Community-based enhanced adherence counseling among people living with HIV in Ethiopia: outcomes and key predictors of viral suppression

Abstract

Background

Community-based enhanced adherence counseling (CEAC) is person-centered intervention provided to PLHIV on ART and with unsuppressed viral load (VL) in community settings. This study assessed the effectiveness of CEAC in achieving viral load suppression (VLS) and explored key predictors to inform scalable strategies in resource-limited settings.

Methods

A quantitative, retrospective follow-up study was conducted between October 2022 and October 2024 on 2839 PLHIVs enrolled to CEAC service. Study participants were HIV-infected individuals who were on ART for at least six months and had unsuppressed VL. Frontline community health workers provided them with three-to-six monthly sessions of counseling to address underlying barriers to treatment adherence. Client-level data were collected using CommCare mobile app and analyzed in SPSS.

Results

A total of 2839 PLHIV were enrolled in CEAC, of which 2365 (83.3%; 95% CI: 82.0−84.5%) clients achieved VLS after receiving three-to-six months of CEAC. Major significant predictors of VLS included age 1–14 years [AOR (95% CI) = 2.08 (1.25–3.41)], having a baseline VL of < 10,000 copies/mL [AOR (95% CI) = 1.75 (1.41–2.17)], and enrolled to other community-based case management services [AOR (95% CI) = 5.77 (4.33–7.71)]. Moreover, PLHIV with adherence related challenges resolved had higher odds of VLS [AOR = 1.86; 95%CI: 1.48 − 2.34].

Conclusion

CEAC service demonstrated encouraging results in supporting VLS among PLHIVs on ART with unsuppressed VL count. Community-based interventions showed potential in addressing individual barriers that were challenging to health facility to improve VLS. We recommend continuing efforts to scale up and integrate CEAC with ongoing health facility EAC services to synergistically improve VLS and accelerate epidemic control.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12981-025-00815-8.

Introduction

Globally, 39.9 million people were living with HIV in 2023, with Sub-Saharan Africa accounting for 52% (20.8 million) of this burden [1–3]. Ethiopia has one of the highest numbers of people living with HIV (PLHIV) in this region (610,000) [4, 5].

Ethiopia has adopted the UNAIDS 95-95-95 targets where 95% of PLHIV know their HIV status, 95% of those diagnosed are on sustained antiretroviral therapy (ART), and 95% of those on ART achieve viral load suppression (VLS) [6–8]. Ethiopia has reported strong performance with achievement of 90%, 94%, and 96% for the 95 targets respectively [9].

The World Health Organization (WHO) recommends VL testing for monitoring ART effectiveness [10, 11]. VL monitoring is vital to track treatment success and identify challenges [11]. Clients with unsuppressed VLs exhibit higher risk of AIDS and mortality. For such clients, WHO recommends Enhanced Adherence Counseling (EAC) to identify barriers and develop individualized strategies to overcome them [12].

WHO recommends that 70% of clients with unsuppressed VL achieve suppression after 3–6 EAC sessions [13, 14]. Outcomes across sub-Saharan Africa have remained inconsistent. A recent meta-analysis of 12 African studies reported only 51.2% suppression post-EAC [15], that showed a significant gap in meeting global targets. Individual studies showed regional variation, this may be due to many reasons like differences in population characteristics, health system context, and EAC implementation approach. For instance, retrospective facility-based studies reported, 91% in a selected Nigerian state [16], 73.8% in Southern Nigeria [17], and 74.2% in Uganda [18] were exceeded the WHO threshold. These higher figures may reflect that strong health system, more stable population with fewer adherence barriers, targeted interventions, or better-resourced facilities. In contrast, similar retrospective facility-based studies showed below WHO threshold, 66.4% in Kampala [19], 61% in Zambia [20], and 45% in Harare [21]. These lower achievements may be due to systemic constraints such as resource limitations, shortage of healthcare workers, or quality of EAC.

Similarly in Ethiopia, retrospective facility-based studies showed below the WHO threshold, for instance, 66.4% in North Wollo [22], 65.2% in Nekemte [23], 51.7% in West Gojjam [24] and 40.9% in Hawassa City [25]. Despite using similar methodology, these differences may highlight the impact of local context, resources, or population characteristics.

