8) Colorectal cancer may be preventable in up to 90% of cases. (5,9) Indeed, the National Polyp Study revealed that polypectomy reduced the incidence of colorectal cancer by 76 to 90%. (6) The overall mortality rate of colorectal cancer approaches 60%. However, advanced colorectal cancer is one of the least curable cancers, with an estimated 5-year survival rate of only 7%. In contrast, the estimated survival can be as high as 92% in patients with early-stage cancer confined to the bowel wall. (10)

Evidence-based practice guidelines for colorectal cancer screening have been developed by professional organizations such as the American Cancer Society, the American College of Physicians, and the United States Preventive Services Task Force. Although the debate over screening average-risk, asymptomatic individuals has been resolved, interest has shifted to determining the preferred screening strategy and finding out the better and more effective ways of implementing screening in the general population. The ongoing drive to perform colorectal cancer screening has been enhanced by the recent data on the accuracy and comprehensiveness of colonoscopy in detecting polyps. (11-13) Medicare's recent decision (14) to reimburse screening colonoscopy has generated both interest and debate on the screening options for average-risk patients over the age of 50. Despite ample evidence supporting colorectal cancer screening, most average-risk individuals have not had screening. Less than 40% of individuals over age 50 report having had a colorectal cancer screening test and less than 30% are up to date with the screening. (15-17) The overall reported compliance rates have been only 15 to 40%. (18) There are unique barriers to colorectal cancer screening, some specific to the individual test strategies and others related to individual subjects who are offered screening. Understanding and addressing these barriers is crucial in designing and implementing screening strategies, as patient compliance is a vital determinant of success and the cost-effectiveness of colorectal screening. (18-21) Most of the current recommendations for colon cancer screening of average-risk individuals have not incorporated patient preferences. Moreover, patient preferences have not been well studied. Patient choices are especially important in colorectal cancer screening because available options differ in effectiveness and likelihood of adverse events. The goal of this study was to identify the attitudes, beliefs, and demographics of adult patients in a community hospital outpatient setting and recognize personal and demographic parameters that may be amenable to change through public education and targeted interventions. The results of this study may help primary care physicians in addressing colon cancer screening with their patients.

Methods

Data Collection

The survey was conducted in the outpatient/ambulatory care setting of a community teaching hospital. A convenience sample of adult patients attending the internal medicine and family practice clinics was chosen on the basis of availability and ease of collecting data. A research assistant handed to all patients in the waiting room packets containing a questionnaire and a detailed description of fecal occult blood testing (FOBT), sigmoidoscopy, and colonoscopy procedures. The research assistant was available to answer only questions about the questionnaire. Questionnaires were available in English and Spanish. Clinical assistants and family members translated the questionnaire for patients who spoke other languages. The questionnaire focused on screening procedures followed in our hospital (i.e., yearly FOBT and sigmoidoscopy every 5 years or colonoscopy every 10 years for average-risk individuals). Patients with history of colonic diseases, significant family history of colon cancer, or inflammatory bowel disease were excluded from the study. Otherwise, everyone who was willing to fill out a questionnaire was eligible. The data collection covered an 8-week period between April and May 2001.

Statistical Analysis

Data were analyzed for the statistical associations between preference for treatment and other factors. For 2 X 2 tables, a Fisher's exact test was used. Otherwise, with nominal variables, a [chi square] test was performed and categories were combined whenever an expected frequency was less than 5. One or two-way analysis of variance was used when age was one of the factors to be investigated. All P values are two-tailed. All confidence intervals are at the 95% level. One hundred ninety-three of 300 questionnaires were completed and appropriate for analysis. Age, sex, ethnicity, religion, education level, annual income, smoking, facility with English, knowledge of someone with colon polyps or cancer, and previous screening history constitute some of the factors of interest (Table 1). Analysis was performed using SPSS software (Release 11.0.1, SPSS, Inc., Chicago, IL).

