Study Aim 1 will explore patient, health care system, and structural factors that influence adherence to AHT (Years 1-2). To achieve this aim, in-depth, semistructured qualitative interviews will be conducted with health care professionals who work in oncology settings and with breast cancer survivors. For purposes of this research, we define a “breast cancer survivor” as anyone who has been diagnosed with invasive breast cancer.

Health care professionals will be recruited to participate in qualitative interviews. The study will target health professionals from key networks and organizations in South Carolina, including professional organizations (eg, oncology nursing, oncology social work), cancer centers and South Carolina’s cancer coalition and department of health. The primary method of recruitment will be through email invitation letters sent by the study’s principal investigator (PI).The invitation letter will describe the study aims, study procedures, funding source, and institutional review board approval protocol information. Interested health care professionals will be able to contact the study PI directly by telephone or email. The PI will schedule an interview at a time and location convenient to the potential participants to determine their eligibility for the study. Eligibility requirements will include (1) being employed as a health care professional (ie, physicians, nurses, pharmacists, social workers, patient navigators); (2) currently working with breast cancer patients, for example in an oncology setting, such as a cancer center or hospital; and (3) age ≥ 21 years.

Before the interview, health care professionals will be asked to complete a brief questionnaire about their personal (eg, gender, age) and professional (eg, current job title, type of health care professional, health care setting) information. The PI will conduct all interviews using a guide developed by the study team. Key topics include: (1) major barriers to AHT adherence; (2) organizational resources to support posttreatment (eg, surgery, chemotherapy) of breast cancer survivors; (3) communication among health care team members; (4) communication with patients and their families; and (5) organizational strategies for addressing problems with AHT adherence. Participants will receive a US $50 cash stipend for their time and effort. At the end of the interview, participants will be asked for their contact information if they are willing to participate in a follow-up feedback session, which will serve to inform the development of a subsequent intervention. All interviews will be digitally recorded and field notes will be taken.

Breast cancer survivors will be recruited from the South Carolina Oncology Associates (SCOA) and the Gibbs Cancer Center and Research Institute (GCCRI). Eligibility requirements include: (1) age≥21 years; (2) diagnosed with HR+, invasive breast cancer; (3) having medical records that confirm the prescription of any hormonal treatment (anastrozole, exemestane, letrozole, tamoxifen) at any point after diagnosis; (4) eligible for or enrolled in South Carolina Medicaid program; and (5) being able to speak and read in English.

To recruit participants, we will apply a recruitment model developed by Heiney and colleagues [30], which combine social marketing with relationship building. The PI and a research assistant will work with SCOA and GCCRI staff to identify participants who meet the study eligibility criteria using electronic medical records. SCOA patients will be mailed a personal invitation letter from the PI explaining the goals of the study, including a Frequently Asked Questions document that addresses potential concerns of research participants. The letter will include a phone number to call if the participant prefers not to be contacted. Within 3-5 days after the recruitment letter is mailed, the PI or research assistant will contact the participant by phone to provide them with an overview of the study. If the participant expresses interest in the study, the PI or research assistant will verify the participant’s study eligibility. If eligibility is met, an interview appointment will be scheduled at a time and location convenient to the participant. If a participant is not eligible, the PI or research assistant will let the participant know that they did not meet the study criteria and will thank them for their interest in participating.

At GCCRI, designated staff will contact potentially eligible breast cancer survivors by phone and provide them with an overview of the study. If the participant shows interest in the study, GCCRI staff will then ensure the participant meets study eligibility. If eligibility is met, GCCRI staff will request verbal consent by phone and set up an interview appointment at GCCRI. Interested and eligible participants will be mailed an “Appointment Form” confirming the date, time, and location of the interview; a copy of the informed consent; and the Frequently Asked Questions document.

