Evidence of reduced treatment adherence among HIV infected paediatric and adolescent populations in Nairobi at the onset of the UNAIDS Universal Test and Treat Program
Kabogo et al. BMC Res Notes
Evidence of reduced treatment adherence among HIV infected paediatric and adolescent populations in Nairobi at the onset of the UNAIDS Universal Test and Treat Program
Joseph Kabogo 0
Elijah Songok Elijah.Songok@umanitoba.c 0
0 Kenya Medical Research Institute (KEMRI) , Mbagathi Road, Nairobi , Kenya
Objective: We conducted a retrospective cohort study to evaluate the efficacy of the World Health Organization (WHO) “Universal Test and Treat” (UTT) policy, initiated in Kenya in September 2016. Under this policy, every human immunodeficiency virus (HIV)-infected person should be initiated on antiretroviral therapy (ART). We compared intraand inter-group viral suppression and ART adherence rates for pre-UTT (initiated on ART in March-August 2016) and UTT groups (initiated in September 2016). The study was conducted in a community outreach Program in Nairobi with 3500 HIV-infected children enrolled. Results: 122 children and adolescents were initiated on first-line ART pre-UTT, and 197 during the UTT period. The 6 month viral suppression rate was 79.7% pre-UTT versus 76.6% UTT (P < 0.05). Suboptimal adherence was higher in the UTT than pre-UTT period (88 of 197, 44.7% and 44 of 122, 34%; P < 0.001). The decrease in adherence was greater among orphans (91.7% pre-UTT and 87.2% UTT, P = 0.001) and children 11-18 years. Our results show that successful implementation of the UTT policy in Africa is challenged by an increased risk of suboptimal adherence. There is a need to develop extra strategies to support adherence, especially among orphans and teenagers.
Universal test and treat; 90-90-90; HIV; Children; Antiretroviral therapy; Adherence; Virologic failure; Kenya
Sub-Saharan Africa accounts for 71% of all Human
Immunodeficiency Virus (HIV)-infected individuals,
and 91% of all HIV-infected children below the age of
15 years [
]. In Kenya, there are approximately 190,000
HIV-infected children and adolescents, of whom only
38% are on antiretroviral therapy (ART) [
]. In 2014,
the Joint United Nations Programme on HIV and AIDS
(UNAIDS) launched the three 90-90-90 targets for 2020
as a major step towards eliminating the AIDS epidemic
]. Related to this, the World Health Organization
(WHO) directed that from the 1st of September 2016,
every HIV-positive person should be initiated on ART
immediately, regardless of their CD4+ T cell count [
This is commonly called the “Universal Test and Treat”
(UTT) strategy. By availing ART to HIV-infected people
and expanding prevention choices to the uninfected, 21
million AIDS-related deaths and 28 million new
infections can be prevented by 2030 [
The 90-90-90 targets recognize and utilize ART as a
life-saving treatment [
3, 5, 6
], a transmission
prevention measure [
] and a human right [
]. Target one
is successfully diagnosing 90% of all HIV-positive people.
In target two, 90% of those diagnosed will be started on
ART, and target three entails achieving viral
suppression for 90% of those on ART. While allowing for serial
10% losses at each subsequent step, the implementation
of these targets should result in 73% of all HIV-infected
individuals achieving viral suppression [
We conducted a retrospective cohort study whose
objective was to determine the impact of the UTT policy
on viral suppression and ART adherence rates among
paediatric and adolescent populations in Nairobi, Kenya.
A secondary objective was to determine the risk factors
for virologic failure and suboptimal ART adherence.
Study site and methods
Study design and population
The Lea Toto Program (LTP) is a multi-Centre
community outreach program in Nairobi, Kenya that cares
for 3500 HIV-infected children and adolescents, aged
1–18 years old. These individuals live in eight low-income
suburbs of Nairobi: Dagoretti, Dandora, Kangemi,
Kariobangi, Kawangware, Kibera, Mukuru, and
Zimmerman. The LTP Centres provide free ART drugs, CD4+
T-lymphocyte cell count monitoring, viral load testing,
clinical management of opportunistic infections, and
This was a retrospective cohort study, which included
all eligible participants going by inclusion criteria [
Two separate groups of patients were studied: the first
group was the pre-UTT group, who began ART in any of
the 6 months before the adoption of the UTT guidelines
(March to August 2016). The second group was the UTT
group: patients who began ART in September 2016, the
UTT launch month. The data of interest for both groups
was viral suppression and ART adherence rates 6 months
after ART initiation. We conducted intra- and
intergroup comparisons for the pre-UTT and UTT groups.