Results

Some of the demographics of the 193 respondents are listed in Table 1. The questionnaire contained several key questions about attitudes, past experience, and colon cancer awareness. As for screening choice, 55% preferred sigmoidoscopy and FOBT, 29% chose colonoscopy, and 16% wanted no screening (those not expressing an opinion were excluded in the following analyses). Ten percent had previously experienced an FOBT and flexible sigmoidoscopy and 17% had experienced a colonoscopy. Thirty-six percent of responders knew someone with either colon polyps or colon cancer.

Information about those variables significantly associated with desire to screen are listed in Table 2. Subjects who knew someone with colon cancer reported a significantly higher preference for screening than those who did not (94 versus 80%, P = 0.015). Similarly, those who knew someone with colon polyps reported a significantly higher preference for screening than those who did not (98 versus 80%, P = 0.006).

Subjects who knew someone with colon polyps or colon cancer also reported a significantly higher preference for colonoscopy than those who did not (39 versus 24%, P = 0.005). There was a significant relationship between religion and desire to screen. This pattern was consistent across gender. Catholics were most likely to prefer no screening compared with non-Catholics (30 versus 9%, P = 0.001).

There was a significant relationship between smoking status and attitudes toward screening. Although nonsmokers and current smokers were fairly comparable in their preferences for different screening methods, ex-smokers (compared with all others) were more likely to want screening (94 versus 80%, P = 0.045).

Previous screening was associated with a higher desire for future screening. Of those with previous experience, 97% preferred future screening versus only 81% in the group with no previous screening (P = 0.011). This same pattern showed up in the subgroups "with versus without previous colonoscopy" and also "with versus without previous sigmoidoscopy," but differences were not significant due to smaller numbers. A significant relationship between age and screening preference was noted (P = 0.021). Those preferring no screening were significantly younger than those who expressed a preference for screening (38.1 versus 47.6 yr, P = 0.008).

Information about screening preferences are listed in Table 3. This table and the following section cover only the "screeners," those who preferred some form of screening. There were 154 screeners in the sample. Of those who had had a previous colonoscopy, 55% preferred it for repeat screening, compared with only 30% of those who had never had a colonoscopy (P = 0.017). The same pattern was apparent when previous sigmoidoscopy was the factor of interest, although the differences were not quite significant. Among screeners, Catholics were least likely to want colonoscopy (18% versus 42% for all others, P = 0.024). Ex-smoker screeners were more likely to choose colonoscopy (48%) versus other screeners (29%, P = 0.039). Ninety-four percent of responders would recommend their personal preference to a friend or family member.

Discussion

Currently, the overall number of patients screened is low. (15,16,18) With emphasis on shared decision-making and the new reimbursement approval by Medicare, colonoscopy is more likely to be offered as a screening procedure to average-risk patients. We expected that more patients would opt for colonoscopy as the more complete and less frequent examination. The response rate to our questionnaires was 64%, which is comparable to other studies. (22,23) Most of our patients who had not been screened in the past preferred sigmoidoscopy. In a cross-sectional study of patients in an academic medical center clinic, Ling et al (22) had a response rate of 76%. The majority of their patients preferred annual FOBT (43%), followed by colonoscopy every 10 years (40%). Only 12% of patients preferred the combination of annual FOBT and flexible sigmoidoscopy every 5 years. In our study, 29% of patients selected colonoscopy as the test of choice. In the study by Ling et al, (22) patients who considered accuracy the most important attribute of a test preferred colonoscopy (62%). The design of this study was different from ours. The authors used a formalized decision aid intended to educate patients about colorectal cancer screening and offered five screening strategies. Pignone et al (23) surveyed the screening preferences of 146 patients in a university clinic. After descriptive test information, the most preferred screening strategy was FOBT alone (45%), followed by the combination of flexible sigmoidoscopy every 5 years and annual FOBT. However, after information about the test performance, the combination of flexible sigmoidoscopy and FOBT was the most preferred choice (47%). Our observations are consistent with these results, as we also provided descriptive information about both the test performance and invasiveness. However, colonoscopy was not offered as a screening test in the study by Pignone et al. (23) Less than 5% of patients in their study wanted no screening, compared with 16% in our study. Again, the setting of this study was a university hospital clinic and the patient population may conceivably be different.