Overall, in resource-limited settings, VLS outcomes are influenced by a variety of social, psychological, economic, health system, and medical factors [23, 26, 27]. A study from Harare showed clients with VL >5000 copies/ml were less likely to achieve VLS [21]; in Uganda, those with initial VL >10,000 copies/ml showed poor response to EAC [28]. In Ethiopia malnutrition, poor adherence, and fewer EAC sessions were significantly associated with low levels of VLS [25].

Consistently low VLS rates following EAC at health facilities suggest that traditional counseling may not fully meet clients’ broader psychosocial and economic needs [29, 30]. Therefore, multi-pronged approaches, such as peer-led and community-based models, may enhance VLS outcomes [31].

EAC is an iterative process that assesses adherence levels, identifies barriers, and develops individualized action plans to improve outcomes and reduce treatment failure [14, 32]. To address these gaps, Ethiopia’s PEPFAR/USAID-supported Community HIV Care and Treatment (CHCT) activity introduced a differentiated model of care known as Community Enhanced Adherence Counseling in 2022. This community-based intervention was delivered in collaboration with health facilities and local implementing partners, engaged trained Community Health Workers (CHWs) to provide tailored adherence support to individuals with unsuppressed VL. CEAC is like health facility level EAC, it is an iterative process that is used to design individualized treatment adherence plans and follow-up support, but with an emphasis on community-based support that aims to improve adherence, VLS, and strengthen linkage to care through targeted strategies implemented in 3- or 6-months long sessions [33], shown below Fig. 1.

Fig. 1.

The process and contents of community-based enhanced adherence counseling (CEAC) intervention to clients with unsuppressed HIV viral load, PEPFAR/USAID CHCT activity

Despite its potential, implementation of CEAC is limited to high-prevalence towns in Ethiopia, and the outcomes of such interventions have not been assessed. Here, we described the implementation procedures and outcomes of CEAC documented the practical lessons, evaluated effectiveness, and offered recommendations for scaling up the similar approaches in resource-limited settings. Sharing this finding may be critical for informing policy, guiding programmatic decisions, and accelerating the progress toward controlling the HIV epidemic.

Methods

Study area and setting

This implementation research was conducted in eight high HIV-burden regions supported by PEPFAR, including Addis Ababa, Amhara, Oromia, Gambella, Sidama, Southwest Ethiopia, Central Ethiopia, and South Ethiopia regions. A total of 189 public health facilities (50 hospitals and 139 health centers) were included.

The CHCT activity is a PEPFAR/USAID-funded initiative implemented by Project HOPE Ethiopia in collaboration with the Ministry of Health and local partners. Since 2017, CHCT has aimed to accelerate and sustain HIV epidemic control in Ethiopia through community-based care and treatment services [33].

To support clients with unsuppressed VL after six months of ART initiation, the CHCT program introduced CEAC. CHWs acting as Community-Engagement Facilitators (CEF) received lists of unsuppressed clients who consented to CEAC service from nearby facilities. Clients are traced and offered CEAC services at the community level. Those who enrolled were provided with three consecutive monthly CEAC sessions. Clients adhering to treatment and not presenting with an opportunistic infection received a follow-up VL test after third CEAC session completed. If unsuppressed, counseling continued to additional three months before considering an alternative ART regimen [15, 33] (Fig. 1).

Study design

A quantitative, retrospective follow-up open cohort study was done between October 2022 and October 2024, on 2839 PLHIVs across 189 health facilities in eight high-burden regions of Ethiopia.

Source population and study population

The source population were PLHIV who received first-line ART and had ongoing unsuppressed VLs (> 50 copies/ml) after six months of treatment ART initiation, identified between October 2022 and October 2024 from 189 hospitals and health centers.

The study population included PLHIV with initial VL of above 50 copies/ml, were enrolled in CEAC, and attended at least three CEAC sessions. And analysis was conducted for clients VL count tested after completed the sessions with documented VL test results.

Clients who did not attend at least three sessions, did not have a follow-up VL test, discontinued the program, died, or transferred out before completing the required counseling sessions were excluded from the study.

Sample size/participants

A total of 2839 clients from 189 health facilities who were enrolled in the CEAC service, completed at least three counseling sessions and had their follow-up VL documented in the CommCare Unified Data System (UDS) during the study period, were included for analysis.

