CDC-funded intervention research aimed at promoting colorectal cancer screening in communities

Authors

  • Steven S. Coughlin PhD, MPH,

    Corresponding author
    1. Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Georgia
    • Epidemiology and Applied Research Branch, Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, 4770 Buford Highway, NE (K-55), Atlanta, GA 30341, U.S.A.
    Search for more papers by this author
    • Fax: (770) 488-4639

  • Mary E. Costanza MD,

    1. University of Massachusetts Medical School, Worcester, Massachusetts
    Search for more papers by this author
  • Maria E. Fernandez PhD,

    1. Center for Health Promotion and Prevention Research, School of Public Health, University of Texas Health Science Center, Houston, Texas
    Search for more papers by this author
  • Karen Glanz PhD, MPH,

    1. Department of Behavioral Sciences and Health Education, Rollins School of Public Health, Emory University, Atlanta, Georgia
    Search for more papers by this author
  • Judith W. Lee PhD,

    1. Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Georgia
    Search for more papers by this author
  • Selina A. Smith PhD, MDiv,

    1. Department of Community Health and Preventive Medicine, Morehouse School of Medicine, Atlanta, Georgia
    2. Prevention Research Center, Morehouse School of Medicine, Atlanta, Georgia
    Search for more papers by this author
  • Leonardo Stroud MD, MPH,

    1. Division of Cancer Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Georgia
    Search for more papers by this author
  • Irene Tessaro DPH,

    1. School of Nursing, West Virginia University, Morgantown, West Virginia
    Search for more papers by this author
  • John M. Westfall MD, MPH,

    1. Department of Family Medicine, University of Colorado at Denver and the Health Sciences Center, Aurora, Colorado
    Search for more papers by this author
  • Joel L. Weissfeld MD, MPH,

    1. Department of Epidemiology, University of Pittsburgh Cancer Institute and University of Pittsburgh, Pittsburgh, Pennsylvania
    Search for more papers by this author
  • Daniel S. Blumenthal MD, MPH

    1. Department of Community Health and Preventive Medicine, Morehouse School of Medicine, Atlanta, Georgia
    2. Prevention Research Center, Morehouse School of Medicine, Atlanta, Georgia
    Search for more papers by this author

  • The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.

Abstract

BACKGROUND.

Although strong scientific evidence has shown that screening for colorectal cancer saves lives, most U.S. adults who are at the recommended age are not being screened. Prior studies suggest that barriers to routine screening vary by race, ethnicity, socioeconomic status, urban/rural residence, health insurance status, and factors related to health care providers and the health care environment. Relatively few studies, however, have identified and tested intervention approaches to promote routine colorectal cancer screening among diverse populations.

METHODS.

The Division of Cancer Prevention and Control at CDC has funded ongoing projects to develop and test interventions to promote routine colorectal cancer screening among medically underserved populations in Appalachia, the Lower Rio Grande Valley in Texas, the High Plains region of Colorado, and other U.S. communities.

RESULTS.

This article provides an overview of colorectal cancer screening intervention studies currently funded by CDC that focus on a wide range of populations, including medically underserved persons who live in predominately rural areas, Hispanic and non-Hispanic persons, urban African Americans, persons with low health literacy, and persons enrolled in managed care organizations.

CONCLUSIONS.

These CDC-funded intervention research projects are likely to contribute importantly to evidence about what works to promote colorectal cancer screening in diverse U.S. communities. Cancer 2006. © 2006 American Cancer Society.

Colorectal cancer is a major cause of cancer morbidity and mortality in the United States, with an estimated 145,290 new cases in 2005.1 An estimated 56,290 persons will die from colorectal cancer in 2005. It is, in fact, the second leading cause of cancer death in the United States. Treatment of colorectal cancer costs about $6 billion a year.2 As shown by other chapters included in this supplement, colorectal cancer incidence and mortality rates are relatively high among some populations, for example, whites in Northern Appalachia and African Americans throughout the country. Screening rates are relatively low among Hispanics, rural residents, persons with low health literacy, and other population subgroups.3 These disparities are the result of a complex array of economic, social, and cultural factors, including disparities in preventive behaviors.4

