Background: World Health Organization (WHO) defines adolescents as those aged 10–19 years.
They represent 25% of the population in sub-Saharan Africa. Eighty-two percent of the estimated
2.1 million adolescents aged 10–19 years living with human immunodeficiency virus (HIV) in 2012
were in sub-Saharan Africa. Adherence to antiretroviral therapy (ART) is vital to HIV-infected
adolescents for survival and quality of life. However, this age group faces many challenges to
remain adherent.
According to the Kenya National surveillance of acquired HIV drug resistance in 2013; only 67% of
adolescents had viral suppression which reflects poor adherence. It is important to define
adherence rates in different groups of adolescents and in the proposed study we explore
adherence in those residing in rural areas.
Objectives: The primary objective was to determine the level of optimal ART adherence among
HIV infected adolescents on ART at the Kangundo District Hospital (KDH). The secondary objective
was to describe the factors associated with adherence in the same population of adolescents.
Methods: This was a retrospective study with a cross-sectional qualitative component. HIV
infected adolescents aged 10 to 19 years who had been on ART and in care at KDH for at least six
months were eligible. Disclosure status was determined from the caregiver and, adolescents who
fulfilled the inclusion criteria were enrolled in to the study after informed written consent from
caregiver and assent from the adolescent. Recent ART adherence was determined by 3, 7 and 30
days recall. Long term adherence was determined as follows: adherence to pharmacy refill with
data abstracted from pharmacy database, and adherence to clinic appointments with data
abstracted from hospital records for six months prior to the interview date.
Factors associated with adherence to ART were sought including socio-demographic factors,
clinical factors, drug related and health system related factors. Focus group discussions were held
with HIV infected adolescents in care and who were aware of their HIV status to determine to a
greater depth the factors impacting adherence
x
Results: A total of 98 adolescents aged between 10 to 19 years were enrolled into the study with a
median (interquartile range (IQR)) age of 14.0 (13.-16.) years. Majority (76.2%) were in their
teenage years (13-19 years) and none of the respondents was married. Most (78.6%) were in, or
had completed primary school while a few (17.3%) were in secondary school or had already
completed secondary school education. Most of the participants (91.8%) were on first line ARV
regimen with a median (IQR) duration on ART 6.3 (4.4-8.1) years. At the time of initiation of ART,
11.2%, 35.7% and 53.1% of the study participants were at Stage one, two and three/four of
HIV/AIDS respectively as described in the WHO guidelines. Majority 65.3% of the participants had
disclosure of their HIV status while the rest (34.7%) had not.
The short term 3, 7 and 30 days optimal adherence, measured by self-report was poor with 76%,
55% and 69% being adherent respectively. Similarly the long term adherence over preceding 6
months as determined from hospital records was poor with only 66.3% being adherent to
clinician’s appointment and 64.3% optimally adhering to drug refill appointments.
From the univariate analysis several factor were shown to influence adherence negatively
including: being in boarding school (OR 8.47, 95% CI 2.37-30.26, P=0.001), feeling tired of taking
drugs daily (OR 0.13, 95% CI 0.05-0.37, P <0.001), reporting that dosing at specific times interfered
with other daily activities (OR 0.27, 95% CI 0.10-0.71, P=0.006), non-disclosure of HIV status
especially for adolescents aged >14 years (OR 8.5, 95% CI 1.57-46.08, P 0.009) and inconvenient
appointment dates (OR 3.17, 95% CI 1,25-8.03, P 0.03). Clinicians’ behavior perceived as
good/excellent (OR 4.79, 95% CI 1.28-17.91, P 0.032), similarly pharmacists’ behavior (OR 5.42,
95% CI 1.25-23.39, P 0.022) and adequate health education on HIV infection (OR 4.44, 95% CI 1.31-
15.01, P 0.021) were associated with better adherence.
Multivariable analysis revealed that adolescents who were in day school were more likely to be
adherent compared to those who were in boarding school (adjusted odds ratio (aOR) 7.99, 95% CI
2.85 – 22.41, p<0.001). Appointments fitting with daily activities was associated with about fourfold
increment in the likelihood of adherence to ART (aOR (95% CI) 4.22 (1.51-11.83), p=0.006).
xi
FGDs with adolescents’ revealed that good social support, disclosure about HIV status to the
adolescents, to other family members and at least one person in school were perceived to
improve adherence. Health providers’ behavior and attitude towards adolescents were also
perceived to influence adherence with most adolescents reporting missed doses due to
disappointments and dissatisfaction with the services provided at the clinic.
Conclusion: We concluded that adherence to ART among this cohort of adolescents in a rural set
up was poor as shown by the self- report, clinicians’ appointments and pharmacy refill
appointments’. Poor adherence was significantly associated with age above 15 years, nondisclosure
of HIV status, being in boarding school, feeling tired of taking drugs daily and dosing
times interfering with other daily activities. Similarly, hostile health providers attitude, inadequate
time spend by clinician addressing the adolescents needs, long waiting times, lack of privacy at the
hospital and inconvenient timings of appointments were perceived as barriers to adherence.
Stigmatization arising from the location of the CCC shows a need to re-design the service to
protect confidentiality of adolescents living with HIV.
Recommendations: Our findings suggest need for strengthening of interventions to assist
adolescents’ in adhering to their medication including: Early disclosure, continuous psychosocial
support, involvement of family and school “buddy” in adolescent care especially those in boarding
schools. There is need to tailor the health services to be adolescent friendly at the level health
provider, timing of appointments and privacy.