The retrospective cohort included 122 individuals in
the pre-UTT group and 197 in the UTT group who were
eligible going by the inclusion criteria: any children and
adolescents who began ART during one of the two
periods of interest (March-August 2016 or September 2016),
and had been followed up for at least 6 months. Among
the UTT group of 197 children, 108 children had tested
positive in the pre-UTT period, but had not qualified for
ART under the old WHO guidelines; either because they
were above 10 years of age, or their CD4+ T-cell count
was above 500 cells/µL [
]. Instead of ART, they had been
on Co-trimoxazole prophylaxis prior to September 2016.
Data collection and analysis
Data was collected from patient files, and analysed using
SPSS version 22. The WHO threshold of 1000 HIV-1
RNA copies/mL, was used to determine treatment
success or virologic failure after 6 months of treatment [
Percentages, medians, and interquartile ranges (IQRs)
were calculated. P values were determined using the
Student’s T test for normally distributed data and the
Mann–Whitney U test for skewed data. Univariate and
multivariate Cox proportional hazards regression
analysis was used to establish the effect of different variables
on virologic failure. A threshold of P < 0.05 for statistical
significance was set.
To measure adherence rates, four parameters were used
]: firstly, the pill counts done by the Clinicians during
the monthly clinic visits; secondly, the pill counts done
by the LTP community health workers (CHWs)
during unannounced home visits, with the latter serving as
verification of the former; thirdly punctuality in
attending clinic sessions, with a maximum allowance of 2 days
late; fourthly, undergoing the required blood tests on
schedule. Each parameter contributed 25% to the overall
score. Within each parameter, full compliance was scored
as 25%, partial compliance was scored as 12.5%, and zero
compliance was scored as 0%. Any overall score between
95 and 100% was defined as optimal, while a score below
95% was defined as suboptimal [
Ethical approval and consent to participate
The study was approved and cleared by the Kenya
Medical Research Institute Scientific and Ethics Review Board,
as well as the Nyumbani Medical Board (NMB); the
NMB oversees all research and treatment operations in
the LTP. All personal identifiers were removed by the
researchers upon copying the data from the patient files.
This was a retrospective cohort non-interventional study
founded on post hoc analysis of data already present in
patients’ files, and collected only for clinical indications
]. Therefore, no written informed consent was asked
of the patients.
Pre‑UTT and UTT viral suppression intra‑group comparisons
The baseline characteristics of the overall cohort, the
preUTT group and the UTT group are shown in Table 1. We
compared intra-group viral suppression rates within the
pre-UTT group based on number of parents in the home
and age group (Fig. 1). Two parent homes: 82.5%;
oneparent homes: 76.2%, and guardian homes: 76.5%. For the
comparison of two-parent homes vs. one-parent homes,
P = 0.04; for two parent homes vs. guardian homes,
P = 0.02. Age group comparisons: 2 to 5 years: 82.1%;
6–10 years: 80.0%; 11–14 years: 77.7%; 15–18 years:
73.9%. For the 2–10 years group vs. the 11–18 years
group, P = 0.01. The UTT intra-group viral suppression
rates are also shown in Fig. 1.
Pre‑UTT and UTT viral suppression inter‑group comparisons
The viral suppression rates were 79.7% pre-UTT vs.
76.6% UTT; P = 0.007. Significant reductions were
Variable at baseline
UTT Universal Test and Treat, IQR interquartile range, WHO World Health Organization, ART antiretroviral therapy, ABC abacavir, 3TC lamivudine, AZT zidovudine, TDF
tenofovir, NVP nevirapine, EFV efavirenz, LPV/r lopinavir/ritonavir
observed for children under the care of guardians (76.5%
pre-UTT vs. 74.1% UTT, P = 0.02) and the 11–14 years
group (77.7% pre-UTT vs. 75.4% UTT P = 0.03). The
15–18 years group reported the lowest viral suppression
rates among all age groups, both in the pre-UTT and
UTT periods (73.9 and 74.3%, P = 0.35) (Fig. 1).