PREVENTING DEATH FROM COLORECTAL CANCER THROUGH ROUTINE SCREENING

Strong scientific evidence shows that screening for colorectal cancer saves lives.4 The United States Preventive Services Task Force recommends initiating screening at the age of 50 years for men and women at average risk for colorectal cancer, a recommendation based on the observed rise in incidence of cancer in persons older than 50 years.5 Additionally, the American College of Gastroenterology has recently recommended that screening of African Americans begin at age 45 because of the higher incidence in this population.6 Research conducted by CDC investigators and others, however, has found that most persons in the target age groups are not being screened.7–11

A sizeable body of observational research indicates that barriers to routine cancer screening vary by race, ethnicity, cultural factors, socioeconomic status, urban or rural residence, family history of cancer, health insurance status, and factors related to providers and the health care environment.7, 8, 11, 12 Relatively few intervention studies, however, have identified and tested approaches to promote routine colorectal cancer screening among diverse populations.13 The need for additional intervention studies to identify effective approaches for promoting colorectal cancer screening is highlighted later. The needed studies include both community-based participatory research, which involves members of the target communities as empowered partners, and clinic-based research.

THE NEED FOR COMMUNITY-BASED, PARTICIPATORY RESEARCH AND CLINIC-BASED RESEARCH

There is evidence that sending client reminders and removing structural barriers (e.g., by improving access to screening via extended clinic hours) are effective interventions for colorectal cancer screening. However, the currently available evidence on the effectiveness of other specific client-oriented interventions (e.g. small media interventions, one-on-one education, and group education) and provider-oriented interventions is not sufficient to justify a Guide to Community Preventive Services recommendation.13 The Community Guide, developed by the Task Force on Community Preventive Services, evaluates community, population, and health care strategies to address a variety of public health and health promotion topics.13 Relatively few community intervention studies have examined methods to increase colorectal cancer screening, and those studies have focused primarily on client reminders. Effective interventions to promote routine screening are needed for all age-eligible persons, particularly those in medically underserved populations. Patients with low literacy may have limited exposure to literature recommending screening and early detection. Low health literacy, which is common among persons with limited education and low incomes, may also be an important factor in patients' decision making about colorectal cancer screening.14

ONGOING INTERVENTION RESEARCH FUNDED BY CDC

The Division of Cancer Prevention and Control at CDC has funded ongoing projects to develop and test interventions to promote routine colorectal cancer screening among a variety of patient populations, including the medically underserved in Appalachia, the Lower Rio Grande Valley in Texas, the High Plains region of Colorado, the Atlanta Metropolitan Area, and other U.S. communities (Table 1). These cooperative agreements or contracts, which are funded through grant or contract mechanisms and approved by institutional review boards, focus on persons who live in predominately rural and frontier areas, Hispanic and non-Hispanic persons, persons with low health literacy, and persons enrolled in managed care organizations. Their aim is to eliminate the disparities in colorectal cancer incidence among these populations. These projects illustrate the beneficial uses of local and central cancer registry data. The purpose of this article is to summarize these 8 ongoing intervention studies, draw some conclusions about their implications, and make suggestions for future directions.

Table 1. CDC-Funded Intervention Research Projects Aimed at Promoting Colorectal Cancer (CRC) Screening in Communities
ProjectPrincipal investigatorPeriodLocation, settingTarget populationStudy designIntervention approachesBehavioral frameworkMain outcome
  • *

    Text in parentheses indicate the institution.

  • Examples of small media include videos, flipcharts, posters, and pamphlets.

  • Unincorporated communities along the Texas–Mexico border.