Catholics, Hispanics, and individuals with less education and at lower income levels were more likely to not want screening in our study. The study by Dominitz and Provenzale (24) suggests that whites are less accepting of endoscopic screening than nonwhites. However, in our survey, whites were the most likely to prefer screening (94%). Ethnic differences in concert with level of education and socioeconomic class among older Americans have been found to determine responses to cancer prevention. (25) Unlike the study by Dominitz and Provenzale, (24) Hispanics in our survey were more likely to not want screening and least likely to prefer colonoscopy. Lower income respondents were also less likely to prefer any screening. These ethnic and socioeconomic groups may have less educational opportunities available to them, or have no time for such interventions because of their working environment. Most Hispanic patients in our area are farm workers. Several studies have suggested that ethnicity is associated with attitudes toward screening. Some ethnic populations may find it embarrassing to discuss such issues openly. These populations may need extra education and advising from their physician to get screened or tested. The incidence of colorectal cancer is higher than in the general population in Hispanics and Native Americans. (26,27) The Surveillance, Epidemiology and End Results program and the American Cancer Society have observed that blacks are at higher risk for developing colorectal cancer at a younger age and have higher mortality than whites. (28) Elderly black men are more likely to not use colorectal cancer screening. (28-31)

An exploration and understanding of beliefs and cultural factors that may affect the decisions of patients in different ethnic and religious groups would be helpful in devising targeted interventional strategies. Our observations suggested that Catholics, Hispanics, and individuals with lower income and less formal education would require the most educational intervention. It is unclear whether the differences in screening preferences stem from religious convictions or cultural background. This aspect has not been looked into and deserves further investigation.

Our patients who preferred colonoscopy may have selected the more invasive, complete, but less frequent procedure because of more knowledge or heightened fear of colon cancer. Like us, Ling et al (22) also noted a significant influence of demographic features on patient preferences for colorectal cancer screening. They found a significant difference in screening preference between genders (P = 0.01), especially in the combination strategy of flexible sigmoidoscopy and FOBT (men, 17%; women, 7%). However, unlike our study, they did not observe any significant differences related to age, race, and educational level. They did note a significant relation between the educational level and the test features considered (P = 0.04), especially the accuracy of the test. Of patients with a greater than high school education, 60% considered accuracy as the most important feature as compared with 48% of those with a high school education or less.

Unwillingness to undergo screening may indicate that the patient perceives invulnerability to colorectal cancer, fears discomfort of the procedure, or believes that early detection of the cancer will not increase survival. Weller et al, (32) in an Australian population-based study, found that although awareness of FOBT was high, only 15% of subjects aged 40 or older had been tested, and only 28% stated the intention to be tested. Many patients in that study denied susceptibility to colorectal cancer, had knowledge deficits related to treatment, or simply felt uncomfortable with the test. Cultural and ethnic barriers may significantly contribute to perceptions and participation in screening. Lacey (33) found that black women perceived personal life stress as a more important health issue than hypertension, diabetes, heart disease, or cancer. Fatalistic cultural perceptions and beliefs of invulnerability may also pose significant barriers to screening. (33-36) Comparison of various Hispanic population subgroups has revealed that Mexican-Americans are least likely to receive cancer screening, based on level of education, income, and insurance coverage. (37,38)

An interesting observation in our survey was a strong preference for colonoscopy among patients who had previously undergone screening with colonoscopy (53 versus 25%) or sigmoidoscopy (53 versus 27%). Leard et al, (39) who noted that 71% of patients who previously had undergone a colonoscopy preferred a repeat colonoscopy, also reported this observation. This may be the result of education and information acquired at previous screening tests or, as speculated by Leard et al, (39) the patients may have chosen colonoscopy because colonoscopy is considered the "gold standard." Ling et al, (22) however, noted almost similar preferences for colonoscopy (41%) and FOBT (42%) in their group of patients. Our survey also revealed that, of the 22% of patients who had previous colon cancer screening with a colonoscopy or sigmoidoscopy, 85% indicated an intention to undergo testing in the future. This suggests that our population is more willing to undergo screening procedures and is open to engaging in a discussion about screening. Again, this may also have been the result of education and awareness acquired from the previous procedures.

