Enhancing Prevention Pathways Toward Tribal Colorectal Health

Despite the effectiveness of CRC screening tests for average risk adults, these tests are under-utilized by AI/ANs. A recent study of AI/ANs who utilize IHS facilities reported that only about 4.0% of average risk AI/ANs age 45-75, compared with nearly 65% of U.S. adults, were current with CRC screening guidelines. The overall objective is to test the efficacy of serially implemented interventions of graded intensity designed to enhance annual CRC screening using the FIT in AI communities. The aims are to:

Aim 1. Finalize and evaluate the efficacy of serially implemented interventions of graded intensity for increasing annual CRC screening uptake.

Aim 2. Determine promoters and barriers to enhancing annual CRC screening practices.

Aim 3. Conduct process evaluation studies on: cost-effectiveness of the interventions, implementation fidelity, and sustainability and scalability.

Protocol A. Aim 1-Qualitative Research (Intervention Finalization)

• Research Design: Qualitative design, using focus group methodology.
• Study Population: Recruited from the project's Community Advisory Council, from the Albuquerque Area Southwest Tribal Epidemiology Center's (AASTEC's) Executive Council, Community Health Representatives, and other tribal stakeholders.
• Recruitment Target: Up to 30-36 eligible (assuming attrition) persons.
• Recruitment Methods: Purposive sampling methods.
• Procedures: Conduct three focus groups, with each group consisting of 6 to 8 participants. Each session will meet for approximately two hours. Before starting the discussion, participants will provide informed consent and complete a survey collecting socio-demographic information. The purpose of these focus groups is to finalize the intervention's educational material, ensuring it is culturally appropriate and contains messages operationalizing various constructs of the theory guiding the study (Health behavior Framework; HBF).

B. Aim 1-Quantitative Research (Efficacy RCT)

• Research Design: Randomly assigned, three-group, pretest-posttest design.
• Research Setting: Six Pueblo Tribes in New Mexico.
• Randomization: Using simple random allocation procedures.
• Participants Identification and Sampling: Identify eligible participants from the IHS Resource and Patient Management System (RPMS) electronic health record system. We will invite all eligible persons (estimated at 1,700) identified through the RPMS to participate.
• Interventions: The interventions are graded in intensity: high intensity, medium intensity, and reference group (usual care).
• Procedures: After randomizing tribes to the three Groups, we will mail a recruitment letter, a consent form for completion of surveys, and a baseline survey. We will mail the first posttest survey 12 months after mailing the baseline survey, with the 2nd and 3rd posttest surveys mailed in successive 12-month intervals. Between the baseline and 1st posttest surveys, participants will receive the group-specific intervention. All eligible adults will be included in the analyses for the primary outcome (ie, completion of the FIT kit) since non-completion of a FIT is a valid response. All eligible participants in Groups 1 and 2 will receive a mailed FIT kit whether or not they return the baseline survey.
• Implementation of the Intervention: We will simultaneously implement the intervention, over three 12-month cycles. For Groups 1 and 2, we will mail the FIT kit, and the educational material about one month following the initial recruitment letter/baseline survey mailing. For Group 1, navigators will offer up to four education outreach sessions. Group 2 will receive the FIT and education material mailing. Group 3 will receive usual care (screening recommendation and a FIT kit at a clinic visit).
• Baseline and Posttest Surveys: These surveys will collect data on CRC-related knowledge, attitudes, perceived control, perceived susceptibility and severity, social support, and self-efficacy. The baseline survey will also collect socio-demographic data.
• Data Management and Analysis: We will summarize categorical and continuous variables using descriptive statistics. For all analyses, the overall type I error rate will be α=0.05. We will use appropriate methods such as Tukey's honest significant difference test or a Bonferroni correction for multiple comparisons.

For the primary outcome, annual completion of the FIT kit, the primary endpoint is the proportion of participants within each Group who completed and returned 0, 1, 2 or 3 FIT kits. All analyses will be intent-to-treat, since all eligible participants will be included in the analysis. We will compare the primary outcome across interventions via a likelihood ratio test with a multiple comparisons adjustment. To summarize intervention effects, we will report odds ratios and their 95% confidence intervals (CIs) for completing all 3 FIT kits as compared to returning each of 2, 1, or no FIT kits.

The secondary outcomes are temporal changes in the HBF constructs (eg, knowledge, attitudes, perceived risk) from baseline to 12-month cycles following the baseline survey. We will assess temporal changes in the HBF constructs with repeated measures linear regression modeling to understand the interventions' effects over time. Predictive value of the interventions and the HBF constructs on the yearly FIT uptake will be assessed by refitting the repeated measures logistic regression model described in the primary analysis for the dependent variable of return of completed FIT kit (yes vs. no) at each time point, but also including as a covariate score from all returned surveys measuring HBF constructs. We will report odds ratios (and 95% CIs) for completion by Groups.

