Increasing African Immigrant Women's Participation in Breast Cancer Screening

The goal of Aim 1 is to collect formative data via a mixed methods approach and use those results to culturally adapt the Witness Project for African immigrant populations. Cultural adaptation will follow the stage model outlined by Barrera et al.48 The model includes five stages: 1) information gathering to determine which intervention components to modify; 2) preliminary adaptation design that integrates information from stage 1 to inform preliminary modification of the original intervention; 3) preliminary adaptation tests to pilot the intervention for feasibility and acceptability, 4) adaptation refinement with decisions informed a leadership team/advisory board; and 5) cultural adaptation trial to test the effectiveness of the intervention in changing health outcomes. As this is the first step in this line of research, the proposed study will only follow stages 1-4. The Witness Project has three main components that will be culturally adapted: 1) educational presentation addressing myths, barriers and values related to breast cancer and mammography, 2) culturally matched, peer led narrative about breast cancer experience, 3) experiential education about breast anatomy and self-examination. The quantitative surveys will be conducted first. The results will inform the refinement of the focus group guide and in-depth interview questions. The "point of interface" (i.e., merging quantitative and qualitative data) will be associations between mammography screening and HBM constructs (e.g., perceived risk, self-efficacy) and any socio-demographic characteristics (i.e., preferred language, health literacy level). The quantitative results will inform the refinement of the qualitative questions. For example, if an association is found between self-efficacy and screening, the study team will ask in-depth questions about self-efficacy and suggestions for how it could be improved in this population.

Participant Recruitment for Focus Groups, Interviews and Surveys. There is a tremendous diversity of African nations and languages; focusing research on individual African nations and/or native languages is not feasible within the proposed study. Given that French and English are the most prevalent non-native languages spoken by African immigrants in New York City, the study team has chosen to focus on targeting individuals who speak these languages as a first step in this line of research. All participants will be recruited from African immigrant communities in NYC. Based on the prior experience of the study team's community partners, the majority of the populations that they serve are from Senegal, Mali, Ivory Coast, and Burkina Faso. In anticipation of wide variations in cultural values, barriers and benefits, the study team will thus use a maximum variation sampling approach to allow discovery of central themes, core elements and/or shared dimensions that cut across a diverse sample of participants. The study team has a well-established collaborative relationship with gatekeepers (see Letters of Support) in the African immigrant community and have successfully recruited African-born persons into prior studies49 and therefore anticipate successful recruitment for this proposed research. Participants will be recruited for in-depth interviews, focus groups, and surveys through IRB approved flyers at community sites. All participants will receive $30 for their participation. HIPAA authorization and informed consent will be obtained prior to participation. Quantitative Surveys: The study team will conduct surveys (N=50) to assess predictors of breast cancer screening among African immigrant women. The survey will be conducted by a bilingual RA an take about 15-30 minutes to complete and participants will receive $30. Measures: Sociodemographic data will be collected for all participants and will include age, gender, education level, primary language spoken, marital status, health insurance status and income (see Appendix). The survey will include questions about breast cancer knowledge, self-efficacy, fear, mistrust, perceived risk, benefits/barriers to screening, fatalism, screening intention, acculturation, and health literacy(see Appendix; Table of Measures). Focus Groups: The study team will conduct three focus groups (n=24) with African immigrant gatekeepers and stakeholders, including African immigrant women. The study team will ask the community partners to assist us with identifying potential participants for the focus groups. The goal of the focus groups is to gather rich, qualitative data from African immigrant stakeholders and gatekeepers about the breast cancer screening needs of women within their communities. The study team will deliver informational presentations about the focus groups and the proposed study at events and meetings held by the community partners. The study team will conduct one focus group with gatekeepers and stakeholders (in English) and the other two focus groups will be conducted with African immigrant women (in French and English). The focus groups will take 75-90 minutes to complete and will be conducted by a bilingual RA. In-depth Interviews: The study team will conduct in-depth interviews (n=20; see Appendix) to identify breast cancer screening barriers and facilitators (e.g., age, gender, country of origin) among eligible African-born immigrant women. An RA will approach potential participants and explain the purpose of the study, verify eligibility and obtain informed consent. Participants will be asked to discuss their beliefs about barriers and facilitators patients may face as they consider breast cancer screening. The duration of the in-depth interview and focus group will be approximately one hour including the interview and completion of a demographic questionnaire. Quantitative Data Analysis: As the assessment will be administered as an interview, The study team does not anticipate much missing data. As needed, missing data will be imputed and then be used to estimate the parameter values for each of scales. The study team estimates that the sample size of 64 participants will generate a +/- 11% margin of error at a 90% confidence level. The primary outcomes are screening status (up-to-date with for mammography screening) and HBM constructs. As screening status is dichotomous, a logistic regression analysis will be conducted using socio-demographics and HBM constructs (e.g., self-efficacy, perceived risk, perceived benefits) as explanatory variables. Since Aim 1 is designed to provide information to inform the adaptation of the subsequent intervention and education, the study team shall adopt a less conservative criterion for statistical significance using a p-value equal to or less than 0.15.