Approximately 30% of our study respondents preferred screening by colonoscopy rather than sigmoidoscopy/FOBT. We do not know whether they chose the former test because the examination of the entire colon was more reassuring or because of the comfort of sedation and a longer interval between procedures. Ling et al (22) observed that 54% of patients in their survey considered accuracy as the most important feature of the screening test and only 15% considered personal discomfort as the most important feature. A majority of our patients (55%) preferred a combination of FOBT and sigmoidoscopy. However, Ling et al (22) noted that only 12% of patients in their survey preferred the combination. In a survey of outpatients, Leard et al (39) reported that colonoscopic screening was preferred to sigmoidoscopy or barium enema. They also noted a relatively weak preference (36%) for repeat FOBT in those who had previously experienced FOBT. They postulate that a patient who had a negative FOBT in the past may feel that a repeat evaluation may not be worthwhile. In a hospital-based randomized trial of screening sigmoidoscopy versus screening colonoscopy, patients thought that preparation for sigmoidoscopy was easier. (40) However, sigmoidoscopy was considered more uncomfortable and embarrassing, perhaps due to lack of sedation during the procedure.

In our survey, more than half (53%) of those individuals who have had a previous sigmoidoscopy expressed a preference for colonoscopy. Of those patients who have not experienced a screening test in the past, just over one in four (27%) preferred colonoscopy. It is possible that these patients either gained more knowledge and information at the time of sigmoidoscopy and wanted a higher level of testing the next time, or were dissatisfied or uncomfortable with sigmoidoscopy and wanted to try something different.

Sigmoidoscopy was the screening test of choice for the majority of patients who answered our survey. Sigmoidoscopy was also the screening test of choice for those who had not had past colorectal cancer screening. This observation was the same for all ethnic groups.

Interestingly, ex-smokers in our survey were more likely than all others, including smokers and nonsmokers, to prefer screening. This may indicate the effectiveness of previous educational interventions to help these patients stop smoking.

More than one third (36%) of our survey respondents reported knowing someone with either colon polyps or colon cancer. Over 90% would recommend their personal preferences to a family member or friend. This leads us to believe that patient awareness of risks and the discussion of colorectal screening information can help other people become more receptive to, and aware of, screening.

Conclusion

Although there are signs of improvement in colon cancer survival, the fact remains that 20% of patients with colorectal cancer present with metastatic disease. This is a disease where health care professionals can truly have a positive impact by discussing the appropriateness of different screening procedures with their patients, with an emphasis on underinformed populations and ethnic groups. Screening decisions are often made with little input from patients. This study shows that patients have preferences and that these preferences are diverse, based on the patient's beliefs, religion, ethnicity, and other social and demographic factors. This makes informed shared decision-making an important means of improving screening compliance, screening choices, and patient satisfaction. The United States is a multicultural society. To be successful, the practice of medicine, especially preventive medicine, must be based on the understanding of diverse beliefs and ethnic and cultural value systems of various patient populations. Colon cancer screening education should be particularly targeted toward minority ethnic groups such as Hispanics, because this group is one of several that may have higher incidences of colon cancer than the general population. (26) The results of this study demonstrate diversity in patient choices for colorectal cancer screening. A focus on patient preference rather than on the test itself may help develop and target appropriate intervention for prevention of colorectal cancer. We hope that this survey of our community hospital population opens up discussion and allows for increased education geared toward target groups. Further research is needed to determine the factors that can causally determine attitudes about screening.

