Open access Original research

Viral load undetectable state and

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
predictors among children and
adolescents living with HIV in South
Gondar, Ethiopia, 2023: an 8-­year
retrospective cohort study

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Ermias Sisay Chanie ‍ ‍,1 Dejen Getaneh Feleke ‍ ‍,2 Tigabu Desie Emiru ‍ ‍,2
Abere Gebru Abuhay,3 Habtamu Shimels Hailemeskel ‍ ‍,4
Astewle Andargie Baye ‍ ‍,5 Berihun Bantie ‍ ‍,5 Abraham Tsedalu Amare ‍ ‍,5
Tadila Dires Nega,5 Denekew Tenaw Anley ‍ ‍,6 Anteneh Mengist Dessie ‍ ‍,6
Sintayehu Asnakew ‍ ‍,7 Natnael Moges ‍ ‍,2 Wubet Muluye Kassahun,8
Mitiku Arayaw Diress,8 Abeje Kumlachew Asfaw,8 Dawit Alemu Ferede,8
Endalk Birrie Wondifraw,9 Melkamu Senbeta Jimma,10
Fisha Alebel GebreEyesus ‍ ‍,11 Sewnet Sisay Chanie12

To cite: Chanie ES, Feleke DG, ABSTRACT

Emiru TD, et al. Viral load Objectives The objective of this study is to examine the
STRENGTHS AND LIMITATIONS OF THIS STUDY
undetectable state and zonal-­wide load undetectable state and predictors among ⇒ This study attempts to determine the rate of un-
predictors among children detectable viral load in children and adolescents
children and adolescents living with HIV at South Gondar
and adolescents living with starting antiretroviral therapy for the first time in
HIV in South Gondar, Ethiopia, health institutions, 2023.
Design A retrospective cohort study. Ethiopia.
2023: an 8-­year retrospective
cohort study. BMJ Open Setting South Gondar Health Institutions, Northwest, ⇒ This study was conducted in multiple health institu-
2024;14:e083206. doi:10.1136/ Ethiopia. tions, encompassing a long-­term follow-­up period.
bmjopen-2023-083206 Participants We recruited 430 children and adolescents ⇒ Incomplete or missing data due to its retrospective
living with HIV who had a follow-­up at the antiretroviral nature may influence the outcomes.
► Prepublication history
therapy (ART) clinic at health institutions from 1 June 2016 ⇒ An 8-­year study duration might not encompass the
and additional supplemental
to 30 April 2023. latest developments in HIV treatment guidelines or
material for this paper are
available online. To view these Outcome measures The primary outcome measure of interventions that could affect the results.
files, please visit the journal ⇒ This study did not investigate certain possible fac-
this study was the viral load undetectable state estimation.
online (https://doi.org/10.1136/​ tors such as monthly income, substance use and
Moreover, the study assessed the median time viral
bmjopen-2023-083206). family size, which might impact reaching an unde-
load undetectable state and its predictors by the Cox-­
tectable viral load.
Received 20 December 2023 proportional hazard model. Data were entered into Epi-­
Accepted 20 September 2024 data V.4.2 and exported to STATA V.17 statistical software
for analysis. had nutritional status ≥−2 Z score increased a viral load
Results The mean follow-­up period was 8.5 (95% CI undetected state by 2.3 times as compared with children
8.1 to 8.9)±4.4 SD months, overall yielding 9151 person-­ and adolescents with nutritional status <−2 Z score
month observations. At the end of the follow-­up, 369 (AHR 2.3 (95% CI 1.3, 4.0)). Furthermore, children and
(85.8%, 95% CI 82.6% to 88.8%) of the children and adolescents who had no treatment failure increased a
adolescents achieved the viral load undetected state. viral load undetected state by 2.1 times as compared with
Moreover, the overall median survival time to develop viral children and adolescents who had treatment failure (AHR
© Author(s) (or their load undetectable state was found to be 6 months. 3.0 (95% CI 1.1, 4.0).
employer(s)) 2024. Re-­use The viral load undetected state in children and adolescents Conclusion The lower proportion of individuals
permitted under CC BY-­NC. No who have a cluster of differentiation 4 (CD4) count above achieving an undetectable viral load compared with the
commercial re-­use. See rights the threshold level was 2.8 times higher than those in target set by the WHO strategic plan (95-­95-­95 target
and permissions. Published by children and adolescents with a CD4 count lower than
BMJ.
by 2025), highlights the need for targeted interventions
the threshold level (adjusted HR (AHR) 2.8 (95% CI 1.5, and improved HIV management strategies. Factors such
For numbered affiliations see 5.3)). Likewise, the viral load undetected state in children as CD4 count, ART adherence, nutritional status and
end of article.
and adolescents who have a good level of ART adherence treatment failure play a significant role in achieving viral
Correspondence to was 2.0 times higher than those children and adolescents load undetectable state. Healthcare providers better
Ermias Sisay Chanie; with a fair/poor level of ART adherence (AHR 2.0 (95% prioritise these factors through comprehensive care
​ermisis1888@​gmail.c​ om CI 1.1, 3.9)). Moreover, children and adolescents who and support. The study emphasises the importance

Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206 1

 Open access

of aligning efforts with the WHO strategic plan to meet targets and METHODS AND MATERIALS
improve treatment outcomes. Therefore, stakeholders better address Study setting