Risk factors for below‑target viral suppression
The variables considered as potential risk factors were
adherence level, caregiver status, gender, and age group.
The treatment success sub-group was compared to the
virologic failure sub-group within the pre-UTT and UTT
groups (Table 2). These variables qualified as risk factors
when three conditions were met: the Cox proportional
hazard ratio (HR) had to be greater than 1.96, both the
lower and upper limits of the 95% confidence interval
(CI) had to be greater than 1, and the P value had to be
less than 0.05 [
]. Suboptimal adherence was the only
predictive risk factor: children with suboptimal
adherence in the pre-UTT group were 5.3 times more likely
to develop virologic failure when compared to
individuals in the same group with optimal adherence [univariate
hazard ratio (HR) = 5.32, 95% confidence interval (CI)
2.75–10.29, P value of 0.0006]. The likelihood of
treatment failure increased to 14.6 times after implementation
of UTT (univariate HR = 14.63, 95% CI 3.18–104.38, P
value of 0.003), a 2.8-fold increase (Table 2).
Risk factors for suboptimal adherence
Children with suboptimal adherence increased from 36%
(44 of 122) in the pre-UTT group to 44.7% (88 of 197) in
the UTT era (P < 0.001). The complete adherence dataset
is found in Additional file 1: Appendix S1. The variables
considered as potential risk factors for suboptimal
adherence were caregiver status, age group, and gender. For
the pre-UTT group, none of these variables met all three
conditions for predictive Cox hazard ratios [
However in the UTT period, both guardian caregiver status
and being 11–14 years old were predictive of suboptimal
adherence: UTT guardian homes: univariate HR = 2.27,
95% CI 1.12–6.72, P = 0.041); children aged 11–14 years:
Univariate HR = 2.16, 95% CI 1.03–5.73, P = 0.018
Our study shows a significant fall in viral suppression
rates among an HIV-infected paediatric and adolescent
population in Nairobi on the transition from the
previous WHO treatment guidelines to the UTT policy. This
was directly correlated with a fall in adherence rates.
Children with suboptimal adherence were 5.3 times more
likely to fail ART pre-UTT, and 14.6 times more likely to
fail in the UTT period, representing a 2.8-fold increase
in the rate of virologic failure. Previous studies have
proved the direct link between poor adherence to ART
and increased risk of treatment failure [
]. This is
even more concerning in the light of UTT ART scale-up
for the 22 million people who were not on ART before
September 2016 [
]. Prior to UTT implementation,
patients received thorough ART counselling before being
initiated on treatment. In the “test and start” UTT
dispensation, patients receive ART counselling concurrent
with treatment . Therefore, patients in the UTT period
are more likely to have a poorer understanding of the
importance of taking their medicine at specific times of
the day, every day. This may result in lower viral
suppression and adherence rates.
Adolescent populations have been shown in previous
studies to be the least adherent to ART. A recent national
survey in Kenya found higher levels of suboptimal
adherence among persons aged 15–29 [
]. This is in
conformity with data from across sub-Saharan Africa which
found that ART adherence by pre-teenagers and
teenagers is influenced by disclosure [
status , treatment fatigue [
], fear of stigmatization
by peers [
], and vigilance by the primary caregivers [
]. In the UTT period, many more pre-teenagers and
teenagers who had not qualified for ART under the old
WHO guidelines were initiated on ART. This transition
may have amplified the above factors.
