Evaluating an Economic Empowerment Intervention on Medication Adherence for HIV Positive Youth in Uganda: Longitudinal Data Collection Challenges Reported from the Field

Background and purpose: Conducting longitudinal studies in poverty-impacted regions, like Sub-Saharan Africa, pose significant data collection and coordination challenges. The Suubi+Adherence trial is a National Institutes of Health (NIH) longitudinal study evaluating an economic empowerment intervention’s impact on ART adherence for HIV+ adolescents in Uganda. Adherence data is collected by local Research Assistants (RAs) in multiple ways, including electronic medication monitoring devices (Wisepill) and pill-counts. RAs conduct visits to participant homes every 30-60 days to monitor Wisepill devices, renew batteries, conduct pill-counts, and answer program-related questions. Visits are documented in field reports that note family concerns, and RA challenges. This article focuses on RA-reported challenges during home visits for the first two years (2014-2015) of the study.

Methods: The research team coded 247 field reports selected randomly from field reports recorded with all participants (N=702) during 2014-2015, to maximize variation across RAs, treatment arm, and time period. Data was coded using Dedoose. Thematic analysis using an inductive approach was utilized to analyze themes and subthemes. The research team first read 10 field reports to get a comprehensive understanding of a subsection of the data. Then, the team independently coded them and discussed the codes. Through discussions and iterative coding exercises, a codebook was finalized.

Results: Two major categories emerged for local RAs: Program-related and context-specific challenges. Program-related challenges included clarifying intervention scope and managing family expectations. Requests for school fees to be paid by the program, and for more frequent visits by the study team were common. Occasionally, RAs were asked high-level health-related questions by participating families, which fell outside of their expertise and responsibility. Pill-counting posed challenges as well. Multiple medications for family members being stored together, sporadic record-keeping on prescription refills, and conditions in some rural homes undermined pill-counts. Wisepill-related challenges included devices being broken/malfunctioning or misplaced/lost.

Wisepill usage and tracking were also impacted by context-specific challenges related to limited infrastructure. More specifically, poor connectivity in the area compromised the ability to track Wisepill cellular-signal and interfered with RAs’ ability to troubleshoot via cell phone. Another context-specific challenge that interfered with data collection was poor road conditions that became especially problematic when long distances needed to be covered during the rainy season. Finally, given that home visits were unannounced, caregivers were often unavailable during home visits primarily due to commitments on their gardens.

Conclusions and Implications: The challenges articulated by RAs carry implications for data collection and field coordination. Difficult caregiver questions and expectations underscore the importance of clear guidelines and training for RAs to appropriately field these concerns.  Problems encountered in pill-counting point to the need for continued efforts to raise family awareness on the importance of ensuring sanitary conditions and tracking of pills.  Caregiver/participant unavailability during unannounced home visits warrant the consideration of alternative meeting places for field visits. Finally, Wisepill posed multiple challenges for RAs that resulted from technical problems with the devices, connectivity issues, and participant/family engagement with the device. Hence, further research is needed to examine the feasibility of Wisepill utilization in resource-poor contexts.