Colorectal Cancer Control in Appalachian ChurchesIrene Tessaro, DPH (West Virginia University)*2001–2006Ohio River Valley of western West VirginiaAdult members of 16 churches from a 7-county area2 × 2 experimental design with cluster randomization by churchNatural helpers or parish nurses provide one-on-one and group educationSocial Support Theory, Health Belief Model, Social Cognitive Theory, Transtheoretical Model, community-based participatory researchIncreased CRC screening, primary prevention (increased physical activity, increased fruit and vegetable, and decreased dietary fat intake)
Colorectal Cancer Screening Trial among African AmericansDaniel S. Blumenthal, MD, MPH (Morehouse School of Medicine)2001–2006Atlanta, Georgia, metropolitan areaLow-income African Americans aged 50 years and older from 5-county areaRandomized controlled trialOne-on-one education, group education, financial incentivesSocial marketing, community coalition building, Health Belief Model, Social Cognitive TheoryIncreased awareness of modifiable risk factors and CRC screening
Empowering People in Central Massachusetts to Increase CRC ScreeningMary Costanza, MD (University of Massachusetts)2001–2006Primary care practices in central MassachusettsRandom sample of (Caucasian) patients aged 50–75 years seen at 1 of 37 physician practicesRandomized controlled trial, with randomization by practiceSmall media, one-on-one educationPrecaution Adoption Process ModelIncreased CRC screening, economic costs
Increasing Utilization of CRC Screening among Hispanics in the Lower Rio Grande ValleyMaria E. Fernandez, PhD (University of Texas at Houston)2004–2008Lower Rio Grande Valley, TexasLow-income Hispanics aged 50 years and older residing in colonias along the Lower Rio Grande ValleyRandomized controlled trial, with colonias as the unit of randomizationSmall media and tailored interactive multimedia delivered by promotorasIntervention mapping, community-based participatory researchIncreased CRC screening and informed decision-making about screening
Promoting CRC Screening in Rural ColoradoJohn Westfall, MD (University of Colorado)2004–2008High Plains Research Network (HPRN), rural and frontier communities in eastern ColoradoAdults aged 50 years and older (primarily Caucasian and Hispanic)Nonrandomized controlled trial, with 9 intervention counties in the HPRN and 5 control countiesSmall group education, small media, mass mediaCommunity engagement model of community-based participatory researchIncreased CRC screening
Communicating Health Options through Information and Cancer Education (CHOICE)Karen Glanz, PhD, MPH (Emory University)2004–2008Health practices in Georgia and Florida belonging to a large health plan30 large group practices and (primarily Caucasian and African American) patients aged 52–75 years from each practiceRandomized controlled trial with randomization by practiceVideotape-based decision aid for patients and provider-directed academic detailingSocial Cognitive Theory and Transtheoretical ModelIncreased CRC screening
Physicians Encouraging Colorectal Screening (PECS)Joel L. Weissfeld, MD, MPH, (University of Pittsburgh Cancer Institute)2001–2005Southwest Pennsylvania, primary medical care practice50- to 79-year-old patients without personal history of CRC or polyp, flexible sigmoidoscopy within 5 years, or colonoscopy within 10 years2 × 2 randomized clinical trial with physician group clustersTailored physician recommendation and coordinated physician office management systemsPut Prevention into Practice, Motivational InterviewingLower endoscopic colorectal cancer screening
Evaluation of an Intervention to Increase CRC Screening in Primary Care ClinicsJudith Lee, PhD (Centers for Disease Control and Prevention)2002–2008Henry Ford Health System in Michigan and Lovelace Health Systems in New MexicoAdult patients seen at 27 primary care clinics belonging to two managed care organizationsRandomized controlled trial with randomization by clinicSmall media, provider education, and office system changes to increase screening among eligible patientsInformed by the Integrated Behavioral ModelIncreased CRC screening, clinician–patient discussions about CRC screening, attitudes, beliefs, and social influence surrounding CRC screening

Study 1: Colorectal Cancer Control in Appalachian Churches

The primary aim of this church-based randomized controlled intervention is to evaluate the independent and combined effects of 2 strategies for education and support that use parish nurses or natural helpers (lay health advisors). These persons are involved in the primary and secondary prevention of colorectal cancer among members of a church population. Churches were eligible to participate if they had at least 100 active members on their roster and had a parish nurse or health ministry, or if they had carried out some health activities not related to colorectal cancer screening. Eligible churches were separated into 2 clusters: those with a parish nurse (n = 25), and those with a health ministry or a history of conducting health activities (n = 49), and randomized for recruitment. Eight churches from each cluster were successfully recruited into the study. Before any data were gathered, an Advisory Group was set up in each church to provide participatory guidance on data collection, educational training sessions, and implementation of church interventions. Self-administered baseline surveys were completed in the 16 churches by 1239 men and women, 839 of whom were aged 50 years or older. Surveys will be readministered at 2-year follow-up.