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
the identified predictors to enhance care and support for this vulnerable The South Gondar health institutions are located in
population. the South Gondar zone, which is located in the Amhara
region in the northwest part of Ethiopia. Besides, Debre
Tabor town is the capital city of the south Gondar zone,
BACKGROUND which is 103 km from Bahir Dar (the capital city of the
In 2019, globally, 1.7 million children were affected by Amhara region) and 665 km from Addis Ababa.
HIV.1 In Ethiopia, in the same year, 44 229 children were There are nine hospitals and eight health centres that
HIV-­positive, and 2055 children died due to AIDS.2 provide ART in the south Gondar zone. ART case team

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Significant advancements have been achieved in has been in the hospitals including physicians, nurses,
expanding access to antiretroviral therapy (ART) in low-­ pharmacists and laboratory technicians who took compre-
income and middle-­income countries. Presently, around hensive ART training. ART case team tries to improve the
the world, approximately 21.7 million individuals are quality of life among children and adolescents living with
receiving ART, with 15 million of them located in sub-­ HIV including voluntary testing and counselling, psycho-
Saharan Africa.3 4 For HIV treatment to be effective in logical support and prevention of HIV transmission from
restoring immune functionality, ART should suppress mother to child. All children and adolescents who have
plasma HIV RNA to undetectable states or levels (<50 been followed up in the health institutions have had viral
copies/mL) and improve the quality of life.5 Hence, load monitoring annually since 2008 E.C.
regular viral load testing is widely recognised as the most
effective method for monitoring ART in individuals living Study design and study participants
with HIV.6 7 A retrospective follow-­up study was conducted from 1
Although being virally suppressed and undetectable June 2016 to 30 April 2023, at South Gondar Health Insti-
is the aim of HIV treatment, there are some reasons why tutions, in 2023. The study included a total of eight hospi-
a person may not be able to achieve an undetectable tals and seven health centres that provide ART services
viral load. These can include factors outside of some- in the South Gondar zone. Four health centres, namely
one’s control, including, drug interactions, side effects, Worta Health Center, Nefas Mewucha Health Center,
Mekane Eyesus Health Center and Addis Zemen Health
adherence challenges and other clinical-­related charac-
Center, were selected randomly for the study. Addition-
teristics.8 9
ally, five hospitals, including Debre Tabor Compressive
The United Nations Programme on HIV/AIDS
Specialized Hospital, Nefas Mewucha Primary Hospital,
(UNAIDS) has set global targets called the 95-­ 95-­
95
Mekane Eyesus Primary Hospital and Addis Zemen
targets to eliminate AIDS as a public health threat by
Primary Hospital, were also randomly selected as part of
2030. The third target aims to achieve viral load suppres-
the study.
sion in at least 95% of individuals living with HIV and
The sample size was estimated using the formula:
receiving ART.10 However, numerous developing coun-
n=(za/2)2×p(1−p)/d2 where p was assumed to be 50%
tries, including Ethiopia, have not yet achieved these
due to the absence of prior studies. n=(1.96)2×0.5
targets. It is essential to compare viral rebound and viral (0.5)/0.052=384. After accounting for a 10% increase to
suppression rates based on different clinical attributes, address potential non-­responses or incomplete data, the
and identifying factors that contribute to increased viral final sample size was adjusted to approximately 430.
load is crucial for adjusting programme-­level strategies. In the ART clinics of the selected hospitals, all children
These interventions play a vital role in achieving the and adolescents living with HIV who initiated ART were
global target goals set by UNAIDS and addressing the identified. The investigator assigned registration numbers
challenges faced by developing countries in attaining the to these individuals in chronological order, based on
desired levels of viral suppression. their ART start dates between 1 June 2016 and 30 April
The viral load testing is a powerful tool, recommended 2023. After a thorough review of medical charts and the
by the WHO for improving the long-­term success of ART ART registration logbook, the investigator selected a total
for HIV. However, less than 20% of ART patients in Africa of 430 samples that met the predetermined inclusion
receive regular viral load monitoring.11 criteria.
Although various studies have examined survival rates,
ART adherence and other aspects among children and Patient and public involvement
adolescents living with HIV, the factors associated with There was no direct public or patient involvement in the
achieving an undetectable viral load have not yet been design and implementation of this study.
investigated in Ethiopia, particularly in the study area.
Hence, this study aims to examine the viral load undetect- Data sources/measurement
able state estimation and its predictors among children A standard checklist containing study variables has been
and adolescents living with HIV at South Gondar health developed from the children and adolescents registry card
institutions, Northwest, Ethiopia, 2023. which was developed by the Ethiopian Federal Ministry of

2 Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206

 Open access

Table 1 Sociodemographic characteristics among children

Censored
and adolescents living with HIV at South Gondar health Children and adolescents were classified as censored in

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
institutions, Ethiopia, 2023 the study if they were lost to follow-­up or transferred to
another healthcare facility before achieving the event of
Frequency
Variables (n=430) % interest (ie, reaching an undetectable viral load state).
Additionally, children and adolescents who remained
Age (year) 0–10 157 36.5 alive until the end of the study period, which was 30 April
10–18 273 63.5 2023, were also considered censored.
Sex Male 193 44.9 CD4 count below the threshold for severe immuno-
Female 237 55.1 deficiency was classified according to age (for infants