Children under the care of guardians in our study
showed decreased adherence rates and a
corresponding increase in treatment failure upon the transition to
UTT. This is in line with previous studies which show
that HIV-infected orphans being cared for by
guardians are at greater risk of poor adherence and treatment
]. In Zambia, higher adherence levels
were observed where the child’s mother was the
primary caregiver and improved further if the child had
multiple caregivers [
]. In Rwanda, double orphans
under the care of guardians or siblings had the lowest
adherence rates to ART [
]. It has been observed that
non-biological caregivers are less motivated to
monitor their children’s ART ingestion as compared to
biological parents [
]. This may be more exacerbated in
an urban slum setting, such as the LTP cohort in our
study, where individuals live on less than 5 US dollars
Our observations show an increase in suboptimal
adherence and a corresponding increase in treatment
failure rates among a paediatric and adolescent
population in Nairobi, Kenya at the onset of implementation
of the WHO UTT Program in September 2016. Though
there is overwhelming scientific evidence that
starting ART as soon as possible has profound advantages,
there is need to develop extra strategies to vigorously
support adherence especially among orphaned children
The main limitation of the study was that it was a
retrospective cohort study with only 6 months of follow
up. A time series analysis over 12–24 months would
likely reveal more. Secondly, although the total number
of children in the LTP is large (3500), only about 20–30
children were initiated on ART monthly in the pre-UTT
period. Thus, we were able to include only 122 children in
the pre-UTT group and 197 in the UTT group who met
inclusion criteria. A larger sample size would have
powered the study.
Additional file 1: Appendix S1. A PDF file with a detailed dataset for
intra- and inter-group Adherence rates, before the adoption of Universal
test and Treat (UTT) and after UTT adoption; found at https://doi.
UTT: Universal Test and Treat; UNAIDS: United Nations Programme on HIV
and AIDS; WHO: World Health Organization; LTP: Lea Toto Programme; AIDS:
acquired immune deficiency syndrome; HIV: human immunodeficiency virus;
ART: antiretroviral therapy; CHWs: community health workers; HR: hazard ratio;
CI: confidence interval; IQR: interquartile range; ABC: abacavir; 3TC:
lamivudine; AZT: zidovudine; TDF: tenofovir; NVP: nevirapine; EFV: efavirenz; LPV/r:
ES and JK conceptualized the study. JK and RM participated in the
implementation of the study. EM and JK performed the data analysis; JK drafted
the manuscript with ES providing substantial input. EM and FW edited the
manuscript. All authors read and approved the final manuscript.
1 Jomo Kenyatta University of Agriculture and Technology (JKUAT), Juja, Kenya.
2 Department of Paediatrics, University of Nairobi, Nairobi, Kenya. 3 Kenya
Medical Research Institute (KEMRI), Mbagathi Road, Nairobi, Kenya. 4 Centre
for Public Health Research (CPHR), Kenya Medical Research Institute (KEMRI),
Mbagathi Road, Nairobi, Kenya. 5 Department of Biochemistry, Jomo Kenyatta
University of Agriculture and Technology (JKUAT), Juja, Kenya.
We thank Sister Mary Owens, Sister Annie Panikulam, the Children of God
Relief Institute (COGRI) and Nyumbani Medical Board (NMB) for granting us
access to the LTP records. We also thank the Director, Kenya Medical Research
Institute (KEMRI) for granting permission for the study and the publication of
The authors declare that they have no competing interests.
Availability of data and materials
Additional data supporting the conclusions of this article is presented in
Additional file 1: Appendix S1.
Consent to publish
Ethics approval and consent to participate
The study was approved and cleared by the Kenya Medical Research Institute
Scientific and Ethics Review Board under protocol number KEMRI/SSC/2500.
Approval was also granted by the Nyumbani Medical Board, which oversees
all Research and Treatment operations in the Lea Toto Program. This was a
retrospective cohort study and it was non-interventional. All personal identifiers
were removed by the researchers upon copying the data from the patient
files. Since this was a retrospective non-interventional study founded on post
hoc analysis of data already present in patients’ files, and collected only for
clinical indications, no written informed consent was asked of the patients.
This work was funded by the Kenya National Commission for Science,
Technology, and Innovation (NaCoSTI), grant NCSTI/ST&I/RCD/4thCallPhD/176 to JK;
the Manitoba Medical Service Foundation (MMSF) in Canada to ES; and the
HIV Research Trust in the United Kingdom, Grant HIVRT14-106 to JK.
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
1. World Health Organization (WHO). Core epidemiological HIV/AIDS estimates . 2015 . http://www.who.int/hiv/data/en/. Accessed 24 Mar 2017 .
2. The Joint United Nations Programme on HIV/AIDS (UNAIDS) . HIV and AIDS estimates for Kenya . 2015 . http://www.unaids.org/en/regionscou ntries/countries/kenya. Accessed 24 Mar 2017 .