Data collection procedures

Project HOPE developed a standardized UDS built on CommCare. UDS is a digital health application designed for data entry on mobile devices. Trained CHWs assigned to specific service delivery units collected individual client data using dedicated accounts. At ART clinics, healthcare providers compiled line lists of clients with unsuppressed VLs after verbal consent, which were entered into the CommCare app by CEFs. Data was stored in a local server for analysis.

Data quality control

Data quality was ensured through validation rules that were built-in to the UDS at design stage in which individuals were granted access to the UDS based on their roles, comprehensive training for data collectors, use of standard user manuals, regular data quality assessments, and continuous follow-up support. Data cleaning and verifications were conducted quarterly. These measures helped maintain data accuracy, completeness, and consistency throughout the implementation of the program and study.

Data processing and analysis

Data were de-identified and exported from UDS/CommCare to Excel and SPSS version 20 for statistical analyses. Descriptive statistics summarized key variables in percentages, tables, and charts. A bivariate logistic regression model was fitted to identify potential predictors (p < 0.2), which were then included in multivariate binary logistic regression models to control confounding. Statistical significance was set at p < 0.05, and results were expressed as adjusted odds ratios (AOR) and 95% confidence intervals (CI). Model fit was assessed by the Omnibus test (p < 0.001) and Hosmer-Lemeshow test (p = 0.58) indicated that the model was well fitted for identifying the predictors.

Study variables

The primary outcome variable was VLS status, measured as a binary outcome. Participants were classified as having either a suppressed VL (≤ 50 copies/ml) or an unsuppressed VL (> 50 copies/ml) based on their most recent VL test result after completing the CEAC sessions.

The independent variables were grouped into socio-demographic factors including participant sex, age, place of residence, and service location. Clinical variables were VL measurement at enrollment, drug side effects, and contact elicitation for HIV testing. Service-related variables included HCW counseling skills, health service delivery setting, and distance to the facility. Client-related factors included forgetfulness, absence of family support, and frequent travel or mobility. Community support factors included whether the participant was enrolled in other community-based HIV care and treatment services.

Operational definition

CEAC: is a targeted intervention designed to help clients identify and overcome barriers to adherence, while developing personalized strategies to improve VLS at the community level.

Follow-up VL count is performed after completing three consecutive monthly CEAC sessions, and the client is adhered to ART and without any opportunistic infections during the EAC sessions.

Suppressed VL: Patient is considered to have a suppressed VL if their count drops to ≤ 50 copies/ml after CEAC completion [14].

Unsuppressed VL: Patient is classified as having unsuppressed VL if their count exceeds 50 copies/ml after CEAC completion [14].

Low-Level Viremia: If patient’s VL is >50 copies/ml but ≤ 1000 copies/ml after CEAC, they are classified as having low-level viremia. They should continue their ART regimen and undergo VL testing every six months.

High viral load: Patient is considered to have a high VL if their count exceeds 1000 copies/ml after completing CEAC.

Ethical considerations

The Institutional Review Boards of the Ethiopian Public Health Association provided ethical clearance. While the study is a retrospective follow-up study and part of routine service delivery, beneficiaries verbally consented to the use of their data for program monitoring, donor reporting, and implementation research. For children, verbal assents were secured from parents and caregivers. Strict protocols were in place to ensure the protection of client data, safeguarding its security, privacy, and confidentiality.

Results

Socio-demographic characteristics

Data of 5529 PLHIV with unsuppressed VL were available, of which, 2839 client records were analyzed for those who completed at least 3 CEAC sessions and underwent follow-up VL testing. More than half 1702 clients (60%) were females, with a mean age of 34 years (SD ± 13.3). The majority, 2148 (76%) of clients, were adults aged 25 years or older. In addition, 1647 (58%) of clients received HIV care and treatment service including ART refilling service at health center level, and 1192 (42%) at hospital level, as detailed in Table 1.

Table 1.

