Feasibility, acceptability and effectiveness of electronic monitoring devices on adherence in adolescents in Zimbabwe

Background: Non-adherence to both prophylactic and therapeutic treatment remains a major challenge among adolescents, particularly those with chronic conditions, who constitute an estimated 10–30% of the global adolescent population. Sub-optimal adherence to long-term therapies remains a major public health concern, contributing to increased healthcare use, medical complications, and mortality. Early identification of treatment lapses is essential for developing effective adherence support strategies; however, accurately measuring adherence remains difficult in the absence of a goldstandard method. Electronic monitoring devices (EMDs) offer objective, day-to-day adherence data and may help identify those at risk of non-adherence. While studies in children and adults from high- and middle-income countries suggest EMDs can improve adherence and clinical outcomes, evidence in adolescents from low-income settings is limited. This PhD thesis aimed to assess the feasibility, acceptability, and effectiveness of EMDs for supporting adherence in adolescents in Zimbabwe. Methods: This PhD comprised of four interrelated studies conducted between February 2022 and December 2024, investigating the use of electronic monitoring devices (EMDs) to support medication adherence among adolescents. Study 1 was a systematic review of randomised controlled trials evaluating EMD-informed interventions and their impact on adherence and clinical outcomes in children and adolescents across diverse geographic settings. Studies 2 and 3, nested within the Zimbabwe site of the VITALITY trial (PACTR202009897660297), evaluated the validity, feasibility, and acceptability of an EMD among 97 adolescents aged 11–19 years with perinatally acquired HIV receiving 48- week vitamin D supplementation. Study 4 was a randomised controlled trial that enrolled 200 young people aged 16–30 years at risk of virological failure (viral load >200 copies/mL) in Harare, Zimbabwe, comparing standard care (existing enhanced adherence counselling services) with an EMD-based intervention involving text message reminders and monthly adherence feedback for six months. Results: The systematic review (Study 1) included 12 RCTs and found that EMD-informed interventions were associated with improved adherence, though evidence for their effect on clinical outcomes was mixed. Most studies focused on younger children in high- or middle-income countries, with limited data from adolescents or low-income settings. The proof-of-concept trial (Study 2) showed that EMD use without feedback or reminders was associated with improved adherence, as indicated by serum 25 hydroxy-vitamin D 25(OH)D levels, and identified distinct adherence patterns. The mixed-methods process evaluation (Study 3) demonstrated a positive correlation between EMD-measured adherence and 25(OH)D levels, suggesting greater validity than pill count, and found EMDs to be feasible and acceptable, though concerns about privacy and design were noted. In the randomised trial (Study 4), a six-month EMD-informed intervention improved viral suppression compared to standard care. However, this effect was not sustained at nine months post-withdrawal of the intervention. Conclusions: This thesis provides evidence that EMDs are feasible, acceptable, and effective for supporting medication adherence among adolescents on long-term therapies in low-income settings. EMD-informed interventions were associated with improved adherence and clinical outcomes, though effects were not consistently sustained. EMDs offered a more reliable measure of adherence than traditional methods and may help identify adolescents needing targeted support. This work supports the integration of digital adherence tools into adolescent care in resource-limited settings. Further research is needed to assess long-term effectiveness, contextual fit, and cost, as well as to develop more affordable, discreet devices that meet adolescents’ needs.