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
CD4<1500/mm3, 12–35 months <750/mm3, 36–59
Residence Urban 255 59.3
months <350/mm3 and ≥5 years <200/mm3).12
Rural 175 40.7 Adherence to ART was categorised based on the
Functional Working 376 87.4 percentage of drug dosage taken in relation to the total
status Ambulatory 39 9.1 monthly doses of ART drugs. Adherence levels were clas-
Bedridden 15 3.5 sified as good if the percentage exceeded 95%, fair if it
ranged between 85% and 94%, and poor if it fell below
Disclosure of Disclosed 369 85.8
HIV status 85%.13
Non-­disclosed 61 14.2
Marital status Single 91 21.2 Statistical methods
of parents Married 238 55.3 The data collected was inputted into Epi Data V.4.4.2
Divorced 61 14.2
software and then exported to Stata V.17 for analysis.
Descriptive statistics such as mean, median and SD were
Widowed 40 9.3
used to summarise continuous variables related to the
Religion Orthodox 277 64.4 study participant’s characteristics. Categorical data were
status of Muslim 107 24.9 summarised using percentages, frequency tables and
parents
Protestant 27 6.3 visual representations such as box plots.
The Akaike information criteria and Bayesian informa-
Catholic 19 4.4
tion criteria were employed to compare different models
Educational Cannot read and write 55 12.8
and determine the most efficient one for the analysis.
status
Able to read and write 231 53.7 The Kaplan-­Meier curve was employed to estimate the
Primary school and 144 33.5 cumulative probability of reaching an undetectable viral
above load state, as well as the median survival time. To iden-
tify predictors of viral load undetectable state, the Cox
proportional hazards model was used. Assumptions of the
Health.12 Data were collected from children and adoles- model, including proportionality assumptions and good-
cents ART record cards and registers from 1 June 2016 to ness of fit, were assessed. Both bivariate and multivariable
30 April 2023. The data were collected in terms of patient analyses were conducted, and a 95% CI was computed for
sociodemographic characteristics, clinical information the HR. Variables that had a p<0.05 in the multivariable
and treatment-­related information. In the data cleaning analysis were considered statistically significant and asso-
process, missing variables were recollected and confirmed ciated with viral load undetectable state.
using paper-­ based patient ART records and registers.
WHO Anthro-­Plus software was used to assess the nutri-
tional status of the children and adolescents. The data RESULTS
were collected by nine BSc health professionals and two Sociodemographic characteristics of the study participants
MSc degrees for supervision who had taken comprehen- The study enrolled a total of 430 children and adoles-
sive HIV care training. Data cleaning and double data cents living with HIV. Among these participants, 193
entry were carried out to check for any inconsistencies. (44.9%) were identified as male. About 255 (59.3%) of
The quality of the data was assured by giving 2 days of the study participants resided in urban areas. Within
training for data collectors and supervisors. A pretest was the group of children and adolescents included in the
conducted on 10% of the sample size. study, 273 (63.5%) fell into the age range of 11–18 years.
When considering their functional status, a significant
Variables proportion—379 (90.0%)—were classified as being in
Viral load undetectable state (event) the working category. Furthermore, it was observed that a
An undetectable viral load is where ART has reduced majority of the participants—369 (85.8%)—had disclosed
your HIV to such small quantities that it can no longer their HIV status.
be detected by standard blood tests usually <50 copies/ The parental status of the children and adolescents
ML.8 9 revealed that a substantial portion of parents, 378

Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206 3

Open access

Table 2 Clinical and treatment-­related characteristics among children and adolescents living with HIV at South Gondar health
institutions, Ethiopia, 2023

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
Variables Frequency (n=430) %
CD4 count Below threshold 66 15.3
Above threshold 364 84.7
ART adherence Good 357 83.0
Fair/poor 73 17.0
Haemoglobin status <10 g/L 79 18.4

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
≥10 g/L 351 81.6
Nutritional status <−2 Z score 84 19.5
≥−2 Z score 346 80.5
History of TB Yes 44 10.2
No 386 89.8
WHO staging Stages I and II 358 83.3
Stages III and IV 72 16.7
CPT and IPT prophylaxis Yes 406 94.4
No 24 5.6
Treatment failure Yes 68 15.8
No 362 84.2
Opportunistic infections Yes 146 34.0
No 284 66.0
Regimen of ART TDF+3TC+DTG and ABC+3TC+DTG 294 68.4
AZT+3TC+DTG and ABC+3TC+EFV 69 16.0
RAL+ABC+3 TC, ABC+3TC+EFV and TDF+3TC+EFV 41 9.5
AZT+3TC+LPV/r and TDF+3TC+ATV/r 19 4.4
Other 7 1.6
Duration of ART in month <60 months 163 38.0
≥60 months 267 62.0
Duration of follow-­up in month ≤6 months 262 60.8
>6 months 168 39.2
Patient status A live 391 90.9
Death 39 9.1
Status of viral load undeletable Yes 369 85.8
state No 61 14.2