Respondents were required to read either English or Spanish. A significant number of patients (36%) did not return the questionnaires. This lack of participation may have contributed to a sample bias. These individuals either did not want to fill out the questionnaire or had no interest in colon cancer screening. This significant number of nonrespondents may indicate a group that may need to be explored for the reasons of their nonparticipation. Also, on the basis of the patient population seen at our hospital, fewer Spanish-speaking patients answered the surveys than we expected. Even with low numbers of respondents, this particular population had some very distinct preferences in preventive testing. Although cost has also been shown to be a determinant of patient preferences, our study did not address the issue of cost. With the new Medicare reimbursement schedule, the cost of screening tests to our patients may be of less concern because Medicare/MediCal covers most of our patients.

Table 1. Demographics

                         No. (%)
Characteristic          (n = 193)

Sex
  Male                   64 (33%)
  Female                129 (67%)
Ethnicity
  White                  92 (48%)
  Hispanic               38 (20%)
  Black                  30 (16%)
  Asian                  15 (8%)
  Other                  16 (8%)
Language
  English-capable       175 (94%)
  Spanish only           12 (6%)
Religion
  Catholic               60 (33%)
  Protestant             37 (20%)
  Other                  87 (47%)
Annual income
  <$15,000               98 (54%)
  $15,000-$30,000        56 (31%)
  >$30,000               28 (15%)
Education
  No school              16 (9%)
  Some high school       38 (20%)
  High school            47 (25%)
  Some college           53 (29%)
  College or higher      32 (17%)
Smoking status
  Never smoked           72 (37%)
  Ex-smoker              53 (28%)
  Present smoker         67 (35%)
Insurance
  None                   85 (45%)
  Only MediCal           43 (23%)
  Only Medicare          11 (6%)
  MediCal and Medicare   11 (6%)
  Other                  38 (20%)

Table 2. Variables associated with desire to screen

                        Prefer      Prefer no
                      screening     screening
                      (n = 154)     (n = 29)
Characteristic       No.  (Row %)  No.  (Row %)  P value

Smoking status                                    0.045
  Never               55  (77.5%)   16  (22.5%)
  Current             51  (83.6%)   10  (16.4%)
  Ex-smoker           48  (94.1%)    3   (5.9%)
Religion                                          0.001
  Catholic            39  (69.6%)   17  (30.4%)
  Protestant          35  (94.6%)    2   (5.4%)
  Other               73  (89.0%)    9  (11.0%)
Education level                                   0.000
  Less than high      35  (67.3%)   17  (32.7%)
    school
  High school or     115  (91.3%)   11   (8.7%)
    more
Knows someone with                                0.006
    colon polyps?
  Yes                 39  (97.5%)    1   (2.5%)
  No                 114  (80.3%)   28  (19.7%)
Knows someone with                                0.015
    colon cancer?
  Yes                 50  (94.3%)    3   (5.7%)
  No                 104  (80.0%)   26  (20.0%)
Previous colon or                                 0.011
    sigmoidoscopy?
  Yes                 37  (97.4%)    1   (2.6%)
  No                 117  (80.7%)   28  (19.3%)
Ethnicity                                         0.000
  Hispanic            20  (55.6%)   16  (44.4%)
  White               84  (94.4%)    5   (5.6%)
  Other               49  (86.0%)    8  (14.0%)

Table 3. Variables associated with screening choice (a)

                        Prefer         Prefer
                    sigmoidoscopy   colonoscopy
                      (n = 100)      (n = 54)
Characteristic      No.  (Row %)   No.  (Row %)  P value

Smoking status                                    0.039
  Ex                25   (52.1%)   23   (47.9%)
  Others            75   (70.8%)   31   (29.2%)
Religion                                          0.024
  Catholic          32   (82.1%)    7   (17.9)
  Protestant        19   (54.3%)   16   (45.7%)
  Other             44   (60.3%)   29   (39.7%)
Had a previous                                    0.017
    colonoscopy?
  Yes               13   (44.8%)   16   (55.2%)
  No                87   (69.6%)   38   (30.4%)
Had a previous                                    NS
    sigmoidoscopy?
  Yes                8   (47.1%)    9   (52.9%)
  No                92   (67.2%)   38   (32.8%)
Ethnicity                                         0.017
  Hispanic          17   (85.0%)    3   (15.0%)
  Caucasian         37   (44.0%)   47   (56.0%)
  Other             13   (26.5%)   36   (73.5%)

(a) NS, not significant.