After baseline surveys were completed, the churches were randomly assigned to 1 of 4 conditions: parish nursing, natural helpers, parish nursing and natural helpers, or minimal intervention. Before the interventions were implemented, focus groups were conducted in the 16 churches to obtain a better understanding of church members' knowledge and perceptions about colorectal cancer prevention and early detection, and to identify barriers and facilitators for behavior change. Feedback on baseline survey and focus group results was given to each church. With the use of health education materials developed specifically for this Appalachian church population, educational training sessions on the prevention and early detection of colorectal cancer and action planning for behavioral change were conducted for the parish nurses and natural helpers at each intervention church. Natural helpers or parish nurses disseminated information about colorectal cancer to church members, provided support for screening and risk reduction, and organized group activities for primary study outcomes.

Study 2: Colorectal Cancer Screening Trial among African Americans in Atlanta, Georgia

The goal of the Colorectal Cancer Screening Trial is to evaluate the effectiveness of 3 interventions to increase screening rates among low-income African Americans from 5 counties in the Atlanta Metropolitan area. The study, which uses both medical care system- and community-based interventions, is designed to increase participants' awareness of modifiable risk factors for colorectal cancer and the early detection benefits of screening. It is also intended to encourage persons aged 50 years and older to make screening a priority.

The phased 5-year trial included 500 African American men and women, aged 50 years and older, recruited from churches, senior centers, federally designated community health centers, public health units, and other sites throughout the Atlanta Metropolitan Area. The participants were randomized either into a control cohort or into 1 of 3 intervention cohorts—one-on-one counseling, small group sessions, or financial incentives. Phase I (the first 12 months of the project) included the recruitment and randomization of participants to either control or intervention cohorts. During this phase, a culturally appropriate video was used to introduce participants to “Down Home Healthy Living,” an educational program aimed at decreasing colorectal cancer risk through lifestyle modification and awareness of early detection (http://www.nblic.org/down_home.htm).

Initial data collection included baseline assessments and medical record abstraction to determine the participants' eligibility, including their colorectal cancer screening history. Baseline assessments included knowledge, attitudes, beliefs and practices, and various psychosocial measures. Reliability, validity, and cognitive testing were completed for any newly developed evaluation instrument. Repeat outcome measurements and screening verification information were collected from participants after the intervention during Phase II (years 2–4).

Study 3: Empowering People in Central Massachusetts to Increase Colon Cancer Screening

This project in Worcester, Massachusetts, is currently in Year 5 of 5. A major impetus for the study was to learn whether an inexpensive, tailored intervention could be successfully delivered to a large number of people and whether it would be effective in promoting cancer screening. The specific aims are 1) to determine, among people aged 50–75 years, the effectiveness of an intervention that combined a print component and tailored computer-assisted telephone counseling; 2) to determine the effectiveness of the print component alone; 3) to determine whether mediating mechanisms suggested by the Precaution Adoption Process Model are associated with the intervention's effectiveness and to explore the patterns of colorectal cancer screening in both the short term (90 days) and the long term (1 year); and 4) to collect cost data.

The setting of the study is a community practice division of a large medical center serving Worcester County in central Massachusetts. All primary care providers (PCPs) in the division were invited to participate. Of the 51 active office-based PCPs, 38 (75%) agreed to allow their patients to participate. One provider's practice was used solely for pretesting. The remaining 37 physicians participating in the study were surveyed about their colorectal cancer screening practices.

A random sample of 3404 patients aged 50–75 years, who had seen 1 of the 37 providers within the past 2 years, was selected for the study. PCPs were asked to review their list of selected patients and exclude any with the following: limited life span, history of colon polyps, colon cancer or bowel disease, not under their care, deceased, cognitive impairment, non-English speaking, or other. Four hundred and forty patients were excluded, leaving 2934 eligible for study. Patients were stratified by their type of health insurance. About 10% received either free care or Medicaid. Patients were then randomized into intervention or control groups. The study used mailed baseline and final surveys to assess participants' sociodemographic and health system characteristics, colorectal cancer screening history and intention, attitudes, knowledge, and beliefs pertaining to screening. The baseline survey was mailed to 2934 eligible patients, of which 2027 surveys were completed and returned (a 69% response rate).