TDF+3TC+DTG and ABC+3TC+DTG: tenofovir disoproxil fumarate (TDF), lamivudine (3TC) and dolutegravir (DTG) or abacavir (ABC),
lamivudine (3TC) and dolutegravir (DTG). AZT+3TC+DTG and ABC+3TC+EFV: zidovudine (AZT), lamivudine (3TC) and dolutegravir (DTG)
or abacavir (ABC), lamivudine (3TC), and efavirenz (EFV). RAL+ABC+3TC, ABC+3TC+EFV and TDF+3TC+EFV: raltegravir (RAL), abacavir
(ABC) and lamivudine (3TC), abacavir (ABC), lamivudine (3TC) and efavirenz (EFV), or tenofovir disoproxil fumarate (TDF), lamivudine
(3TC) and efavirenz (EFV). AZT+3TC+LPV/r and TDF+3TC+ATV/r: zidovudine (AZT), lamivudine (3TC) and lopinavir/ritonavir (LPV/r), or
tenofovir disoproxil fumarate (TDF), lamivudine (3TC) and atazanavir/ritonavir (ATV/r).
ART, Antiretroviral Therapy; CD4, Cluster of Differentiation 4; CPT, Cotrimoxazole Preventive Therapy; IPT, Isoniazid Preventive Therapy;
TB, Tuberculosis.

(87.9%) were orthodox, 254 (59.1%) were married under WHO clinical stages III and IV and 79 (18.4%) had a
and 235 (54.7%) were able to read and write (table 1). haemoglobin level below 10 g/L. Moreover, 84 participants
(19.5%) had a nutritional status below −2 Z score.
Clinical and treatment-related characteristics of the study
Out of the total participants, 146 (34.0%) experi-
participants
During the course of the study, a considerable proportion of enced opportunistic infections, 73 (17%) exhibited
children and adolescents living with HIV displayed certain fair or poor adherence to ART, 68 (15.8%) had treat-
health indicators. Specifically, 66 participants (15.3%) had ment failure and 44 (10.2%) were diagnosed with
a CD4 count below the threshold level, 72 (16.7%) fell tuberculosis.

4 Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206

 Open access

Kaplan-Meier curve and median survival time of viral load

undetected state

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
The average duration of follow-­ up was 8.5 months
(95% CI 8.1 to 8.9)±4.4 SD months. This resulted in a total
of 9151 observations over the course of the study. By the
end of the follow-­up period, 369 children and adolescents
(85.8%, 95% CI 82.6% to 88.8%) achieved an undetect-
able viral load. Furthermore, the median survival time for
children and adolescents to reach an undetectable viral
load was found to be 6 months (figure 2).

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Bivariable and multivariable analysis of viral load
Figure 1 The box plot of viral load level among children
undetectable state
and adolescents living with HIV at South Gondar health In the Cox proportional hazard model, the bivariable
institutions, Ethiopia. analysis identified factors associated with the develop-
ment of an undetectable viral load, with a p value of less
than 0.25. These factors included age, residence, educa-
The majority of participants—406 (94.4%)—received tional status, disclosure status, CD4 count, level of ART
both cotrimoxazole preventive therapy (CPT) and isoni- adherence, haemoglobin status, nutritional status, WHO
azid preventive therapy (IPT). Additionally, 294 children staging, CPT and IPT prophylaxis, treatment failure,
and adolescents (68.4%) were prescribed TDF+3TC+DTG opportunistic infections and duration of follow-­ up in
(4i) and ABC+3TC+DTG (4j) as their ART regimen. months. These factors were then included in the multi-
Regarding the duration of follow-­up, 262 participants variable analysis.
In the multivariable models, the following factors were
(60.8%) had a follow-­up period of less than or equal to
found to be associated with achieving an undetectable
6 months, while 267 (62.1%) had been on ART for 60
viral load during the follow-­up period: CD4 count above
months or more.
the threshold level, good level of ART adherence, nutri-
Throughout the follow-­ up period, 369 participants
tional status ≥−2 Z score and WHO stages I and II. Chil-
(85.8%) achieved an undetectable viral load and 391 dren without treatment failure were also more likely to
(90.9%) survived (table 2). achieve an undetectable viral load. Specifically, children
and adolescents with a CD4 count above the threshold
The level of viral load of the study participants over time level had a 2.8 times higher chance of achieving an unde-
The level of viral load of the study participants was esti- tectable viral load compared with those with a CD4 count
mated from 2016 to 2023. Over the follow-­up period, below the threshold level (adjusted HR, AHR 2.8, 95% CI
the level of viral load in 2016 was estimated to be 1240 1.5, 5.3). Similarly, children and adolescents with a good
copies/mL, while in 2023, it had decreased significantly level of ART adherence were 2.0 times more likely to
and was found 103 copies/mL (figure 1). achieve an undetectable viral load compared with those
with fair/poor adherence (AHR 2.0, 95% CI 1.1, 3.9).
Children and adolescents with nutritional status ≥−2 Z
score had a 2.3 times higher chance of achieving an unde-
tectable viral load compared with those with nutritional
status <−2 Z score (AHR 2.3, 95% CI 1.3, 4.0). Further-
more, children and adolescents without treatment failure
had a 2.1 times higher chance of achieving an undetect-
able viral load compared with those with treatment failure
(AHR 2.1, 95% CI 1.1, 4.0) (table 3 and figure 3A–D).