Accepted January 15, 2003.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9703-0224

References

1. American Cancer Society. Cancer Facts and Figures 2001. Atlanta, GA, American Cancer Society, Inc. Available at: http://www.cancer.org/downloads/STT/F & F2001.pdf. Accessed May 16, 2003.

2. Parkin DM, Pisani P, Ferlay J. Global cancer statistics. CA Cancer J Clin 1999;49:33-64, 1.

3. Peeters M, Haller DG. Therapy for early-stage colorectal cancer. Oncology (Huntingt) 1999;13:307-321.

4. Winawer SJ, Zauber AG. The advanced adenoma as the primary target of screening. Gastrointest Endosc Clin N Am 2002;12:1-9, v.

5. Winawer SJ, Fletcher RH, Miller L, et al. Colorectal cancer screening: Clinical guidelines and rationale. Gastroenterology 1997;112:594-642.

6. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy: The National Polyp Study Workgroup. N Engl J Med 1993;329:1977-1981.

7. Winawer SJ, Zauber AG, O'Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps: The National Polyp Study Workgroup. N Engl J Med 1993;328:901-906.

8. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61:759-767.

9. Lieberman DA. Cost-effectiveness model for colon cancer screening. Gastroenterology 1995;109:1781-1790.

10. American Cancer Society. ACS cancer facts and figures. Cancer Pract 2000;8:105-108.

11. Lieberman DA, Weiss DG, Bond JH, et al. Use of colonoscopy to screen asymptomatic adults for colorectal cancer: Veterans Affairs Cooperative Study Group 380. N Engl J Med 2000;343:162-168.

12. Lieberman DA, Weiss DG. One-time screening for colorectal cancer with combined fecal occult-blood testing and examination of the distal colon. N Engl J Med 2001;345:555-560.

13. Imperiale TF, Wagner DR, Lin CY, et al. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. N Engl J Med 2000;343:169-174.

14. Scala M. Medicare preventive services. Issue Brief Cent Medicare Educ 2001;2:1-8.

15. Anderson LM, May DS. Has the use of cervical, breast, and colorectal cancer screening increased in the United States? Am J Public Health 1995;85:840-842.

16. Bloom Y, Figgs LW, Baker EA, et al. Data uses, benefits, and barriers for the behavioral risk factor surveillance system: A qualitative study of users. J Public Health Manag Pract 2000;6:78-86.

17. Ounpuu S, Krueger P, Vermeulen M, et al. Using the U.S. Behavior Risk Factor Surveillance System's health related quality of life survey tool in a Canadian city. Can J Public Health 2000;91:67-72.

18. Burt RW. Colon cancer screening. Gastroenterology 2000;119:837-853.

19. Costanza ME, Hoople NE, Gaw VP, et al. Cancer prevention practices and continuing education needs of primary care physicians. Am J Prev Med 1993;9:107-112.

20. Pommerenke FA, Dietrich A. Improving and maintaining preventive services: Part 2--Practical principles for primary care. J Fam Pract 1992;34:92-97.

21. Wender RC. Cancer screening and prevention in primary care: Obstacles for physicians. Cancer 1993;72(3 Suppl):1093-1099.

22. Ling BS, Moskowitz MA, Wachs D, et al. Attitudes toward colorectal cancer screening tests. J Gen Intern Med 2001;16:822-830.

23. Pignone M, Bucholtz D, Harris R. Patient preferences for colon cancer screening. J Gen Intern Med 1999;14:432-437.

24. Dominitz JA, Provenzale D. Patient preferences and quality of life associated with colorectal cancer screening. Am J Gastroenterol 1997;92:2171-2178.

