Abstract:
Background: Kisumu County’s HIV prevalence of 19.9% is about three times higher
than the national average. In 2015, 52% of new HIV infections reported in the County
were among adolescents. Globally, HIV viral suppression rates among adolescents are
low. Kenya has adopted the 90-90-90 targets by joint United AIDS (UNAIDS), one of
which is to have 90% of all people receiving anti-retroviral therapy (ART) virally
suppressed by 2020.
Objectives: The objective of study was to estimate prevalence of viral non-suppression
(VNS) and identify factors associated with VNS among HIV positive adolescents on
ART at Chulaimbo Sub-County Hospital.
Methods: This was a cross-sectional study among HIV positive adolescents (10-19
years) on ART for >6 months. They must have had a routine viral load (VL) test done
in ≤ 6 months with valid laboratory results. Study sample was randomly selected from
laboratory VL register. A VL of ≥1000 RNA copies/ml was considered non-suppressed.
Using a pre-tested structured questionnaire, disclosure status and socio-demographic
data were obtained; adherence to ART was measured by pill counts and abstraction for
clinical data was done from medical records. Data analysis was done using Epi info
version 7 and STATA software. To identify factors associated with VNS, we conducted
bivariate analysis for crude odds ratio (cOR) and stratified logistic regression to control
for confounding and check for effect modification by computing adjusted OR (AOR).
Results: Out of 398 adolescents on ART, 212 had been tested for routine VL ≤ 6 months
with valid VL results. Two hundred adolescents were randomly selected and
interviewed; 103 (51.5%) were female; younger adolescents (10-14 years old) were 102
(51%) and 98 (49%) adolescents were 15-19 years old. The median duration on ART
was 7.6 years (range 0.8-12.4years). One hundred and forty-two (71%) adolescents
were on first line regimen and 58 (29%) on second line regimen; 110 (55%) had good
(≥ 95%) adherence to ART. Seventy-one (35.5%) adolescents had VL result of ≥1000
RNA copies/ml. The odds of VNS were higher among those who had poor adherence
to ART and there was effect modification with increasing age; older adolescents (AOR
13.93, 95% CI 3.20-60.74) compared to younger adolescents (AOR 4.41, 95% CI 1.18-
16.54), and in males (AOR 22.33, 95% CI 4.35-114.63) compared to females (AOR
3.92, 95% CI 1.09-14.04). Being on first line regimen was negatively associated with
VNS for males (AOR 0.06, 95% CI 0.00-0.87) and older adolescents (AOR 0.02, 95%
CI 0.00-0.20).
Conclusion: The proportion of VNS was more than three times higher than the
acceptable based on the 90% viral suppression target. Adolescents on second line
regimens and those who are non- adherent to ART have higher odds of VNS, the odds
being higher in males and older adolescents.
Recommendations: County ART program should reinforce targeted interventions to
motivate adherence in HIV infected males and older adolescents towards attaining the
90% target and improve the treatment outcomes.