DISCUSSION
This study aimed to analyse the undetectable viral load
and identify factors among children and adolescents
living with HIV at South Gondar health institutions in
Northwest Ethiopia. The study revealed that the level of
undetectable viral load in South Gondar health institu-
tions was found to be 85.8% (95% CI 82.6% to 88.8%).
Figure 2 Kaplan-­Meier curve of viral load undetected state Over time, there was a significant decrease in viral load
among children and adolescents living with HIV at South levels, with the level of 1240 copies/mL in 2016 and 103
Gondar health institutions.19 copies/mL in 2023. Additionally, the median survival

Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206 5

Open access

Table 3 Bivariable and multivariable analysis of viral load undetectable among children and adolescents living with HIV at
South Gondar health institutions, Ethiopia, 2023

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
Viral load undetectable HR (95% CI)
Yes No
Characteristics −369 −61 CHR AHR P value
Age (year) 0–10 125 32 Ref Ref
10–18 244 29 1.2 (1.0 to 1.5) 1.1 (0.8 to 1.3) 0.9
Sex Male 167 26 Ref –

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Female 202 35 1.0 (0.8 to 1.2) –
Residence Urban 228 27 1.2 (1.0 to 1.4) 0.9 (0.8 to 1.2) 0.3
Rural 141 34 Ref Ref
Functional status Working 331 45 1.0 (0.6 to 1.8) –
Ambulatory and 30 9 1.0 (0.5 to 1.5) –
Bedridden 8 7 Ref –
Disclosure of HIV Disclosed 334 35 2.7 (1.8 to 3.8) 0.8 (0.5 to 1.2) 0.43
status Non-­disclosed 35 26 Ref Ref
Marital status of Single 80 11 Ref –
parents Married 199 39 0.9 (0.7 to 1.3) –
Divorced 56 5 1.0 (0.7 to 1.5)
Widowed 34 6 1.5 (1.0 to 2.1) –
CD4 count Below threshold 17 49 Ref Ref
Above threshold 352 12 6.8 (4.1 to 11.1) 2.8 (1.5 to 5.3) 0.001**
ART adherence Good 350 7 9.9 (5.8 to 17.8) 2.0 (1.1 to 3.9) 0.035*
Fair/poor 14 59 Ref Ref
Haemoglobin status <10 g/L 32 47 Ref Ref
≥10 g/L 337 14 3.8 (2.6 to 5.5) 1.3 (0.9 to 2.0) 0.136
Nutritional status <−2 Z score 29 55 Ref Ref
≥−2 Z score 340 6 7.6 (4.5 to 12.0) 2.3 (1.3 to 4.0) 0.004**
History of TB Yes 36 8 Ref –
No 333 53 1.0 (0.7 to 1.4) –
WHO staging Stages I and II 341 17 3.8 (2.6 to 5.7) 1.4 (0.9 to 2.2) 0.09
Stages III and IV 28 44 Ref Ref
CPT and IPT Yes 350 56 0.9 (0.6 to 1.3) –
prophylaxis No 19 5 Ref –
Treatment failure Yes 15 53 Ref Ref
No 354 8 8.1 (4.8 to 13.8) 2.1 (1.1 to 4.0) 0.001**
Opportunistic Yes 110 36 Ref Ref
infections No 259 25 1.3 (1.1 to 1.6) 0.8 (0.7 to 1.1) 0.19
Regimen of ART TDF+3TC+DTG and 256 38 Ref –
ABC+3TC+DTG
AZT+3TC+DTG and 62 7 1.2 (0.9 to 1.5) –
ABC+3TC+EFV
RAL+ABC+3 TC, ABC+3TC+EFV 32 9 1.1 (0.8 to 1.5) –
and TDF+3TC+EFV
AZT+3TC+LPV/r and 15 4 0.9 (0.6 to 1.7) –
TDF+3TC+ATV/r
Other 4 3 1.5 (0.7 to 3.2) –
Duration of ART in <60 months 140 23 Ref –
month ≥60 months 229 38 1.0 (0.8 to 1.2) –
Continued

6 Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206

 Open access

Table 3 Continued

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
Viral load undetectable HR (95% CI)
Yes No
Characteristics −369 −61 CHR AHR P value
Duration of follow-­ ≤6 months 225 46 Ref –
up in month >6 months 144 15 0.7 (0.5 to 1.6) –

TDF+3TC+DTG and ABC+3TC+DTG: tenofovir disoproxil fumarate (TDF), lamivudine (3TC) and dolutegravir (DTG), or abacavir (ABC),
lamivudine (3TC), and dolutegravir (DTG). AZT+3TC+DTG and ABC+3TC+EFV: zidovudine (AZT), lamivudine (3TC), and dolutegravir

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
(DTG), or abacavir (ABC), lamivudine (3TC), and efavirenz (EFV). RAL+ABC+3TC, ABC+3TC+EFV and TDF+3TC+EFV: raltegravir (RAL),
abacavir (ABC) and lamivudine (3TC), abacavir (ABC), lamivudine (3TC), and efavirenz (EFV), or tenofovir disoproxil fumarate (TDF),
lamivudine (3TC), and efavirenz (EFV). AZT+3TC+LPV/r and TDF+3TC+ATV/r: zidovudine (AZT), lamivudine (3TC) and lopinavir/ritonavir
(LPV/r), or tenofovir disoproxil fumarate (TDF), lamivudine (3TC), and atazanavir/ritonavir (ATV/r).
*p<0.05, **p<0.01.
AHR, adjusted HR; CD4, cluster of differentiation 4; CHR, crude HR; CPT, cotrimoxazole preventive therapy; IPT, isoniazid preventive
therapy; TB, tuberculosis.

