Fax: (206) 744-9917
Promoting culturally appropriate colorectal cancer screening through a health educator
A randomized controlled trial
Article first published online: 24 JUL 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 5, pages 959–966, 1 September 2006
How to Cite
Tu, S.-P., Taylor, V., Yasui, Y., Chun, A., Yip, M.-P., Acorda, E., Li, L. and Bastani, R. (2006), Promoting culturally appropriate colorectal cancer screening through a health educator. Cancer, 107: 959–966. doi: 10.1002/cncr.22091
- Issue published online: 21 AUG 2006
- Article first published online: 24 JUL 2006
- Manuscript Accepted: 11 MAY 2006
- Manuscript Revised: 27 APR 2006
- Manuscript Received: 7 NOV 2005
- National Cancer Institute. Grant Number: CA 92432
- colorectal cancer screening;
- fecal occult blood test;
- health educator;
- Chinese American;
- Asian American
Colorectal cancer (CRC) is a leading cause of cancer mortality in the US. Surveys reveal low CRC screening levels among Asians in the US, including Chinese Americans.
A randomized controlled trial was conducted with Chinese patients to evaluate a clinic-based, culturally and linguistically appropriate intervention promoting fecal occult blood test (FOBT) screening. The multifaceted intervention included a trilingual and bicultural health educator, bilingual materials (a video, a motivational pamphlet, an informational pamphlet, and FOBT instructions), and three FOBT cards. Patients in the control arm received usual care. Our primary outcome measure was FOBT screening within 6 months after randomization. The proportion of FOBT completion in the intervention and control arms was compared by using a chi-square test, and logistic regression analysis was performed to adjust for the effects of sociodemographic variables and prior screening history. Potential effect modifications were also tested by using logistic regression models.
Our intervention had a strong effect on FOBT completion (intervention group, 69.5%; control group, 27.6%), and the adjusted odds of FOBT slightly increased to over 6-fold greater in the intervention arm compared with the control arm. No effect modification by age, gender, language, insurance, or prior FOBT was found.
The authors' multifaceted, culturally appropriate intervention significantly increased FOBT screening in a group of low-income and less-acculturated minority patients. Given the large effect size, future research should determine the effective core component(s) that can increase CRC screening in both the general and minority populations. Cancer 2006. © 2006 American Cancer Society.
In the US, colorectal cancer (CRC) is the second leading cause of cancer mortality and will contribute to 55,170 deaths in 2006.1 The lifetime risk of dying from CRC approaches 3%.2 Survival from CRC is closely related to stage of disease at the time of diagnosis, with 5-year survival rates exceeding 90% for those with localized disease. In comparison, 5-year survival is about 60% when regional lymph nodes are involved and less than 10% for persons with distant metastases.2, 3 However, over one-half of all CRC cases are diagnosed with regional or distant disease.2
A growing body of evidence indicates that CRC deaths could be greatly reduced through appropriate screening.4 However, CRC screening continues to be underutilized, particularly by racial/ethnic minorities.2, 5–7
Randomized controlled trials have provided compelling evidence for the effectiveness of fecal occult blood testing (FOBT) in reducing CRC mortality by using samples from three successive stools.4, 8–10 The US Preventive Services Task Force, the American Cancer Society, and the American Gastroenterological Association all recommend CRC screening for average-risk persons age 50 and over using 1 of the following techniques: annual FOBT; sigmoidoscopy every 5 years; annual FOBT combined with sigmoidoscopy every 5 years; barium enema every 5 years; or colonoscopy every 10 years.4, 11–13
There is up to a 25-fold variation in CRC occurrence worldwide, with the lowest incidence in Asia and Africa and the highest incidence in North America, Western Europe, and Australia/New Zealand.14 Over the past few decades, however, there have been remarkable changes in the CRC incidence in Asian countries, particularly those more developed and westernized. For example, in Hong Kong the CRC incidence in men increased from 29.5/100,000 in 1983 to 59.5/100,000 in 2002.15 Increases in CRC incidence are also reported in China, Japan, South Korea, Singapore, and Taiwan.16 In fact, 1993–1997 data from the International Agency for Research on Cancer show that age-standardized CRC incidence is greater in Singapore and in several areas in Japan than in the US.17