Patients were excluded from the study if any of the earlier-mentioned exclusion criteria were noted on the returned survey. After survey exclusions, there were 1180 patients in the intervention arm who were still eligible. The intervention arm consisted of a mailed colon cancer screening brochure that addressed screening knowledge and risks of getting colon cancer. The booklet described the various screening options, the process of getting or performing them, their relative risks and benefits. After 90 days, the intervention group received a tailored counseling call. This 90-day interval will permit an eventual evaluation of the brochure alone as an intervention. Computer-assisted telephone counseling interviews were used as the second part of the intervention activities. The counseling protocol directed the telephone counselors through a series of questions to classify patients according to the Precaution Adoption Process Model, and then helped them to counsel patients on the basis of the individual's intention to be screened and on his or her specific screening barriers. A total of 1025 patients were reached, and of these 725 (71%) were actually eligible to receive counseling. Reasons for ineligibility included respondent's report of a history of colonic polyps; primary care physician's nonparticipation in the study; patients being deceased, having cognitive impairment, being non-English speaking, having a history of colorectal cancer, and other.

Study 4: Increasing Utilization of Colorectal Cancer Screening among Hispanics in the Lower Rio Grande Valley

The goals of this recently initiated study are to develop and assess the effectiveness of a community-based intervention to increase colorectal cancer screening among low-income Hispanics in the Lower Rio Grande Valley of Texas, an area along the U.S.–Mexico border. The specific aims are 1) to identify personal and external factors influencing colorectal cancer screening and informed decision-making about screening; 2) to develop a detailed intervention plan using intervention mapping; 3) to develop 2lay-health-worker (promotora)–delivered community interventions (namely, a small media intervention using video, flipchart, and pamphlets, and a tailored interactive multimedia intervention); 4) to implement and evaluate the effectiveness of the 2 interventions for increasing screening and decision-making about screening among Hispanic men and women.

The project team is completing Phase I of the study and has begun phase II. Phase I uses qualitative inquiry and secondary analysis of data to identify personal and external factors influencing colorectal cancer screening and decision-making about it. Phase II includes the development and pilot testing of both interventions using Intervention Mapping, a method for planning theory and evidence-based health promotion interventions.15 Following completion of the intervention-development and pilot-test phase, a randomized controlled trial of the interventions will be conducted, with randomization at the level of colonias (unincorporated communities along the Texas–Mexico border). Researchers will compare the effectiveness of the interventions on increasing screening. They will also assess changes in intermediate impact variables such as knowledge about colorectal cancer and colorectal cancer screening, colorectal cancer screening self-efficacy, perceived cancer survivability, decisional balance, and other factors that seem to influence screening decisions. About 733 women and men (about 50% women) aged 50 years and older will be included as participants in this study.

As mentioned earlier, decisions regarding the development of interventions will be informed by results from focus groups conducted in January 2004 in Brownsville, El Paso, and Laredo, Texas, with low-income Hispanic men and women,16 and by analysis of data (resulting from a survey of 600 Hispanic men and women conducted in the same 3 cities) collected via one-on-one interviews. Researchers will also conduct focus groups and interviews with promotoras and providers to further inform development decisions. Both the focus groups and survey addressed demographic, psychosocial, and cultural factors potentially related to colorectal cancer screening, as well as general issues such as access to health care services, perceptions about the importance of preventive health care, and factors surrounding health care decision making. Focus group findings indicate that both men and women reported a heavy reliance on home remedies, herbal remedies, and prescription drugs bought across the border, in part, due to financial barriers and lack of insurance. The participants' knowledge, attitudes, and beliefs related to cancer reflected a mixture of misconceptions and accurate information, and participants expressed varying beliefs about the survivability of cancer. Most participants in these focus groups knew very little about colorectal cancer or about colorectal cancer screening.16 Individual-level barriers to colorectal cancer screening suggested by participants included embarrassment, machismo, lack of knowledge and information, procrastination, fear of questioning physicians, fear of the screening procedures, fear of receiving a cancer diagnosis, and lack of health insurance or financial resources.