time for children and adolescents to reach an undetect- plasma viral load in our study, as referenced in Tchouwa
able viral load was found to be 6 months. et al,16 was lower than a study conducted in Oromia, Ethi-
The results of this study indicate that the findings fall opia, where it was reported as 2.03×105 cells/µL.13 The
below the target set by WHO in their strategic plans to observed difference in viral load levels could be attributed
achieve the Sustainable Development Goal of 95-­95-­95 by to variations in the monitoring period and study duration.
2025.10 The results of this study are highly significant as Recently, viral load monitoring has become the gold stan-
they offer valuable evidence for designing targeted inter- dard for assessing HIV treatment effectiveness, surpassing
ventions in healthcare institutions. These interventions CD4 count measurements. In our study, conducted more
aim to align with and support the ambitious plans estab- recently, regular viral load monitoring was implemented,
lished by the WHO. allowing for continuous feedback to healthcare providers
In contrast, the proportion of children and adolescents and study participants. This approach aims to improve
who achieved an undetectable viral load in this study the quality of life for individuals living with HIV.
was higher compared with a study conducted in eight Our study found that children with good adherence
countries, including South Africa, the USA, Thailand, to ART had 2.1 times higher likelihood of achieving an
Mali, Burkina Faso, Swaziland and India.14 Additionally, undetectable viral load compared with children and
studies conducted in Cameroon, Cambodia and Uganda adolescents with fair/poor adherence. This finding is
reported viral undetectable rates of 72.1%, 76.8% and consistent with a study conducted in Uganda.11 This can
89%, respectively.15 16 The observed difference in viral be attributed to the fact that maintaining a high level of
load undetectable rates could be attributed to varia- adherence to ART is crucial in preventing viral failure,
tions in the measurement cut-­off point and study design. reducing the risk of viral transmission and decreasing
Most of the aforementioned studies were cross-­sectional, HIV/AIDS-­ related deaths in order to achieve viral
whereas our study followed a cohort design. Additionally, suppression.19 20 Additionally, in our study, children and
differences in the study area, duration of time on ART adolescents with a nutritional status of ≥−2 Z score had
and varying lengths of follow-­ up periods might have 1.8 times higher likelihood of achieving an undetectable
contributed to the disparities. viral load compared with children with a nutritional status
The median survival time required to achieve an >−2 Z score. This finding aligns with the results of a study
undetectable viral load in this study was similar to a conducted in a different setting.21
study conducted in Oromia, Ethiopia, which reported This could be attributed to the fact that malnutrition
a median survival time of 181 days (95% CI 140.5 to is a significant contributor to childhood morbidity and
221.4).13 However, the median survival time observed in can weaken the immune system, leading to an increase in
our study was longer compared with a study conducted viral load levels, both independently and in combination
in the Netherlands, where the median survival time to with other illnesses.22 23
achieve an undetectable viral load was reported as 60 days In our study, children and adolescents with a CD4 count
(12–168 days).17 Furthermore, the median plasma viral above the threshold level had 2.0 times higher likelihood
load observed in our study was similar to that reported of achieving an undetectable viral load compared with
in a study conducted in Kenya, which noted a value of children with a CD4 count below the threshold level. This
1.65×103 cells/µL,18 The median plasma viral load in finding is consistent with a study conducted in Uganda.11
our study, as reported in reference Tesfahunegn et al,19 This can be explained by the fact that low CD4 cell counts
was lower compared with a study conducted in Uganda, and immune activation are both associated with higher
where it was found to be 8.1×104 cells/µL.15 The median viral load levels and increased risk of non-­AIDS-­related

Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206 7

Open access

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
Figure 3 (A–D) Kaplan-­Meier of viral load undetected state by CD4 count, level of ART adherence, nutritional status and Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
treatment failure among children and adolescents living with HIV at South Gondar health institutions.19 ART, antiretroviral
therapy; CD4, cluster of differentiation 4.

mortality. These conditions make children more suscep- medication, genetic mutations in the HIV virus and incor-
tible to opportunistic infections, leading to difficulties in rect dosing of antiretroviral medication.26 27
achieving an undetectable viral load.24 25 Despite the efforts made in this study to estimate the
In our study, children and adolescents who did not undetectable viral load among children and adolescents
experience treatment failure had 3.0 times higher like- initiating ART in Ethiopia across multiple health institu-
lihood of achieving an undetectable viral load compared tions with a long-­term follow-­up period, there are several
with children who had treatment failure. This finding limitations to consider. First, the retrospective design
aligns with the results of studies conducted in Cameroon of the study introduces the possibility of incomplete or
and Uganda.15 16 This can be attributed to several missing data due to reliance on existing medical records,
reasons that can lead to an increase in viral load. These potentially impacting the accuracy and comprehensive-
reasons include inconsistent adherence to antiretroviral ness of the analysis. Second, the study’s 8-­year duration

8 Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206

 Open access

may not fully capture the most recent advancements in provided critical revisions to the article. The final article was reviewed and approved
HIV treatment guidelines and interventions that could by all authors.