Socio-demographic characteristics of clients enrolled in CEAC in Ethiopia (n = 2839)

Variables	Frequency (N)	Percentage (%)
Sex
Female	1702	60
Male	1137	40
Age (in years)
1–10	98	3
11–14	93	3
15–19	287	10
20–24	213	8
25–29	274	10
30–39	863	30
40–49	639	23
50+	372	13
Service region
Amhara	1036	36
Oromia	655	23
Addis Ababa	577	20
SNNPR	405	14
Gambella	166	6
Client residence where HIV care & treatment service received
Capital City	577	20
Regional Town	267	9
Zonal Town	907	31
Woreda Town	1088	38
Type of health facility where clients have received HIV care & treatment services
Health center	1643	58
Hospital	1196	42

Clinical and service-related characteristics

At the time of CEAC enrollment, 1861 (65%) clients had a VL between 50 and 10,000 copies/ml, while the remaining 998 (35%) clients had a VL of ≥ 10,000 copies/ml. Regarding enrollment timing, 1133 clients (40%) were linked to CEAC within one month of their most recent VL count. However, over half (1706, 60%) of the clients experienced delay of more than a month after receiving the unsuppressed VL result before being linked to CEAC.

Furthermore, 39% (1098 clients) were enrolled in other CHCT case management (CM) services (such as including linked to support groups, enrolled in village saving and loan associations (VSLA), community-based management of depression, screening and referrals, and disclosure and food support) alongside CEAC, in contrast, 61% (1748 clients) were not enrolled in any supplementary CM services, as detailed in Table 2.

Table 2.

Clinical and service-related characteristics of clients enrolled to CEAC in Ethiopia (n = 2839)

Clinical and service-related characteristics	Frequency (N)	Percentage (%)
VL count at time of enrollment to CEAC
50–1000	783	28
1001–5000	804	28
5001–10,000	254	9
> 10,000	998	35
The duration between the initial VL test and CEAC start (in weeks)
1–4 weeks	1133	40
5–8 weeks	676	24
9–12 weeks	507	18
>12 weeks	523	18
Clients who missed one or more CEAC session
Not missed	2603	92
Missed	236	8
EAC session started at the health facility before list received to CEAC
Yes	1133	40
No	1706	60
Duration between 1 st CEAC and follow-up VL count (in weeks)
< 8 weeks	522	18
9–12 weeks	1743	61
13–16 weeks	242	9
> 16 weeks	332	12
Contact elicitation for HIV testing
Elicited contacts with unknown HIV status after enrolled into CEAC	568	20
Clients had no contacts (with unknown HIV status) to elicit at time of enrollment	2271	80
Clients enrolled in CHCT care and support service in addition to CEAC service
Enrolled to CM service	1098	39
Not enrolled in CM service	1741	61

Adherence-related barriers

Various barriers to ART adherence were reported by clients at the time of enrollment into CEAC. Personal challenges were the most common, affecting 1114 clients (39%), followed by socio-economic factors 781 (28%), medication related factors 155 (5%), and health system-related factors 77 (3%), as detailed in Table 3.

Table 3.

Adherence-related barriers among clients enrolled in CEAC in Ethiopia (n = 2839)

Adherence related barriers	Frequency (N)	Percentage (%)
Client related adherence problems including forgetfulness, travel away, no family support on ART reminder
Yes	1114	39
Not mentioned	1725	61
Socio economical related adherence problems including lack of community support, lack of many for transport, or lack of food
Yes	781	28
Not mentioned	2058	72
Health system (health service delivery setting) related adherence problems including incontinent and frequent appointments, long waiting, lack of confidentiality, long distance, insufficient counseling, and/or bad healthcare worker attitude
Yes	77	3
No or not mentioned	2762	97
Medication related adherence problems including pill burden, drug side effect
Yes	155	5
No or not mentioned	2684	95

Viral load suppression status following completion of CEAC sessions

In this study, 2365 of the 2839 patients (83.3%; 95%CI: 82.0−84.5%) achieved VLS after completing CEAC sessions.

Of those who achieved suppression, 2118 clients (90%) attained suppression following their third CEAC session. The remaining 247 clients (10%) achieved suppression after completing six CEAC sessions.

Overall, 474 clients (17%) did not achieve VLS. Among them, 285 clients (10%) had high VL levels (≥ 1000 copies/ml), while 189 clients (6.7%) showed low-level viremia, with VL levels ranging between 51 and 1000 copies/ml, these outcomes are shown in Fig. 2.

Fig. 2.