Study 5: Promoting Colorectal Cancer Screening in Rural Colorado

A community-based participatory research process was used to plan and initiate a community-wide intervention to increase colorectal cancer screening in rural Colorado (High Plains area). The research team is working with a Community Advisory Council (CAC) and an associated Joint Planning Committee (JPC). The CAC includes farmers, ranchers, school teachers, a college student, and several other community members. The JPC includes all the CAC members and several physicians, hospital administrators, and public health officials. On the basis of prior research, the Committee is designing appropriate educational messages and delivery approaches likely to result in behavioral change (obtaining colorectal cancer screening). Intervention approaches are likely to be multicomponent, including small group education, small media (for example, a “farm auction” flyer that actually has information on preventing colon cancer), and mass media (newspaper stories and advertisements), and an incentive for discussing prevention with their physician (travel mug, free beverage). A pilot study is being conducted in 2 High Plains communities this year. A similar rural area in southeast Colorado will serve as control communities. A random digit dialing telephone survey will be used to determine colorectal cancer screening rates, beliefs, and attitudes in intervention and control counties. A common practice-based research network data collection approach will be used to determine colorectal cancer screening rates among individuals seeking primary health care in the High Plains Research Network. Qualitative methods will be used to test the content and potential impact of the pilot messages and strategies for increasing colorectal cancer screening. Men and women older than 50 years will be part of the target population for this intervention study. All individuals regardless of race or ethnicity will be included. About 17% of residents in the High Plains area are Hispanic. English–Spanish interpretation and translation will be included as needed.

Study 6: Increasing Colorectal Cancer Screening in Health Plan Members in Georgia and Florida

Because key predictors of colorectal cancer screening utilization include having a physician's recommendation and health insurance, and given the paucity of evidence-based strategies for increasing screening uptake, it is important to conduct effectiveness studies in health plan member populations. A research team is conducting participatory research with a major health insurer to test the transferability of community-based behavioral interventions to increase provision of colorectal cancer screening.

The study, known as CHOICE (Communicating Health Options through Information and Cancer Education), is a cluster-randomized trial being conducted in health practices in Georgia and Florida to test the effectiveness of a videotape-based decision aid and academic detailing for increasing adherence to colorectal cancer screening guidelines. Thirty large group practices will be recruited and randomized to receive usual care (routine reminders) or a combined videotape-based decision aid and academic detailing intervention. “Academic detailing” combines physician education with consultation on organizing health care practices to make it easier for patients to obtain colorectal cancer screening. This is a patient- and provider-directed intervention.

In each practice, 30 patients, aged between 52 and 75, previously unscreened for colorectal cancer, will be enrolled into the study. The intervention will continue for up to 2 years for still-unscreened participants. The main outcome will be provision (or receipt) of screening.

The research team has already begun a participatory research process with key leaders in the Quality Improvement and the Medical Economics units at Aetna, one of the nation's largest health insurers. Unique features of the proposed study include its potential to establish systems to increase screening uptake that will help fulfill HEDIS requirements. This project will improve our understanding of how screening-promotion interventions work in both white and black populations, and it will forge collaborative relationships between public health and health care researchers and the affected communities of health plans and health care providers.