• Sample Size Considerations: We powered our study to measure significant changes by Group in the annual completion of the FIT (primary outcome). With hypothesized proportions of participants with 0, 1, 2 or 3 returned FIT kits that reflect an average of 4%, 14% and 24% completing the FIT each year for the three Groups, respectively, the power of the study is ~99% with 566 participants per Group (total=1,698). We adjusted for multiple comparisons in our power calculation by using a Bonferroni correction.

Aim 2-Qualitative Research (Understanding Behavior Change Context)

• Research Design: Qualitative design.
• Study Population: Participants "largely adherent" (completed FIT 2 or 3 of 3 times) and "not adherent" (completed FIT 0 or 1 of 3 times) to guidelines; navigators; providers and medical directors of IHS facilities.
• Recruitment Target: Up to 120-144 participants (assuming attrition) who took part in the Aim 1 Efficacy RCT; with 60 participants "largely adherent" and 60 participants "not adherent" to the FIT guidelines. All the navigators (n=3) working on the project and medical directors (n=3) of the three participating IHS facilities. One health care provider from each of the three participating IHS facilities (n=3).
• Recruitment Methods: Purposive sampling methods.
• Procedures: Conduct 12 focus groups, with each group consisting of 6 to 8 participants. Each session will meet for approximately two hours. Before starting the discussion, participants will provide informed consent. The moderator will explore HBF factors most conducive to being "largely adherent" to screening guidelines. Among participants "not adherent" to CRC screening guidelines, we will inquire about barriers to completing a FIT and their potential solutions. We will also inquire about how we can sustain and scale-up the interventions for broader dissemination.

Conduct interviews with navigators, and health care providers and medical directors of the IHS facilities. Before starting the interview, we will obtain informed consent. Interviews with the navigators will focus on delivery of the intervention, barriers and promoters to providing navigation, participants' reasons for "largely adhering" or "not adhering" to CRC screening guidelines, and strategies to sustain and scale-up the interventions. Interviews with the health care providers and medical directors will focus on systems-level factors that may or may not facilitate delivery of CRC screening services, viability of using navigation for cancer screening in community settings, and strategies to sustain and scale-up the interventions. We will collect baseline socio-demographic information from the participants.

Aim 3-Process Evaluation Studies (Cost-Effectiveness Analysis, Implementation Fidelity, and Sustainability and Scalability)

• Procedure-Cost-Effectiveness Analysis: We will collect and use primary and secondary data to define parameters for the Markov models of cost-effectiveness. We will include data on CRC epidemiology in New Mexico, screening and treatment costs, and screening compliance rates for each cycle of the serially implemented interventions. We will collect data on all costs including navigation, mailing of FITs, educational material, and follow-up costs by each Group. We will assess direct medical costs through the IHS RPMS, which contains patient encounter-level data by outpatient ICD-9 diagnoses and procedure codes. We will extract data on sensitivity, specificity and efficacy of screening tests from the peer-reviewed literature. We will begin by constructing a decision tree for each of the Groups. We will compute the costs of each decision node (FIT completed/FIT not completed and received colonoscopy/did not receive colonoscopy) based on directly accrued medical and non-medical costs obtained from data sources identified above. The result of the cost-effectiveness analysis will be an incremental cost-effectiveness ratio (ICER) for the three Groups.
• Procedure-Implementation Fidelity: We will evaluate fidelity of implementation based on standard components, including: reach to potentially eligible participants; recruitment; resources and expertise needed for the research; the broader content (ie, social, physical, cultural, and structural factors); fidelity of delivery of the interventions; amount of the intervention actually delivered; amount of the intervention actually received; challenges encountered in implementing the research; potential contamination; and number and quality of partnerships.
• Procedure-Sustainability and Scalability: We will operationalize concepts for sustaining and scaling-up efficacious interventions, including: the nature of the interventions; the interventions' efficacy; the cost and cost-effectiveness of the interventions; partnerships; organizational capacity; contextual factors; promoters and barriers; and generalizability to other populations.

A. Aim 1--Qualitative Research (Intervention Finalization)

Inclusion Criterion:

• Adult (45-75 years) AIs.

Exclusion criteria

• None

Aim 1--Quantitative Research (Efficacy RCT)

Inclusion criteria

• AIs at average-risk for CRC,
• Not "up to date" with US Preventive Services Task Force (USPSTF) CRC screening guidelines (i.e., no guaiac fecal occult blood test [FOBT] or FIT in past year; no flexible sigmoidoscopy in past 5 years; no colonoscopy in past 10 years), and
• Residing in one of the six Pueblo Tribes participating in the research.

Exclusion criteria

meeting at least one of the following criteria:

• History of CRC, total colectomy, adenomatous polyps, or history of inflammatory bowel disease,
• Up-to-date with CRC screening,
• Severe comorbidity,
• No visit to IHS facilities in the past three years,
• Incarceration,
• Family history of CRC, or
• Hospice/terminal care status.

C. Aim 2--Qualitative Research (Understanding Behavior Change Context)

Inclusion criteria

• Participants who completed a FIT (Aim 1),
• Participants who did not complete a FIT (Aim 1),
• Navigators participating in the high intensity interventions (Aim 1),
• Health care providers and medical directors of the IHS facilities participating in the study.

Exclusion criteria

• None