Viral load suppression status of client’s after completed CEAC session in Ethiopia (n = 2839), 2025

VLS levels varied when disaggregated by age: the highest viral load suppression was observed among children aged 1–14 years (87%), followed by adults aged 25 and above (84%), while lower rates were seen among adolescents aged 15–19 (79%) and youth aged 20–24 (78%) years. However, our analysis showed no significant differences in VLS levels between females (83%) and males (84%).

Regional comparisons showed relatively consistent outcomes. The Gambella region recorded the highest rate at 87%, followed by Amhara (86%), Oromia (85%), and Addis Ababa (81%). In contrast, SNNPR (including Sidama, Southwest Ethiopia, Central Ethiopia, and South Ethiopia) reported a lower VLS level of 76%, as detailed in Table 4 and (Annex supplementary Table 1).

Table 4.

Viral load suppression status among selected sociodemographic and clinical and service-related characteristics of clients after CEAC session in Ethiopia (n = 2839)

Selected variables	# Clients N (%)	VL suppression status after CEAC
		Suppressed N (%)	Unsuppressed N (%)
Sex
Female	1702 (60)	1407 (83)	295 (17)
Male	1137 (40)	958 (84)	179 (16)
Age (in years)
1–14	191 (7)	167 (87)	24 (13)
15–19	287 (10)	228 (79)	59 (21)
20–24	213 (8)	167 (78)	46 (22)
≥ 25	2148 (76)	1803 (84)	345 (16)
Service region
Amhara	1036 (36)	890 (86)	146 (14)
Oromia	655 (23)	555 (85)	100 (15)
Addis Ababa	577 (20)	468 (81)	109 (19)
SNNPR	405 (14)	307 (76)	98 (24)
Gambella	166 (6)	145 (87)	21 (13)
The type of health facility clients received C&T service received
Government health center	1643 (58)	1418 (86)	225 (14)
Government hospital	1196 (42)	947 (79)	249 (21)
VL count at time of enrollment to CEAC
50–10,000 copies/ml	1841 (65)	1586 (86)	255 (14)
> 10,000 copies/ml	998 (35)	779 (78)	219 (22)
Contact elicitation for HIV testing
Elicited contacts with unknown HIV status after enrolling to CEAC service	568 (20)	462 (81)	106 (19)
Clients had no unknown HIV status contact at time of enrollment	2271 (80)	1903 (84)	368 (16)
Clients enrolled in CHCT care and support service in addition to CEAC service
Enrolled to CM service	1098 (39)	1036 (94)	62 (6)
Not enrolled in CM service	1741 (61)	1329 (76)	412 (24)
Client related adherence problems
Yes	1114 (39)	981 (88)	133 (12)
Not mentioned	1725 (61)	1384 (80)	341 (20)

Predictors of VLS after completion of CEAC services

To determine the factors associated with VLS following CEAC, a binary logistic regression analysis was first conducted for each independent variable. These included socio-demographic characteristics (sex, age, region, and population type), clinical and service-related factors (health facility type, baseline VL count, timing of CEAC initiation, and attendance), as well as adherence-related challenges and community care engagement. Variables with a p-value less than 0.2 in the bivariate analysis and biologically plausible variables were entered into a multivariable logistic regression model to identify independent predictors.

The multivariable analysis revealed several statistically significant predictors of VLS. Clients residing in the Amhara and Gambella regions had higher odds of achieving VLS compared to those in the SNNPR, with [AOR = 1.65; 95% CI:1.22–2.24] and [AOR = 1.81; 95% CI: 1.05–3.12] respectively. Age was also a strong predictor, as children aged 1 to 14 years were twice as likely to achieve VLS than adolescents and young peoples aged 15 to 24 years [AOR = 2.08; 95% CI: 1.25–3.41].

The type of facility where clients received HIV care and treatment including ART refilling services also influenced the VLS outcomes. Those who received the HIV care and treatment service at health center level were 1.4 times more likely to achieve VLS than those treated at hospital level [AOR = 1.40; 95% CI: 1.13–1.74].

Clients with a baseline VL count of < 10,000 copies/ml at enrollment to CEAC were 1.75 times more likely to achieve VLS than those with higher baseline levels [AOR = 1.75; 95% CI: 1.41–2.17].

Clients who had no contacts with unknown HIV status at the time of enrollment to CEAC were 1.4 times more likely to achieve VLS than those who had reported contacts with unknown HIV status lately after CEAC enrollment [AOR = 1.41; 95% CI = 1.10–1.83].