Study 7: Physicians Encouraging Endoscopic Screening Study in Pittsburgh

The Physicians Encouraging Colorectal Screening (PECS) study is a collaboration that involves the University of Pittsburgh Cancer Institute and a physician research network organized and supported by the University of Pittsburgh School of Medicine (Pittsburgh Research in Office Network, PRONET). In a primary medical care setting, PECS is using a 2 × 2 factorial randomized clinical trial to measure the effects of a tailored physician recommendation and coordinated physician office management systems on compliance with colorectal cancer screening recommendations. The study population consists of 50- to 79-year-old patients seen at 1 of 10 PRONET group practices. To solicit study participants, PECS mailed an explanatory letter, a baseline questionnaire, and a written informed consent document to age-eligible patients. Persons eligible for randomization and intervention included patients who gave consent and did not report a personal history of colorectal cancer or polyp, flexible sigmoidoscopy within 5 years, or colonoscopy within 10 years. One dimension of the 2 × 2 factorial design served to randomize primary care physician group practices (matched according to size and clientele) to a high versus low level intervention (intensive versus basic help for physicians' offices with patient management). The high-level physician group practice intervention used Put Prevention into Practice and motivational interviewing paradigms to stimulate referrals for lower endoscopic colorectal cancer screening and to help nonadherent patients overcome perceived barriers to screening, respectively. The second dimension of the 2 × 2 factorial design served to randomize eligible patients within each group practice to high versus low level intervention (personalized versus generic recommendations). That is, each study participant received a written communication, prepared by PECS, that recommended colorectal cancer screening with either flexible sigmoidoscopy or colonoscopy and asked the participant to contact his/her personal physician to make the necessary arrangements. The personalized recommendation letter also addressed subject-specific risk factors and negative beliefs or anxieties self-reported on the PECS baseline questionnaire. PECS used multiple overlapping approaches, including direct subject contact and review of electronic and paper medical records, to ascertain screening end points. The PECS-defined primary clinical trial endpoint is the medical-record–verified occurrence of flexible sigmoidoscopy or colonoscopy between the date of completion of the baseline questionnaire and the 1-year anniversary of the mailing date of the PECS screening recommendation letter.

Study 8: Evaluation of an Intervention to Increase Colorectal Cancer Screening in Primary Care Clinics in Michigan and New Mexico

This CDC research study is a 4-arm randomized controlled trial to implement and evaluate the relative effects of 1) a patient-focused intervention, 2) a clinic-focused intervention, 3) a combined patient- and clinic-focused intervention, and 4) usual care on the provision of colorectal cancer screening in primary care clinics. The study will also examine the relative effects of the intervention conditions on secondary behavioral outcomes (e.g., clinician–patient discussions about screening) and on intermediate outcome measures of attitudes, beliefs, opinions, and social influence surrounding screening among patients, clinicians, and clinical staff.

Primary and intermediate outcomes will be measured at baseline (before intervention) and during the 1-year period after the intervention is begun. Both the patient- and clinician-level primary outcome measures of colorectal cancer screening will be obtained through electronic medical records and through patient and clinician surveys. Medical record data will be supplemented with self-report survey data. Pre- and postintervention surveys of patients and clinicians/clinical staff will be conducted to assess both the primary (colorectal cancer screening) and intermediate (e.g., attitudes, beliefs) outcomes.

The study will be conducted within the patient care environment (primary care clinics) of 2 managed care organizations: Henry Ford Health System in Detroit, Michigan, and Lovelace Health Systems in Albuquerque, New Mexico. The unit of randomization to a study arm will be primary care clinics. A total of 27 primary clinics across both the managed care organizations will be randomized to the 4 study arms.

The patient-focused intervention will involve sending an education packet on colorectal cancer screening (namely, CDC-developed educational text tailored for use within the managed care organizations), accompanied by a letter from the patient's physician, to patients meeting study criteria. In addition, the CDC has developed fecal occult blood test (FOBT) kit instructions that are easier to read (i.e., clearer and printed in a larger font) than those that manufacturers include with their FOBT kits.

The clinic-based intervention consists of training sessions for clinicians and clinical staff on colorectal cancer screening. The provider training sessions will include statistics and information and will be heavily interactive and skills-based. The provider training session will also focus on the provision of tools for clinical staff– and clinician–patient interactions about screening and for office system changes or improvements to identify eligible patients and to track screening.

Among patients, the primary outcome is colorectal cancer screening via any of the 4 modalities (FOBT, colonoscopy, flexible sigmoidoscopy, and double contrast barium enemas). Among clinicians, the primary outcome is rate of providing screening to age-appropriate patients. These primary outcome measures will be obtained through self-administered surveys and through managed care organization's electronic medical records. Additionally, to measure the effect of the intervention on self-reported colorectal cancer screening beliefs and behavior (the secondary behavioral and intermediate outcomes listed earlier), pre- and postintervention surveys will be administered to patients, clinicians, and clinical staff.

DISCUSSION

CDC is supporting ongoing projects that will help identify effective intervention approaches for promoting colorectal cancer screening in diverse populations. As detailed in other articles in this monograph, colorectal cancer incidence and mortality rates in the United States vary by race, ethnicity, socioeconomic status, residence, and other factors. For example, racial disparities have been observed in mortality rates and in stage at diagnosis. Effective interventions are needed to eliminate such disparities by promoting screening (to eliminate polyps and detect cancer early) and reducing risk factors such as physical inactivity and poor diet (primary prevention).