Earlier studies of migrants from low-risk to high-risk areas in the 1970s and 1980s revealed a rapid increase in CRC incidence within the first generation.18–20 Parkin et al.21 reported lower CRC incidence rates in Asia than in the US for Japanese, Chinese, Filipino, and Korean men. Similarly, Yu et al.22 found that Americans had CRC incidence rates four times higher than Chinese rates; however, CRC incidence rates in Chinese Americans were only slightly lower than American rates and the differences were not statistically significant. In a more recent study, foreign-born Chinese men had about the same incidence of CRC as US-born white men, whereas US-born Chinese men experienced slightly reduced rates.23 Based on data from 2002, Japanese individuals born in the US now have higher CRC rates than those of US whites, with the rates of Japanese residing in Hawaii and Los Angeles among the highest in the world.14
According to ethnicity-specific data from California, CRC ranked among the top 3 most common cancers in both men and women from several Asian ethnic groups.24 Cancer registry data from 1990–2002 indicate that CRC incidence rates among Asians residing in the greater Bay area of northern California were the highest among Chinese and Japanese.24 Elevated risks for AJCC Stage III CRC have also been observed in several Asian ethnic groups.25 Moreover, foreign-born Asian Americans and Pacific Islanders presented at a later stage of disease than US-born individuals.26
A recent study showed Asian men were significantly less likely to have received a prior sigmoidoscopy compared with white men.27 Two convenience surveys of the largest US Asian subgroup, Chinese Americans, revealed low CRC screening rates: 11% and 16% reported FOBT in the past year; 22% reported FOBT in the past year with sigmoidoscopy in the past 5 years; and 30% reported sigmoidoscopy in the past 5 years.28, 29
To promote CRC screening among lower-income and less-acculturated Chinese Americans, we conducted a randomized controlled trial to evaluate a clinic-based, culturally appropriate program that promoted FOBT screening through a health educator.
MATERIALS AND METHODS
This study was conducted in collaboration with the International Community Health Services (ICHS), a community clinic serving predominantly Asians in the metropolitan area of Seattle, Washington. ICHS provides comprehensive primary care services at its 2 clinics, the International District Clinic and the Holly Park Clinic. This study only involved patients at the International District Clinic, and all procedures were approved by the Human Subjects Committee of the University of Washington in Seattle.
At the time of the study, ICHS did not have any special policies or procedures addressing CRC screening, and medical charts included only a form to document preventive tests and services. Usual care for CRC screening at ICHS consisted primarily of FOBT that had to be ordered by ICHS primary care providers (physicians, physician assistants, and nurse practitioners). Medical assistants implemented all orders for FOBT and instructed patients to return completed FOBT cards to the clinic laboratory for processing and documentation of results on a laboratory form. The laboratory forms were then forwarded to medical records and incorporated into the patients' medical charts.
Barium enema is rarely used by primary care physicians for CRC screening.30–32 Similarly, ICHS providers did not recommend barium enemas for this purpose. With respect to endoscopic screening, ICHS had the capacity to perform sigmoidoscopies, but this screening modality was infrequently used, and ICHS primary care providers referred patients for screening colonoscopy only at the patients' request.
The focus of our intervention therefore was a clinic-based education promoting FOBT screening through a trilingual (English, Cantonese, and Mandarin) and bicultural Chinese American health educator. The health educator previously worked at the International District Clinic for several years as a medical assistant. We developed educational materials based on findings from our earlier qualitative study using semistructured interviews and focus groups.33 Our motivational video on CRC screening was produced in Cantonese and dubbed into Mandarin. We also developed a bilingual motivational pamphlet and an FOBT instruction sheet. A bilingual CRC informational pamphlet from the Federation of Chinese American and Canadian Medical Societies completed our intervention print materials.
Using the ICHS administrative computer database, we identified 1,623 unique Chinese patients who were 50–78 years of age and spoke Cantonese, Mandarin, and/or English. We excluded 24 patients who had participated in our qualitative interviews to discuss CRC screening. Based on health insurance account numbers, home addresses, and telephone numbers, a computer program then randomly excluded all but 1 member from households with 2 or more patients eligible for our study sample. ICHS employees and patients without addresses were also excluded from the study sample.