3. The Joint United Nations Programme on HIV/AIDS (UNAIDS) . 90 -90 -90 An ambitious treatment target to help end the AIDS epidemic; JC2684 . 2014 . http://www.unaids.org/sites/default/files/media_asset/90-90- 90_en_0.pdf. Accessed 24 Mar 2017 .
4. World Health Organization (WHO). Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection . 2016 . http:// www.who.int/hiv/data/en/. Accessed 24 Mar 2017 .
5. Levi J , Raymond A , Pozniak A , Vernazza P , Kohler P , Hill A . Can the UNAIDS 90-90-90 target be achieved? BMJ Glob Health . 2016 . https://doi. org/10.1136/bmjgh-2015-000010.
6. Lundgren JD , Babiker AG , Gordin F , The INSIGHT START Study Group, et al. Initiation of antiretroviral therapy in early asymptomatic HIV infection . N Engl J Med . 2015 ; 373 : 795 - 807 .
7. Cohen MS , Chen YQ , McCauley M , et al. Prevention of HIV-1 infection with early antiretroviral therapy . N Engl J Med . 2011 ; 365 : 493 - 505 .
8. Gruskin S , Tarantola D. Universal access to HIV prevention, treatment and care: assessing the inclusion of human rights in international and national strategic plans . AIDS . 2008 ; 22 ( Suppl 2 ): S123 - 32 .
9. Charan J , Biswas T. How to calculate sample size for different study designs in medical research? Indian J Psychol Med . 2013 ; 35 ( 2 ): 121 - 6 .
10. Focà E , Odolini S , Sulis G , et al. Clinical and immunological outcomes according to adherence to first-line HAART in a urban and rural cohort of HIV-infected patients in Burkina Faso, West Africa . BMC Infect Dis . 2014 ; 14 : 153 .
11. Kamya MR , Mayanja-Kizza H , Kambugu A , Bakeera-Kitaka S , Semitala F , Mwebaze-Songa P , Castelnuovo B , Schaefer P , Spacek LA , Gasasira AF , Katabira E , Colebunders R , Quinn TC , Ronald A , Thomas DL , Kekitiinwa A . Predictors of long-term viral failure among Ugandan children and adults treated with antiretroviral therapy . J Acquir Immune Defic Syndr . 2007 ; 46 ( 2 ): 187 - 93 .
12. Ware NC , Idoko J , Kaaya S , Biraro IA , Wyatt MA , Agbaji O , Chalamilla G , Bangsberg DR . Explaining adherence success in sub-Saharan Africa: an ethnographic study . PLoS Med . 2009 ; 6 ( 1 ): e11 .
13. Haberer JE , Cook A , Walker AS , Ngambi M , Ferrier A , Mulenga V , Kityo C , Thomason M , Kabamba D , Chintu C , Gibb DM , Bangsberg DR . Excellent adherence to antiretrovirals in HIV+ Zambian children is compromised by disrupted routine, HIV nondisclosure, and paradoxical income effects . PLoS ONE . 2011 ; 6 ( 4 ): e18505 .
14. Walters S. What is a Cox model? 2009 . http://www.whatisseries.co. uk/ what-is-a-cox-model/ . Accessed 25 Mar 2015 .
15. Bigna JR , Plottel CS , Koulla-Shiro S. Challenges in initiating antiretroviral therapy for all HIV-infected people regardless of CD4 cell count . Infect Dis Poverty . 2016 ; 5 : 85 . https://doi.org/10.1186/s40249-016-0179-9.
16. Lima VD , Reuter A , Harrigan PR , Lourenco L , Chau W , Hull M , Mackenzie L , Guillemi S , Hogg RS , Barrios R , et al. Initiation of antiretroviral therapy at high CD4+ cell counts is associated with positive treatment outcomes . AIDS . 2015 ; 29 ( 14 ): 1871 - 82 .
17. Juma JM , Tiberio JK , Abuya MI , Kilama BK , Somi GR , Sambu V , Banda R , Jullu BS , Ramadhani AA . Monitoring prevention or emergence of HIV drug resistance: results of a population-based foundational survey of early warning indicators in mainland Tanzania . BMC Infect Dis . 2014 ; 14 : 196 .
