Abstract
BACKGROUND: We have a limited understanding on how to best integrate technologies to support antiretroviral therapy (ART) adherence in routine HIV care. METHODS: We conducted semi-structured interviews with multidisciplinary providers caring for pregnant and postpartum people with HIV and asked providers about their perspectives on utilizing adherence support technologies such as text messages, video check-ins with providers or automated with facial recognition for directly-observed-therapy, signaling pill bottle, and signaling pill to support ART adherence. Each approach generated an adherence report. The interview instrument was guided by the Consolidated Framework for Implementation Research and included questions on the implementation climate, barriers, and facilitators to the clinical integration of the adherence approach and strategies that could be used to maximize this integration. The order of adherence support technologies was randomized to minimize bias. We used a modified grounded theory to develop the coding structure and two coders applied the codebook to the transcripts after establishing strong inter-rater reliability with 20% of interviews (kappa = 0.82). RESULTS: Between March and December 2020, we conducted 26 in-depth, semi-structured interviews with providers who weighed several factors when considering each approach, including the approach's effect on patient-provider interaction in and outside of the clinic visit, timing for and duration of the approach's utility, threat of disclosing status, and added burden to providers (e.g., needing to act on generated information) or to patients (e.g., needing to hide the signaling pills, responding to text messages). Providers' most preferred approach was text-messages, and the least preferred was the signaling pill. Barriers to acceptability varied by approach and included perceived surveillance, violation of privacy, added time demand for providers, potential inaccuracy of the adherence data generated, and negative impact on the patient-provider relationship, particularly if the approach was perceived as coercive. Payers anticipated regulatory hurdles with unfamiliar approaches, particularly the signaling pill and signaling pill bottle. Facilitators included strengthened therapeutic alliance, predictable reminder mechanisms, and options for customization according to patient preference. CONCLUSIONS: Our study elucidates barriers and facilitators to integrating technology-based adherence support approaches in clinical care to support adherence of pregnant and postpartum people with HIV.