For people living with HIV/AIDS, Differentiated Service Delivery (DSD) has focused on enhancing resilience, self-efficacy, and engagement. For people co-infected with HIV/AIDS and multidrug resistant tuberculosis (MDR-TB), particularly in subSaharan Africa, there are severe challenges associated with treatment, including stigma, social and structural barriers. We used empirical adherence data and qualitative research to identify longitudinal barriers to medication adherence to inform MDR-TB HIV DSD models.
Adults with MDR-TB and HIV initiating bedaquiline (BDQ) and receiving antiretroviral therapy (ART) in KwaZulu-Natal, South Africa were prospectively enrolled and followed through the end of MDR-TB treatment. Separate electronic dose monitoring devices (EDM) (Wisepill RT2000) measured BDQ and ART adherence through six months, calculated as observed versus expected doses aggregated at a weekly level. We defined severely adherence challenged as < 85% cumulative EDM measured doses of ART and BDQ. Longitudinal focus groups were conducted by trained staff and transcripts were analyzed thematically to describe early, middle, and late-stage treatment challenges.
From November 2016 through February 2018, 199 participants with MDR-TB and HIV were enrolled and followed through treatment completion (median 17.2 months IQR 12.2–19.6). 12 focus groups were conducted. While the majority (83.2%, 166/199) maintained high adherence, a severely adherence challenged subpopulation (16.8%, 33/199) had a precipitous decline in mean BDQ adherence from 91.9% to 44.7% and mean ART adherence from 84.5% to 21.6% over six months (F1, Panel A, B). Qualitative analysis identified discrete treatment stages associated with specific barriers (F1, Panel C) which, when aligned with quantitative data, suggests that declining medication adherence may relate to psychosocial, behavioral, and structural barriers.
Based on these data, MDR-TB HIV DSD frameworks should 1) intensify support for severely adherence challenged subpopulations while adherent patients may require less intensive support, 2) address decreased adherence over time and 3) account for psychosocial, behavioral, and structural challenges linked to discrete treatment stages. DSD models that offer evaluation and intervention at key stages, tailored to needs of both vulnerable and adherent populations, have the potential to improve adherence and outcomes in MDR-TB HIV treatment.
Panel A: Weekly mean antiretroviral (ART) adherence with 95% confidence intervals calculated weekly based on observed versus expected electronic pill box opening through six month stratified by severe adherence challenged ( < 85% adherence). (Nf199) Panel B: Weekly mean bedaquiline (BDQ) adherence with 95% confidence intervals calculated weekly based on observed versus expected electronic pill box opening through six month stratified by severe adherence challenged ( < 85% adherence). (Nf199) Panel C: Qualitative stages of MDR-TB HIV treatment with thematically derived stage specific care & treatment challenges.