Predictors of suboptimal adherence to isoniazid preventive therapy among adolescents and children living with HIV

OUTPUT TYPE: Journal Article
TITLE AUTHOR(S): A.W.Kay, N.Thivalapill, D.Skinner, G.S.Dube, N.Dlamini, B.Mzileni, P.Fuentes, P.Ustero, L.V.Adams, A.M.Mandalakas
DEPARTMENT: Public Health, Societies and Belonging (HSC)
Print: HSRC Library: shelf number 11907
HANDLE: 20.500.11910/15932

If you would like to obtain a copy of this Research Output, please contact Hanlie Baudin at


This study identified factors associated with adherence to a 6-month isoniazid preventive therapy (IPT) course among adolescents and children living with HIV. Forty adolescents living with HIV and 48 primary caregivers of children living with HIV completed a Likert-based survey to measure respondent opinions regarding access to care, quality of care, preferred regimens, perceived stigma, and confidence in self-efficacy. Sociodemographic data were collected and adherence measured as the average of pill counts obtained while on IPT. The rates of suboptimal adherence (< 95% adherent) were 22.5% among adolescents and 37.5% among the children of primary caregivers. Univariate logistic regression was used to model the change in the odds of suboptimal adherence. Independent factors associated with suboptimal adherence among adolescents included age, education level, the cost of coming to clinic, stigma from community members, and two variables relating to self-efficacy. Among primary caregivers, child age, concerns about stigma, and location preference for meeting a community-health worker were associated with suboptimal adherence. To determine whether these combined factors contributed different information to the prediction of suboptimal adherence, a risk score containing these predictors was constructed for each group. The risk score had an AUC of 0.87 (95% CI: 0.76, 0.99) among adolescents and an AUC of 0.76 (95% CI: 0.62, 0.90), among primary caregivers suggesting that these variables may have complementary predictive utility. The heterogeneous scope and associations of these variables in different populations suggests that interventions aiming to increase optimal adherence will need to be tailored to specific populations and multifaceted in nature. Ideally interventions should address both long-established barriers to adherence such as cost of transportation to attend clinic and more nuanced psychosocial barriers such as perceived community stigma and confidence in self-efficacy.