A bilingual letter, in Chinese and English, from the medical director was subsequently sent to the 1,240 age-eligible patients to introduce the study. This letter did not mention CRC or CRC screening and contained the following explanation:
“The staff of the International District Clinic is always interested in improving the quality of care we provide. So sometimes we try new ways of doing things with some of our patients but not with others. For example, we might try a different method of giving health information to patients. If the new method proves to be better than the old, we can then adopt the change for all our patients. We compare different methods only in ways that do not interfere with your medical care. You and your physician are always free to decide the medical tests or treatments you receive.” The letter also included a telephone number that patients could call if they had any questions or chose not to participate.
After the introductory mailing, 102 patients or their family members called to indicate that they did not want to participate, leaving 1,138 age-eligible patients for the randomized trial.
Our intervention trial extended over 14 months, from July 2003 to September 2004. During this period, a computer program identified the 1,138 age-eligible patients from the computerized scheduling system of ICHS. Using appointment lists of patients, the health educator reviewed medical records before the scheduled appointments of any age-eligible patient to determine the following exclusion criteria for the randomized trial: less than 12 months of medical care at ICHS; history of CRC; end-stage disease (e.g., congestive heart failure and chronic obstructive pulmonary disease requiring oxygen); gastrointestinal symptoms requiring diagnostic work-up; and adherence to CRC screening. Adherence to CRC screening was defined as FOBT in the past year (with or without sigmoidoscopy) or colonoscopy in the past 10 years. Because some CRC screening guidelines include sigmoidoscopy every 5 years combined with annual FOBT, we elected to use this criterion in our study's definition of adherence to CRC screening instead of sigmoidoscopy alone every 5 years.4, 11–13
On the day of the scheduled clinic appointments, the health educator checked for any clinical problems that would further exclude eligible patients from the trial (e.g., uncontrolled hypertension). The health educator then introduced the trial to eligible patients and presented them with a note giving them the option to opt out of the trial. This note contained the same explanation as the introductory letter. Because the study was considered minimal risk, review by the University of Washington Human Subjects Committee concluded that a signed written informed consent was not required.
After patients agreed to participate in the trial, the health educator strictly followed a randomization table to assign eligible patients in chronological order into intervention or control status. This table, prepared by the study biostatistician using the random number generator of R software, randomly assigned patients to intervention or control status.
Patients randomized into the intervention arm received the CRC screening education from the health educator, including our motivational video, our bilingual CRC motivational pamphlet, the bilingual informational pamphlet, and an FOBT kit with instructions in Chinese and English. The FOBT kit consisted of three FOBT cards in a stamped envelope with the ICHS mailing address. Health education occurred either before or after the participants' medical visit, and depending on their preference, participants had the option to take the video home to review. Patients randomized into the control arm received usual care.
When 2 or more eligible patients were scheduled too closely in time, the health educator used a selection process that randomly picked 1 patient for the intervention trial. The other patient(s) were returned to the sample to be recruited at a subsequent visit. Eligible patients who canceled, missed, or rescheduled their appointments were also returned to the sample for a possible future entry into the trial.
Our outcome measure, FOBT screening within 6 months of randomization, was based on chart audit by a study personnel who was blinded to the patients' trial arm assignments. Only patients with 3 FOBT cards documented in their medical records were considered as having completed FOBT screening.
The primary statistical analysis of our trial results was a 2-sample binomial test of FOBT completion proportions between the intervention and control arms (a chi-square test). We used logistic regression models to adjust for potential confounding effects of sociodemographic variables and prior screening history. A priori, we also hypothesized 5 potential modifiers of intervention effects, namely, age, gender, language, insurance, and prior FOBT. Each of the 5 potential effect modifiers was tested by using logistic regression models with the main effects and interaction terms of the trial arm and the potential effect modifier.