Interviews with participants recruited from SCOA will be scheduled and conducted at a location (eg, home, library) and time convenient to the participant, while GCCRI participant interviews will be scheduled in person and onsite at GCCRI. Before the interviews, the following informed consent process will be conducted: a research team member (PI or research assistant) will read aloud a consent form that will inform the participant about the following components of the study: title of the research project, the identity of the project PIs and their respective contact information, an introduction about the study, the aims of the study, reason for the project, a summary of why the individual has been asked to participate, a description of the study procedures (ie, audiotaping of interviews), an outline of any risks and/or benefits, an overview of how confidentiality will be maintained, a discussion of any benefits, and a method of securing additional information or asking questions. After the consent form is read aloud and the participant agrees to participate in the study, the research team member and participant will sign and date the form. Participants will be given a copy of the form for their records.

Following consent, participants will be asked to complete a brief information form about their personal information (eg, age, race, marital status), hormonal therapy (eg, type of drug, daily use), and type of AHT resources used and preferred. The research team will conduct all interviews using a semistructured interview guide. Breast cancer survivors will be asked to describe their experiences with AHT, including experiences communicating with health care professionals about AHT, and strategies used for addressing AHT nonadherence. All participants will be given a US $50 cash stipend for their time and effort. At the end of the interview, participants will be asked if they are willing to participate in the intervention development phase of this study (Aim 2).

All interviews will be recorded and transcribed verbatim by a professional transcription service. Each participant’s audio file and transcript will be assigned the same unique identifier as their corresponding brief questionnaire. Quality control checks will be conducted while reviewing the transcripts to ensure that no personal identifiers are used. Audiotaped transcripts and field notes will be used for data analyses. Transcripts will be independently read and reviewed to check for accuracy and authenticity.

To identify potential modifiable factors and future intervention targets for Aim 2, Assarroudi et al’s [31] qualitative content analysis approach will be used to analyze interview transcripts. The PI will develop a preliminary codebook of main categories and subcategories derived from the WHO Multidimensional Adherence Model Framework [25] and Multilevel Context of Cancer Care [26]. The PI will work with other team members to discuss the codebook as well as the final main categories and subcategories and their related meanings. The PI and two other research team members will independently read and code a selection of transcripts using the preliminary codebook. Following this initial coding, all team members will meet, compare coding results, and discuss differences in coding interpretations and concerns with the preliminary codebook. The PI will revise the codebook as needed. Once the codebook is finalized, a research assistant who is knowledgeable in the study topic area will, independently, code all the transcripts, and the PI will code a random sample of the transcripts. The PI and research assistant will then meet and discuss coded transcripts, reach a consensus among any discrepancies and calculate intercoder agreement [32].

Multiple sources will be used for the development of a theory-based, multilevel intervention program, including findings from Aim 1, current scientific literature, insights from participant feedback sessions (described below), and expertise of the project mentoring team. After analyzing the qualitative data from the breast cancer survivor and health care professional interviews, we will develop a modified, data-informed logic model of potential interventions aimed at health care professionals and their organizations, and survivors and their family/social support. Logic models, representing an outcomes hierarchy, will assist with a broad understanding of the components of each potential intervention [33]. Potential intervention strategies are shown in Table 1 [27-29]. The research team will systematically review and compare findings from the survivor and health care professional qualitative interviews and quantitative questionnaire data. Of interest is not only the variation within the survivor and health care professional groups, but also the variation between the survivor and health care professional reports. We will, therefore, identify themes that occur within and across groups. We will employ a logic model based on the PRECEDE approach [34] to illustrate survivors’ experiences with AHT and how knowledge, perceptions and attitudes at different levels impact their ability to adhere to AHT. We will apply appropriate multi-level theories to refine the intervention strategies and messages [35-37].

From the group of participants that agreed to participate in a follow-up feedback session, we will select a purposive sample of breast cancer survivors (n=5; and their designated family member/friend, n=5) and health care professionals (n=5), based on their brief questionnaire and interview responses. We will host two, 1 to 2 hours feedback sessions: one for breast cancer survivors and their family member/social support, and another one for health care professionals. The feedback sessions will provide participants with the opportunity to judge the accuracy and credibility of the qualitative research findings, as well as to provide alternative interpretations and explanations when necessary [38]. We will also discuss any differences of opinion that may appear between the survivor and health care professional interviews to reach a resolution. Participants will receive US $50 and a light meal for their time and effort upon completion of the session. The project mentoring team will provide guidance on how to refine the intervention design based on the data from the feedback sessions and resolve any data discrepancies.