Assessing the Feasibility, Acceptability, and Appropriateness of Implementation Strategies Developed to Support an Enhanced Oral Anticancer Agent Adherence Intervention in Academic, Rural Community, and Urban Community Cancer Settings

Background: Oral anticancer agents (OAAs) have changed oncologic treatment and improved patient outcomes in clinical trials (adherence ≥90%), but due to lack of monitoring, side effects, cost, and forgetfulness, benefits have not translated into practice where adherence rates are~ 40%. Our team designed an effective adherence intervention and worked with a stakeholder panel to develop implementation strategies. Overall, we developed 6 distinct strategies for adoption (data driven presentation for program leaders and formal agreement materials), implementation (motivational interviewing (MI) provider course and standard operating procedure (SOP)), and maintenance (key performance indicators (KPIs) and electronic health record (EHR)-integrated adherence assessment) using focus group discussions with clinicians and administrators. Here we use clinician surveys to assess feasibility, acceptability, and appropriateness of strategies.

Methods: We administered surveys to establish feasibility, acceptability, and appropriateness of implementation strategies of a medication adherence intervention for OAAs. We adapted surveys using Weiner et. al.'s 2017 psychometric assessment of newly developed implementation outcomes that assigns 15 domains amongst constructs of feasibility, acceptability, and appropriateness to evaluate implementation effort success. The assessment was adapted to make it relevant to the OAA intervention. Participants were queried about feasibility (if it was possible, implementable, doable, easy), appropriateness (if it was a good match, good fit, applicable, suitable), or acceptability (if it was welcome, liked, showed acceptability). Participants rated agreement that implementation strategies exhibited domain characteristics using 5-point Likert scale- 1 (strongly disagree) to 5(strongly agree). Participants also answered open-ended questions by indicating their favorite aspects and suggested recommendations for each strategy. We pilot tested the survey with an advisory panel and edited it using feedback received prior to administration. We performed descriptive statistics for all findings.

Results: We administered a survey to hematology and oncology clinicians (n=33) in academic medical centers (n=23), rural community-based centers (n=2), and urban community-based centers (n=7), (no response, n=1). Survey response rate was 31.4%.

There were high rates of agreement on the feasibility of SOPs, data driven presentation, and EHR-integrated assessments with 89%, 82%, and 95% of participants indicating a score of ≥4 on a 5-point Likert scale, respectively. There were a moderate rates of agreement on feasibility of KPIs (75%), formal agreement materials (66%), and MI course (50%).

There were high rates of agreement on the appropriateness of SOPs, data driven presentations, EHR-integrated assessments, and KPIs with 89%, 93%, 93%, and 82% of participants indicating a score of ≥4 on a 5-point Likert scale, respectively. There were moderate rates of agreement for the MI course (77%) and formal agreement materials (66%).

There were high rates of acceptability on data driven presentations, EHR-integrated assessments, KPIs (94-98%), and moderate rates of acceptability on SOPs, formal agreement materials and MIs (67% for all items) as indicated by a score of ≥4 on a 5-point Likert scale.

Strategy components that showed a high rate of agreement from participants for inclusion are summarized in Table 1.

Participants’ favorite part of the presentation was benefit explanation, but they wanted brevity. They appreciated formal agreement for strategic accountability but feared negative consequences. SOPs provides clear roles and helpful diagrams but no flexibility. Participants felt KPIs identify financial impact, but ED visits should be excluded. MI was found effective, but without the hands-on training component. Smart forms were needed for discrete data but difficult to operate.

Conclusions: There were moderate to high levels of agreement on most strategies designed to mitigate barriers to implementation of an OAA intervention. Formal agreement materials and motivational interviewing course were shown to be less appropriate and feasible than other strategies. Efforts must be focused on maintaining brevity, highlighting benefit, and clearly defining clinician roles.