Qualitative Interviews

Interviews were conducted by authors JBT and RB, who are both bilingual in the local language (Runyankole) and English and highly experienced, well-trained male qualitative research assistants. One-time interviews were digitally recorded for later transcription and took place in private settings; most occurred in the study office or near the clinic, although all MoH interviews were conducted in the participants’ offices in Kampala, and some participants were interviewed at home, per their preference. Interviews with health care administrators/clinicians were conducted in English, which is commonly used in professional settings; interviews with ART clients were conducted in Runyankole or English per participant preference. Interviews began with an introduction to the research assistants, followed by statements of no conflicts of interest, a desire for honest perceptions (favorable or unfavorable), and the overall goals of the study. Participants were then asked for basic demographic data. A description of the electronic adherence monitors, associated interventions, evidence for their use, logistical requirements, and costs was subsequently read to participants (Multimedia Appendix 1). Participants were also shown an electronic adherence monitor and the software interface for displaying adherence data (Figure 2). Interview guides (Multimedia Appendix 2 and 3) were designed to obtain unbiased impressions of the technology and its potential for supporting ART in routine care, while also assessing each of the 5 domains in the CFIR. The guides were tailored for anticipated perspectives of health care administrators/clinicians versus ART clients. Initial questions in both guides asked about likes and dislikes of the features and functions of monitor and associated interventions and were followed by questions about their potential to influence HIV care, suggestions on how to measure their value, and recommendations for their use in routine care. Health care administrators/clinicians were also asked about the technology in relation to other health care priorities (ie, the outer setting). Questions were informally pretested with KCRC staff and clients and revised to ensure clarity and utility.

Adherence data display: each dot indicates the date and time of a monitor opening as a proxy of medication ingestion.

Research assistants wrote debriefs after each interview to capture body language, participant mood, and any other nonverbal aspects of the interviews. Transcripts were reviewed for quality among authors LG, BFB, JBT, and RB and corrected as needed. Participants were interviewed until thematic saturation was achieved.