Among 386 patients who had a scheduled appointment during the 14-month trial period, 152 patients were determined to be ineligible by chart review. Four patients were ineligible because of history of CRC, 2 because of end-stage disease, and 1 because of Alzheimer disease. Of the remaining 145 patients, 62 had only FOBT within the past year, 34 had only colonoscopy within the past 10 years, and 3 had both FOBT within the past year and colonoscopy within the past 10 years. The remaining 46 patients had colonoscopies with abnormal findings (e.g., polyps) in the past 10 years.
At the index clinic visit, 18 patients refused to participate in the trial, 5 patients were excluded because of acute medical problems, and 1 patient was excluded because of dementia. A total of 210 patients were enrolled in the trial, with one-half randomized into the intervention arm and the other half randomized into the control arm.
As presented in Table 1, over half of the trial participants were in the 50–64 age group and a majority spoke Cantonese. Reflecting the patient demographics at the clinic, nearly two-thirds of the patients were women and over 80% had public health insurance.
|Intervention (n = 105) %||Control (n = 105) %||P|
Based on medical records, 40% of the trial participants had a prior FOBT, only 5% had a prior sigmoidoscopy, and 0% had a prior colonoscopy. Chi-square tests and, when appropriate, Fisher exact tests, showed no statistically significant differences in age, language, gender, insurance, and prior screening between intervention and control patients.
Results from this trial demonstrate that our intervention program had a strong effect. Among the intervention patients, 69.5% received FOBT screening, compared with 27.6% of control patients, with a resulting intervention effect size of 42 percentage points. The crude odds ratio (OR) for FOBT screening within 6 months of randomization by the intervention arm was 5.98 (95% CI = 3.29, 10.85). Adjusting for age, the OR of FOBT screening within 6 months of randomization by the intervention arm was 5.91 (95% CI = 3.25, 10.75). In logistic regression models adjusting for additional covariates, the odds of FOBT screening within 6 months of randomization slightly increased to over 6-fold greater in the intervention arm than in the control arm (Table 2). No effect modification by age (P = .59), gender (P = .29), language (P = .82), insurance (P = .50), or prior FOBT (P = .99) was found.
|Crude odds ratio||Odds ratio adjusted for sociodemographic variables||Odds ratio adjusted for sociodemographic and screening variables|
|Intervention||5.98 (3.29, 10.85)||6.18 (3.36,11.39)||6.38 (3.44, 11.85)|
|50–64||1.47 (0.77,2.80)||1.53 (0.79, 2.95)|
|Cantonese||0.87 (0.42, 1.80)||0.88 (0.42, 1.84)|
|Female||1.09 (0.58, 2.04)||1.08 (0.57, 2.04)|
|Private||0.38 (0.14, 1.04)||0.36 (0.13, 1.00)|
|None||0.37 (0.09, 1.53)||0.30 (0.07, 1.31)|
|Yes||0.88 (0.47, 1.68)|
|Yes||3.05 (0.76, 12.29)|
To our knowledge, this is the first intervention study promoting CRC screening using a health educator in the primary care setting. Our results confirm the notable effectiveness of a multicomponent, culturally appropriate health education program to promote FOBT screening within an ethnic minority group.
As with other screening modalities, lack of adherence to CRC screening has been found to be associated with lower income, lower educational level, lack of insurance, and racial/ethnic minority status.2, 6, 34–37 Specific barriers to FOBT screening include low levels of knowledge about CRC and lack of familiarity with screening guidelines and tests.2, 38–40 Additional contributors to low screening include lack of perceived susceptibility to CRC, perceptions that screening is unnecessary in the absence of symptoms, beliefs that the test is embarrassing and unpleasant, and fears of abnormal results and surgery.41 Studies have also identified inconvenience, concern about financial costs, lack of availability and access to screening tests, and lack of a doctor's recommendation as deterrents to CRC screening.39–42
Among studies including Asians, similar barriers have been identified.29, 43–46 Bastani et al.47 found that feeling “violated,” not knowing where to obtain screening, and difficulties in obtaining an appointment were additional barriers to CRC screening. Of note, several studies show that foreign birth and low acculturation (fewer years of residence in the US and limited English proficiency) are negatively associated with CRC screening.28, 29, 48–50
Our study targeted lower income and less acculturated Chinese immigrants. The culturally and linguistically appropriate intervention addressed knowledge and barriers through the health educator, the video, and the pamphlets. Intervention participants also had access to FOBT kits. Such a multifaceted approach proved effective in promoting FOBT screening.