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
influence treatment outcomes. As HIV care and manage- Funding Debre Tabor University: Grant no: NO/HP/712/01/2023 G.C.
ment continue to evolve, the findings of this study may Competing interests None declared.
not fully reflect the current landscape of HIV treatment Patient and public involvement Patients and/or the public were not involved in
practices. Additionally, it is important to acknowledge the design, or conduct, or reporting, or dissemination plans of this research.
that certain potential predictors, such as monthly income, Patient consent for publication Not applicable.
substance use and family size, were not thoroughly inves- Ethics approval Ethical clearance was obtained from the ethics review committee
tigated in this study. These factors have the potential to at Debre Tabor University. Additionally, permission letters were obtained from each
influence the achievement of an undetectable viral load of the South Gondar health institutions involved in the study. In order to maintain

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
confidentiality, the names and medical identification numbers of the children were
but were not examined in detail. The absence of these not collected.
variables limits the comprehensive understanding of
Provenance and peer review Not commissioned; externally peer reviewed.
the various factors that can impact treatment outcomes
among children and adolescents living with HIV. Data availability statement All data relevant to the study are included in the
article or uploaded as online supplemental information.
Supplemental material This content has been supplied by the author(s). It has
not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been
peer-­reviewed. Any opinions or recommendations discussed are solely those
CONCLUSION of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and
The lower proportion of individuals achieving an unde- responsibility arising from any reliance placed on the content. Where the content
tectable viral load compared with the target set by the includes any translated material, BMJ does not warrant the accuracy and reliability
of the translations (including but not limited to local regulations, clinical guidelines,
WHO strategic plan (95-­95-­95 target by 2025) highlights terminology, drug names and drug dosages), and is not responsible for any error
the need for targeted interventions and improved HIV and/or omissions arising from translation and adaptation or otherwise.
management strategies. Factors such as CD4 count, ART Open access This is an open access article distributed in accordance with the
adherence, nutritional status and treatment failure play Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which
a significant role in achieving viral load undetectable permits others to distribute, remix, adapt, build upon this work non-­commercially,
state. Healthcare providers better prioritise these factors and license their derivative works on different terms, provided the original work is
properly cited, appropriate credit is given, any changes made indicated, and the use
through comprehensive care and support. The study is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
emphasises the importance of aligning efforts with the
WHO strategic plan to meet targets and improve treat- ORCID iDs
Ermias Sisay Chanie http://orcid.org/0000-0002-3124-5380
ment outcomes. Therefore, stakeholders better address Dejen Getaneh Feleke http://orcid.org/0000-0003-0855-3630
the identified predictors to enhance care and support for Tigabu Desie Emiru http://orcid.org/0000-0001-9535-0190
this vulnerable population. Habtamu Shimels Hailemeskel http://orcid.org/0000-0002-5972-4818
Astewle Andargie Baye http://orcid.org/0009-0003-4387-0305
Author affiliations Berihun Bantie http://orcid.org/0000-0002-8438-785X
1
Pedatric and Neonatal Nursing, Debre Tabor University, Debre Tabor, Ethiopia Abraham Tsedalu Amare http://orcid.org/0000-0002-0186-9627
2 Denekew Tenaw Anley http://orcid.org/0000-0002-5612-6943
Department of Pediatrics and Child Health Nursing, Debre Tabor University, Debre
Anteneh Mengist Dessie http://orcid.org/0000-0002-5919-9670
Tabor, Ethiopia
3 Sintayehu Asnakew http://orcid.org/0000-0002-1241-4468
Debre Tabor University, Debre Tabor, Ethiopia
4 Natnael Moges http://orcid.org/0000-0002-1274-5857
Department of Nursing, Debre Tabor University, Debre Tabor, Ethiopia
5 Fisha Alebel GebreEyesus http://orcid.org/0000-0001-7358-0577
Department of Comprehensive Nursing, Debre Tabor University, Debre Tabor,
Ethiopia
6
Department of Public Health, Debre Tabor University, Debre Tabor, Ethiopia
7
Department of Psychiatric, Debre Tabor University, Debre Tabor, Ethiopia
8
Department of Sport Science, Debre Tabor University, Debre Tabor, Ethiopia
9 REFERENCES
Department of Pediatrics and Child Health Nursing, Wollo University, Dessie, 1 ‘2019-­UNAIDS-­Data. 2019 Available: https://Www.Unaids.Org/Sites/
Ethiopia Default/Files/Media_asset/2019-UNAIDS-Data_en.Pdf
10
Department of Pediatrics and Child Health Nursing, Assosa University, Asosa, 2 Abdella S. HIV related estimats and projections in ethiopia for the
Ethiopia year-­2019. 2020.
11
Department of Pediatrics and Child Health Nursing, Wolkite University, Welkite, 3 Haas AD, Keiser O, Balestre E, et al. Monitoring and switching of
first-­line antiretroviral therapy in adult treatment cohorts in sub-­
Ethiopia
12 Saharan Africa: collaborative analysis. Lancet HIV 2015;2:e271–8.
Debre Berhan University, Debre Berhan, Ethiopia 4 WHO. Antiretroviral therapy (ART) coverage among all age groups.
n.d. Available: http://www.who.int/gho/hiv/epidemic_response/ART/​
Acknowledgements First and foremost, we would like to extend our sincere en/
appreciation to Debre Tabor University for its support. Secondly, we would like to 5 Viral load test | NIH. n.d. Available: https://clinicalinfo.hiv.gov/en/​
glossary/viral-load-test
express our heartfelt gratitude to the study participants, the ART focal person, the
6 Shey ND, Dzemo KO, Siysi VV, et al. Quality of life of HIV patients on
staff of each health institution and the dedicated data collectors who made this highly active antiretroviral therapy: a scoping review. J Public Health
study possible. Epidemiol 2020;12:63–73.
Contributors ESC played a role in conceiving the study, designing the research, 7 Moolasart V, Chottanapund S, Ausavapipit J, et al. The effect
of detectable HIV viral load among HIV-­infected children during
acquiring and interpreting the data and writing the original article. Authors; DGF,
antiretroviral treatment: a cross-­sectional study. Children (Basel)
TDE, AGA, HSH, and AAB Formulating the research question(s) and designing the 2018;5:6.
study. Authors; BB, ATA, TDN, DTA, AMD, SA and NM were involved in statistical 8 What is an undetectable viral load? 2017. Available: https://
analysis, interpreting the results and writing the original article. Authors; SSC, www.avert.org/living-with-hiv/antiretroviral-treatment/what-does-​
EBW, MSJ, FAG, MAD, WMK, AKA, and DAF were involved in data acquisition, and undetectable-mean

Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206 9

Open access

9 NIH. Viral suppression. n.d. Available: https://clinicalinfo.hiv.gov/en/​ 19 Tesfahunegn TB, Berhe N, Abraha TH, et al. Adherence to
glossary/viral-suppression antiretroviral therapy and associated factors among HIV-­infected
10 Bain LE, Nkoke C, Noubiap JJN. UNAIDS 90–90–90 targets to end

BMJ Open: first published as 10.1136/bmjopen-2023-083206 on 26 October 2024. Downloaded from http://bmjopen.bmj.com/ on February 13, 2025 by guest .
children in public health institutions of Adwa, Axum, and Shire towns
the AIDS epidemic by 2020 are not realistic: comment on “Can the of Tigray, Northern Ethiopia: a cross-­sectional study. HIV AIDS
UNAIDS 90–90–90 target be achieved? A systematic analysis of (Auckl) 2023;15:217–24.
national HIV treatment cascades” BMJ Glob Health 2017;2:e000227. 20 Shiferaw MB, Endalamaw D, Hussien M. Viral suppression rate
11 Ssebunya R, Wanyenze RK, Lukolyo H, et al. Antiretroviral therapy among children tested for HIV viral load at the Amhara Public Health
initiation within seven days of enrolment: outcomes and time to
Institute, Bahir Dar, Ethiopia. BMC Infect Dis 2019;19:419.
undetectable viral load among children at an urban HIV clinic in
21 Mannheimer S, Friedland G, Matts J, et al. The consistency of
Uganda. BMC Infect Dis 2017;17:439.
12 WHO, Regional Office for Africa. Guideline for HIV care/ART clinical adherence to antiretroviral therapy predicts biologic outcomes for
mentoring in Ethiopia. n.d. Available: https://www.afro.who.int/​ human immunodeficiency virus-­infected persons in clinical trials. Clin
publications/guideline-hiv-careart-clinical-mentoring-ethiopia Infect Dis 2002;34:1115–21.
13 Ali JH, Yirtaw TG. Time to viral load suppression and its associated 22 World Health Organization. Nutrition for living with HIV infants and

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
factors in cohort of patients taking antiretroviral treatment in East children. In: Antiretroviral therapy for HIV infection in infants and
Shewa zone, Oromiya, Ethiopia, 2018. BMC Infect Dis 2019;19:1084. children: towards universal access: recommendations for a public
14 Bonner K, Mezochow A, Roberts T, et al. Viral load monitoring as a health approach: 2010 revision. 2010.
tool to reinforce adherence: a systematic review. J Acquir Immune 23 Okafor C, Fadupin G, Oladokun R. Nutritional status and virological
Defic Syndr 2013;64:74–8. outcomes of children HIV positive attending anti-­retroviral clinic at
15 Bulage L, Ssewanyana I, Nankabirwa V, et al. Factors associated University College Hospital, Ibadan. FNS 2021;12:1088–97.
with virological non-­suppression among HIV-­positive patients on 24 Phillips AN, Staszewski S, Weber R, et al. HIV viral load response to
antiretroviral therapy in Uganda, August 2014–July 2015. BMC Infect antiretroviral therapy according to the baseline CD4 cell count and
Dis 2017;17:326. viral load. JAMA 2001;286:2560–7.
16 Tchouwa GF, Eymard-­Duvernay S, Cournil A, et al. Nationwide
25 NIH. Poor cd4 cell recovery and persistent inflammation despite
estimates of viral load suppression and acquired HIV drug resistance
in cameroon. EClinMed 2018;1:21–7. viral suppression. 2021. Available: https://clinicalinfo.hiv.gov/en/​
17 Martin-­Blondel G, Sauné K, Vu Hai V, et al. Factors associated with guidelines/hiv-clinical-guidelines-adult-and-adolescent-arv/poor-​
a strictly undetectable viral load in HIV-­1-­infected patients. HIV Med cd4-cell-recovery-and-persistent
2012;13:568–73. 26 Aldous JL, Haubrich RH. Defining treatment failure in resource-­rich
18 Kunzweiler CP, Bailey RC, Mehta SD, et al. Factors associated with settings. Curr Opin HIV AIDS 2009;4:459–66.
viral suppression among HIV-­positive Kenyan gay and bisexual men 27 HIV viral load: why this number matters. 2017. Available: https://
who have sex with men. AIDS Care 2018;30:S76–88. www.healthline.com/health/hiv-aids/your-hiv-viral-load

10 Chanie ES, et al. BMJ Open 2024;14:e083206. doi:10.1136/bmjopen-2023-083206