The Prevalence and factors associated with nonadherence with ARV treatment and cotrimoxazole preventive therapy among HIV infected adolescents attending out-patient HIV clinics in Kenya

Date of Award


Document Type


Degree Name

Master of Medicine (MMed)

First Supervisor/Advisor

Prof. Fred Were

Second Supervisor/Advisor

Prof. Elizabeth Anne Bukusi

Third Supervisor/Advisor

Dr. Josephine Omondi


Paediatrics and Child Health (East Africa)


Background: There is an increase in the burden of HIV infected adolescents, both those perinatally infected as well as those acquiring HIV during adolescence. Female adolescents continue to be at the highest risk for acquiring HIV; in many Southern African countries, they have a three fold risk of acquiring HIV compared to their male counterparts. Adherence to treatment among adolescents has been shown to range between 30-70%, which is unacceptably low for antiretroviral therapy (ART). Psychosocial wellbeing and social support have been shown to be possible contributing factors to adherence to ART. This relationship has not been previously evaluated among Kenyan adolescents.

Objectives: The overall objective of this study was to determine prevalence of non-adherence to medication among HIV infected adolescents aged 13-18 years attending selected outpatient HIV clinics in Kisumu, Kenya. The secondary objectives included determining the effect of psychosocial well being on adherence, and determining factors, including peer group support and their effect on adherence to medication. Methods We conducted a multi-center retrospective cohort study at seven outpatient HIV clinics in Kisumu, Kenya; enrolling 285 adolescents aged 13-18 years. Adherence data was obtained from pharmacy refill data and for each subject and a percentage adherence computed as the proportion of completed scheduled pharmacy visits. The main v predictor variable, psychosocial well being data was collected using a validated tool and a psychosocial score calculated using the corresponding score sheet and categorized as good (score was >22), moderate (15-22) and poor (<15). The maximum possible score was 30. Demographic data on potential determinants of adherence were collected in a face-to-face interview using a structured questionnaire. Characteristics of study participants were summarized using means and standard deviations for continuous variables; counts and proportions for categorical variables. The associations between adherence, psychosocial well-being and other factors were assessed using univariate and multivariate logistic regression.

Results: The mean age of the participants was 15 years (mode 13, median15), 59% of whom were female. The majority (67%) had been enrolled into care when less than 12 years old and therefore considered to have been infected perinatally; 52% were on ART. The overall average adherence was 86%. Adolescents were categorized as adherent (adherence >95%) or non-adherent (adherence <95%) and 65% of them were adherent. Poor psychosocial well being was associated with increased likelihood of poor adherence (OR 3.37 CI 1.17 to 9.69; p=.017). Mental health showed a tendency to affect adherence negatively (p=.09). Other factors such as treatment buddies and peer support groups were not significantly associated with adherence. In addition attending peer group meetings did not significantly affect psychosocial well being. vi

Conclusions: The overall adherence among adolescents on HIV care was acceptable. Psychosocial wellbeing was associated with higher adherence.

Recommendations: Psychosocial well being assessment may be a useful addition to adolescent HIV programs. Interventions to improve psychosocial well being should be evaluated.

This document is available in the relevant AKU library