Consistent with experts' recommendations, our intervention was designed to integrate CRC screening into the primary care setting and involve nonphysician members of the primary care team.30, 37, 51 Previous studies have focused on the role of physician assistants in performing endoscopy as well as the role of nurses in promoting CRC screening.52–55 Thompson et al.55 found that having licensed practical nurses order FOBT in addition to the existing practice significantly increased FOBT orders. A recent study also demonstrated increased compliance with FOBT screening through intensive patient education by primary care nurses.56
With respect to screening in ethnic minority groups, the paucity of data on effective primary care cancer control interventions prompted Zapka and Lemon36 to highlight the need to understand and address unique cultural characteristics in order to reduce inequalities in screening rates.36, 51, 57
Promoting cancer screening through a health educator has been previously tested, albeit in a community setting focusing on Pap testing in Native American women.58 In this study, a health educator assured successful FOBT screening by addressing many challenges encountered in busy primary care practices. Our method capitalized on the primary care encounter to provide cancer prevention education, but depending on the specifics of each patient encounter, the health educator adapted the education program in order not to disrupt clinic flow. The salary of a health educator is lower than that of primary care nurses. Most important, promoting CRC screening through a health educator did not encroach on the ever-decreasing encounter time between patients and primary care providers.
A major strength of our study is its research design. The effectiveness of our intervention was determined by a randomized controlled trial, with FOBT completion among control patients reflecting background clinical activity. The trial most likely contributed to increased awareness of CRC screening among clinic staff in general, and may therefore have led to a conservative estimate of the intervention effect. Another strength of the study is its assessment of outcome through chart audits by a staff member who was blinded to the patients' trial status. Overestimation of self-reported cancer screening has been documented, and it presents a challenge to determining accurate screening rates.59–64 Recently, a study of Vietnamese participating in focus groups revealed confusion of FOBT with stool testing for ova and parasites. This led Walsh et al.46 to caution against overestimation of self-reported FOBT use.
Our study must be interpreted in consideration of some potential limitations. Whereas we used the gold standard of chart audits to determine our screening outcome, it is possible that ICHS patients sometimes seek care from other providers and may obtain FOBT screening through other providers. Given the culturally and linguistically appropriate services of ICHS as well as the comprehensive care it provides, clinical experience indicates that patients who do seek care from another provider usually do so for acute issues rather than for preventive care. In fact, it has been postulated that the low cancer screening rates among Asian Americans are partly due to the lack of a preventive orientation among providers trained in Asia.65, 66
The small sample size of this study also limited the precision of some estimates (Table 2). Specifically, whereas our results showed that patients with no insurance and patients with private insurance were less likely to complete FOBT testing than were patients on public insurance, due to our small sample size there were no statistically significant differences. Similarly, neither prior FOBT nor prior sigmoidoscopy were significantly associated with FOBT completion in the 6 months after randomization.
Lastly, although study eligibility was limited to patients who were not adherent to CRC screening, patients had to visit the International District Clinic during the intervention period to enroll in the trial. Study participants may therefore represent a group that is less recalcitrant to CRC screening. Given the scope of the study, we were unable to collect comparison data by chart audits from age-eligible patients who did not enroll in the trial.
Our findings indicate that an education program conducted by a health educator can significantly improve CRC screening among minority patients in a primary care setting. Chart audit data also suggest that the intervention may have increased CRC screening by sigmoidoscopy. In the intervention arm, 1 patient completed only sigmoidoscopy and 2 patients completed both FOBT and sigmoidoscopy.
With respect to colonoscopy, 1 patient from each trial arm completed only colonoscopy. Additionally, 3 control patients and 9 intervention patients completed both FOBT and colonoscopy. As a diagnostic procedure, colonoscopy is performed to evaluate positive FOBTs. Among the 12 patients, colonoscopies were all performed after FOBT completion, as expected for diagnostic rather than screening purposes.