As part of the projects summarized here, efforts are being made by individual research teams to maximize the quality and completeness of reported results, e.g., by use of the Transparent Reporting of Evaluations of Interventions with Nonrandomized Designs (TREND) statement.17 For intervention approaches found to be effective in promoting colorectal cancer screening in communities and in health care settings, future steps may include dissemination and diffusion research for the interventions. Research-tested intervention materials may be placed on the Cancer PLANET website18 to facilitate their dissemination.

Results from these ongoing projects will help to fill in gaps in the Guide to Community Preventive Services13 and other evidence-based systematic reviews about what works to promote routine colorectal cancer screening and thus address health disparities. At present, the Community Guide offers only 1 strong recommendation to increase screening—reducing structural barriers (e.g., expanding clinic hours, providing child care, etc.)—and 1 “sufficient evidence” recommendation—client reminders. There is “insufficient evidence” for other intervention approaches for promoting screening. Thus, these CDC-funded intervention research projects will contribute importantly to evidence about what works to promote colorectal cancer screening in diverse U.S. communities.

Acknowledgements

We are grateful to coinvestigators, project directors and staff, and members of community advisory committees for their important contributions to these projects: Colorectal Cancer Control in Appalachian Churches (Principal Investigator Irene Tessaro, Carol Mangone, Sam Zizzi, David Goodrich, Bill Carlton, Sheila Rye, and Kim McCourt); Colorectal Cancer Screening Intervention Trial among African-Americans in Atlanta, Georgia (Principal Investigators Daniel Blumenthal and Selena Smith, Lee Caplan, Charlyne Taylor, Lisa Hinton, Latoshia Wells, Joshua Shifrin, Ernest Alema-Mensah, Falah Shamsa, and members of Community Coalition Board); Empowering People in Central Massachusetts to Increase Colon Cancer Screening (Principal Investigator Mary Costanza, Roger Luckmann, Milagros Rosal, Anne Stoddard, Lynn Clemow, Jill Arunin, Mary Jo White, and Jennifer R. Stark); Increasing Utilization of Colorectal Cancer Screening among Hispanics in the Lower Rio Grande Valley (Principal Investigator Maria Fernandez, Theresa Byrd, Evelyn Chan, Wenyaw Chan, Pamela Diamond, David Lairson, Maureen Sanderson, Sally Vernon, Isabel Torres, Janet Williams, Rosario Wippold, Angelia Hall, Marian Harris, and Diana Isachiavici); Promoting Colorectal Cancer Screening in Rural Colorado (Principal Investigator John Westfall, Becky Van Vorst, Elizabeth Staton, Linda Zittleman, Rodrigo Araya-Guerra, Deborah Main, Walter Young, Grace Linn, and members of Joint Planning Committee [Saeid Ahmadpour, Arlene Harms, Denise Hase, James Miller, Kindra Mulch, Fred Grover, Tim Byers, Kaia Gallagher, Shirley Cowart, Garry Haynes, Connie Haynes, Mike Hernandez, Ned Norman, Maret Felzien, Mary Petra Rodriquez, Cameron Walsh, Steve Winkelman, and Kathy Winkelman]); Increasing Colorectal Cancer Screening in Health Plan Members in Georgia and Florida (Principal Investigator Karen Glanz, Tracy Scott, Lauren Taglialatela, Michael Pignone, Barbara K. Rimer, and Jennifer Griffith); Physicians Encouraging Endoscopic Screening (Principal Investigator Joel Weissfeld, Robert Schoen, Jeanette Trauth, Bruce Ling, Francis Solano, Deborah Simak, and Theresa Eury); and Evaluation of an Intervention to Increase Colorectal Cancer Screening in Primary Care Clinics in Michigan and New Mexico (Principal Investigator Judith Lee, Mary White, Steven Leadbetter, Daniel Mantano, Danuta Kasprzyk, Alvah Bittner, Linda Carlin, Margaret J. Gunter, Shelley Carter, Jennifer Elston Lafata, and Christina Moon)

Ancillary