25. Hegarty V, Burchett BM, Gold DT, et al. Racial differences in use of cancer prevention services among older Americans. J Am Geriatr Soc 2000;48:735-740.

26. Chao A, Gilliland FD, Hunt WC, et al. Increasing incidence of colon and rectal cancer among Hispanics and American Indians in New Mexico (United States), 1969-1994. Cancer Causes Control 1998;9:137-144.

27. Brown MO, Lanier AP, Becker TM. Colorectal cancer incidence and survival among Alaska Natives, 1969-1993. Int J Epidemiol 1998;27:388-396.

28. Mayberry RM, Coates RJ, Hill HA, et al. Determinants of black/white differences in colon cancer survival. J Natl Cancer Inst 1995;87:1686-1693.

29. Baquet CR, Commiskey P. Colorectal cancer epidemiology in minorities: A review. J Assoc Acad Minor Phys 1999;10:51-58.

30. Hoffman-Goetz L, Breen NL, Meissner H. The impact of social class on the use of cancer screening within three racial/ethnic groups in the United States. Ethn Dis 1998;8:43-51.

31. Roetzheim RG, Pal N, Tennant C, et al. Effects of health insurance and race on early detection of cancer. J Natl Cancer Inst 1999;91:1409-1415.

32. Weller DP, Owen N, Hiller JE, et al. Colorectal cancer and its prevention: Prevalence of beliefs, attitudes, intentions and behaviour. Aust J Public Health 1995;19:19-23.

33. Lacey L. Cancer prevention and early detection strategies for reaching underserved urban, low-income black women: Barriers and objectives. Cancer 1993;72(3 Suppl):1078-1083.

34. Gregg J, Curry RH. Explanatory models for cancer among African-American women at two Atlanta neighborhood health centers: The implications for a cancer screening program. Soc Sci Med 1994;39:519-526.

35. Lantz PM, Reding D. Cancer: Beliefs and attitudes of migrant Latinos. JAMA 1994;272:31-32 (letter).

36. Lantz PM, Dupuis L, Reding D, et al. Peer discussions of cancer among Hispanic migrant farm workers. Public Health Rep 1994;109:512-520.

37. Solis JM, Marks G, Garcia M, et al. Acculturation, access to care, and use of preventive services by Hispanics: Findings from HHANES 1982-1984. Am J Public Health 1990;80(Suppl):11-19.

38. Trevino FM, Moyer ME, Valdez RB, et al. Health insurance coverage and utilization of health services by Mexican Americans, mainland Puerto Ricans, and Cuban Americans. JAMA 1991;265:233-237.

39. Leard LE, Savides TJ, Ganiats TG. Patient preferences for colorectal cancer screening. J Fam Pract 1997;45:211-218.

40. Elwood JM, Ali G, Schlup MM, et al. Flexible sigmoidoscopy or colonoscopy for colorectal screening: A randomized trial of performance and acceptability. Cancer Detect Prev 1995;19:337-347.

RELATED ARTICLE: Key Points

* There is a significant diversity of patient choices for colorectal cancer screening.

* Patients' preferences are related to religion, ethnicity, and other social and demographic factors.

* A focus on patient preference rather than on the test itself may help develop and target appropriate interventions for prevention of colorectal cancer.

* Colorectal cancer screening education should be targeted toward minority ethnic and educationally underprivileged populations.

Rafiq A. Sheikh, MD, MRCP(UK), Sheela Kapre, MD, Olga M. Calof, MD, Coburn Ward, PHD, and Ashu Raina, MD

From the San Joaquin General Hospital, Stockton, CA.

Reprint requests to Rafiq A. Sheikh, MD, Division of Gastroenterology, Department of Medicine, Medicine A, Kaiser Permanente Medical Center, 6600 Bruceville Road, Sacramento, CA 95823. Email: [e-mail]