The National Cancer Institute (NCI) recently published a compendium of articles addressing lessons learned and future directions for cancer screening research and practice. An article from the Applied Research Program's Office of the Director queried, “When can interventions be simplified? What is the minimal set of activities that will achieve the desired result? How do we develop interventions that are viable in clinical practice?”67 Among recommendations for future research, Meissner et al.68 emphasized the need to identify minimal intervention components essential to improve screening and follow-up that can be generalized to diverse populations and settings.
The large effect of our intervention suggests the remarkable impact of culturally appropriate health education among populations with limited health information. Results from other cancer control intervention trials with limited English-speaking populations also support this finding. In a randomized controlled study promoting Pap testing among North American Chinese women, women who received a mailed intervention with a motivational video and pamphlet significantly increased Pap testing compared with the controls.69 In a recent study by Zapka et al.,70 intervention patients who reported viewing a CRC screening video were significantly more likely to report sigmoidoscopy, although there was no significant difference in screening between patients who were mailed the video and those who were not mailed the video. In fact, to accommodate the busy environment of a primary care clinic, our intervention provided the flexibility for intervention patients to take the motivational video home to review. Of interest, our intervention process evaluation data show that 77% of intervention patients elected to take the video home to review. We postulate that members of minority communities with limited health information are more likely to review health information (e.g., videos and pamphlets) provided to them either by clinic staff or by mail. Thus, simpler and more viable interventions may contribute significantly to promoting CRC screening.
Future research should therefore determine the effective core component(s) of our primary-care based intervention that can increase CRC screening in both the general and minority populations. Through the identification of low intensity and clinically viable intervention component(s), dissemination of CRC screening to other populations and settings can be optimally enhanced.
We thank Dominica Lau for valuable contributions to this intervention study, as well as the many staff members at International Community Health Services who ensured the successful completion of the project. In particular, we thank Terry Tran, Roger Chiu, Mindy Au, Phan Lac, Ping Wong, and Marciano Rodriguez. Assistance with article preparation by Raymond Harris is much appreciated, and we also thank the 2 anonymous reviewers for helpful comments and suggestions
- 1American Cancer Society. Cancer Facts and Figures 2006. Atlanta: American Cancer Society, 2006.
- 15Hospital Authority of Hong Kong Special Administrative Region Key Statistics: Hong Kong Cancer Registry. Available at: http://www3.ha.org.hk/cancereg/ Accessed on March 9, 2006.
- 17Cancer incidence in five continents: volumes 1–8. Lyon, France: International Agency for Research on Cancer, 2005.
- 19Cancer patterns among migrant and native-born Japanese in Hawaii in relation to smoking, drinking, and dietary habits. In: GelboinHV, MacMahonB, MatsushimaT, eds. Genetic and environmental factors in experimental and human cancer. Tokyo: Japan Scientific Societies Press, 1980: 327–340., , .
- 21ParkinDM, WhelanSL, FerlayJ, RaymondL, YoungJ (eds.). Cancer incidence in five continents: volume 7. Lyon, France: International Agency for Research on Cancer, 1997.
- 24Cancer incidence among Asians in the greater Bay Area, 1990–2002. Fremont, CA: Northern California Cancer Center, 2005., , , et al.
- 27Variation in quality of men's health care by race/ethnicity and social class. Med Care. 2005; 43(3 Suppl ): I72–81., , , , , .
- 31Screening for colorectal cancer on the front line. Am J Gastroenterol. 2003; 98: 915–923., , , , .Direct Link:
- 49Predictors of colorectal cancer screening participation in the United States. Am J Gastroenterol. 2003; 98: 2082–2091., , .Direct Link:
- 54The practices and attitudes of primary care nurse practitioners and physician assistants with respect to colorectal cancer screening. Am J Gastroenterol. 2000; 95: 3259–3265., , , et al.Direct Link:
- 65Pathways to early cancer detection for Vietnamese women: Suc Khoe La Vang! (Health is gold!). Health Educ Q (Suppl). 1996; 23: 60–75., , , , , .