18. Maqutu D , Zewotir T , North D , Naidoo K , Grobler A . Factors affecting firstmonth adherence to antiretroviral therapy among HIV-positive adults in South Africa . Afr J AIDS Res . 2010 ; 9 ( 2 ): 117 - 24 .
19. Maqutu D , Zewotir T. Optimal HAART adherence over time and time interval between successive visits: their association and determinants . AIDS Care . 2011 ; 23 ( 11 ): 1417 - 24 .
20. Mukui IN , Ng'ang'a L , Williamson J , Wamicwe JN , Vakil S , Katana A , Kim AA . Rates and predictors of non-adherence to antiretroviral therapy among HIV-positive individuals in Kenya: results from the second Kenya AIDS indicator survey , 2012 . PLoS ONE . 2016 ; 11 ( 12 ):e0167465. https://doi. org/10.1371/journal.pone. 0167465 .
21. Nachega J , Hislop M , Nguyen H , Dowdy D , Chaisson R , Regensberg L , Cotton M , Maartens G. Antiretroviral therapy adherence, virologic and immunologic outcomes in adolescents compared with adults in Southern Africa . J Acquir Immune Defic Syndr . 2009 ; 51 ( 1 ): 65 - 71 . https://doi. org/10.1097/qai.0b013e31819902e.
22. Aderomilehin O , Hanciles-Amu A , Ozoya OO . Perspectives and practice of HIV disclosure to children and adolescents by health-care providers and caregivers in sub-Saharan Africa: a systematic review . Front Public Health . 2016 ; 4 : 166 . https://doi.org/10.3389/fpubh. 2016 . 00166 .
23. Bermudez LG , Jennings L , Ssewamala FM , Nabunya P , Mellins C , McKay MM . Equity in adherence to antiretroviral therapy among economically vulnerable adolescents living with HIV in Uganda . AIDS Care . 2016 ; 28 ( sup2 ): 83 - 91 .
24. Haberer JE , Mellins C . Paediatric adherence to HIV antiretroviral therapy . Curr HIV/AIDS Rep . 2009 ; 6 ( 4 ): 194 - 200 .
25. Wiener L , Mellins CA , Marhefka S , Battles HB . Disclosure of an HIV diagnosis to children: history, current research, and future directions . J Dev Behav Paediatr . 2007 ; 28 : 155 - 66 .
26. Thurman TR , Jarabi B , Rice J . Caring for the caregiver: evaluation of support groups for guardians of orphans and vulnerable children in Kenya . AIDS Care . 2012 ; 24 ( 7 ): 811 - 9 . https://doi.org/10.1080/09540 121. 2011 . 644229 .
27. Mokgatle MM , Madiba S. The burden of disease on HIV-infected orphaned and non-orphaned children accessing primary health facilities in a rural district with poor resources in South Africa: a cross-sectional survey of primary caregivers of HIV-infected children aged 5-18 years . Infect Dis Poverty . 2015 ; 4 : 18 . https://doi.org/10.1186/s40249-015-0049-x (eCollection 2015 ).
28. Popoola T , Mchunu G . HIV orphanhood research and the representation of older orphans in Sub-Saharan Africa: a literature review . J Assoc Nurses AIDS Care . 2016 ; 27 ( 1 ): 57 - 65 . https://doi.org/10.1016/j.jana. 2015 . 09 .002.
29. Kikuchi K , Poudel KC , Muganda J , Sato T , Mutabazi V , Muhayimpundu R , Majyambere A , Nyonsenga SP , Sase E , Jimba M. What makes orphans in Kigali, Rwanda, non-adherent to antiretroviral therapy? Perspectives of their caregivers . J Int AIDS Soc . 2014 ; 17 ( 1 ): 19310 .
30. Kikuchi K , Poudel KC , Muganda J , Sato T , Majyambere A , Otsuka K , Sato T , Mutabazi V , Nyonsenga SP , Muhayimpundu R , Jimba M , Yasuoka J . High risk of ART non-adherence and delay of ART initiation among HIV-positive double orphans in Kigali, Rwanda . PLoS One . 2012 ; 7 ( 7 ):e1998. https://doi. org/10.1371/journal.pone. 0041998 .