Qualitative Data Analysis. The PEN-3 model25 will guide the interpretation and analysis of the focus groups and in-depth interviews. Using this model as a framework, the analysis approach will identify the unique barriers and facilitators related to African-born immigrants' health care engagement. Transcripts will be reviewed using the N'Vivo qualitative data management software. 50 Analysis will use tri-phasic coding - including open, axial and selective coding. In the open coding phase, transcripts will be examined for salient categories of information supported by the text (e.g., barriers and facilitators to uptake of cancer screening in African immigrants). Themes that are extensively discussed by participants will be coded. Using a constant comparative approach, categories will be saturated until new information is no longer added. The research team will meet prior to conducting analysis to define coding labels and definitions. Potential labels and definitions will be drawn from identified barriers and facilitators for other preventive cancer screening exams (e.g., pap test, colonoscopy). All transcripts (including those translated from French to English) will be independently coded by PI Sly and Co-I Jandorf using the same framework to ensure consistency. They will meet multiple times to discuss the coding strategy for the first focus group and first half of the interviews and discuss and resolve any coding discrepancies prior to coding of the remaining focus groups and interviews, and again to resolve any remaining discrepancies by consensus. When there is disagreement on coding labels, the research team will meet to discuss reasons for disagreement and modify coding labels appropriately. Analyses will also examine subgroup differences (i.e., country of origin), age, and language. The study team's proposed sample size of 24 focus group participants and 20 in-depth interview participants will be stratified by language and the study team anticipate achieving saturation with these sample sizes. If saturation is not achieved with the 44 participants, the study team will recruit more participants with collaborators and community partners.

The educational aspect of the program will be modeled after the study team's current community based programs, as noted above, using narrative formats through stories from role models who have undergone breast cancer screenings. The programs will disseminate relevant, up-to-date information for breast cancer to eligible African-born immigrant women.

Recruitment and Iterative Program Development: After the completion of the interviews and focus groups in Aim 1, a 3×3 table of themes categorized using the PEN-3 model 25 will be produced for each of the three components of the Witness Project: 1) addressing myths, barriers and values, 2) culturally relevant breast cancer narrative, 3) experiential breast education. Themes will be categorized in the following domains: cultural empowerment and relationships and expectations. These themes will inform how the intervention components for the educational narrative program should be adapted for an African immigrant audience. The themes will first be categorized according to barriers and facilitators to screening and then the team will decide which themes are appropriate. For example, barriers and facilitators from the cultural empowerment domain may include concerns about deportation, modesty, religiosity and healthcare screenings and will be addressed in the Witness Project component 1. Another example might be that barriers and facilitators from the relationships domain could include concerns about the role of patriarchy in women's healthcare decision making - such as having a mammogram. This could potentially be addressed in the culturally relevant narrative (component 2). The cultural facilitators to screening identified in the interviews will be incorporated into the program by highlighting them as reasons to participate in screening. Cultural barriers identified will also be addressed in the culturally relevant narrative by providing examples and suggestions for coping with these perceived risks. The study team will invite participants (N=20) to give feedback on the first iteration of the educational program components, procedures and feasibility and