Clients who were enrolled in community-based HIV care and support services beyond CEAC had significantly improved outcomes, with an AOR of [AOR = 5.77; 95% CI: 4.33–7.71], indicating they were nearly six times more likely to reach VLS compared to those not enrolled in additional community care.

Regarding adherence-related factors, clients who reported personal adherence-related challenges at enrollment but subsequently resolved these issues were [AOR = 1.86; 95% CI: 1.48–2.34] nearly two times more likely to achieve VLS than those who reported no such challenges. In contrast, other adherence-related barriers (socioeconomic, health care delivery setting, and medication-related), did not show a statistically significant association with VLS. Likewise, other service-related variables, such as whether clients began EAC at health facilities prior to CEAC enrollment or the timing between the last VL test and CEAC initiation, were not significantly associated with treatment outcomes, as detailed in Table 5.

Table 5.

Bivariate and multivariate binary logistics analysis of selected predictors that determine the VLS among unsuppressed VL count clients after CEAC (n = 2839)

Significant predictors for VLS	VLS status after CEAC sessions		Binary logistic regression analysis
			Bivariate	Multivariate
	Suppressed N (%)	Unsuppressed N (%)	COR (95% CI)	AOR (95% CI)
Clients service region
Addis Ababa	468 (81)	109 (19)	1.37 (1.01–1.87)	0.98 (0.69–1.37)
Amhara	890 (86)	146 (14)	1.95 (1.46–2.59)	1.65 (1.22–2.24*)
Oromia	555 (85)	100 (15)	1.77 (1.29–2.42)	1.23 (0.88–1.72)
Gambella	145 (87)	21 (13)	2.20 (1.32–3.67)	1.81 (1.05–3.12*)
SNNPR	307 (76)	98 (24)	1	1
Age range
Children (1–14)	167 (87)	24 (13)	1.85 (1.15–2.99)	2.08 (1.25–3.41*)
Adult (≥ 25)	1803 (84)	345 (16)	1.39 (1.09–1.77)	V (0.94–1.58)
Adolescent & Young (15–24)	395 (79)	105 (21)	1	1
Type of health facilities clients served
Health Center	1418 (86)	225 (14)	V (1.34–2.02)	1.40 (1.13–1.74*)
Hospital	947 (79)	249 (21)	1	1
Client VL result at the time of CEAC enrollment
50–10,000 copies/ml	1586 (86)	255 (14)	1.75 (1.43–2.14)	1.75 (1.41–2.17*)
> 10,000 copies/ml	779 (78)	219 (22)	1	1
Contact elicitation for HIV testing
Elicited untested contact after enrolling to CEAC service	462 (81)	106 (19)	1	1
Clients had no untested contact at time of enrollment	1903 (84)	368 (16)	1.19 (0.93–1.51)	1.41 1.10–1.83*)
Clients enrolled to CHCT care and support services
Enrolled to CM service	1036 (94)	62 (6)	5.18 (3.92–6.85)	5.77 (4.33–7.710*
Not enrolled to CM service	1329(76)	412 (24)	1	1
Clients related adherence problems at enrollment and problem solved
Had problem	981 (88)	133 (12)	1.82 (1.46–2.57)	1.86 (1.48–2.34*)
Hadn’t any problem or not mentioned	1384 (80)	341 (20)	1	1

*Significant association to VLS in multiple binary logistic regression analysis with P < 0.05. 1 = reference

Discussion

In this study, we examined the level of VLS and associated factors among clients who received CEAC in Ethiopia. The overall VLS rate was 83.3% (95% CI: 82.0–84.5%), exceeded the WHO recommendation of 70% following EAC [13, 14] and a recent meta-analysis study in 12 African counties (51.2%) [15] post EAC. this finding also higher VLS rate than many African countries reported similar retrospective facility-based studies such as 74.2% in Uganda [18]. 73.8% in Southern Nigeria [17], 66.4% in Kampala [19], 61% in Zambia [20], and 45% Harare [21]. In this study the higher rates of VLS may be not only due to differences in health systems, demographics, and country-specific contexts, but also to the impact of additional support mechanisms. A recent systematic review in sub-Saharan Africa reported that community-based models of care were associated with improved viral suppression outcomes [34]. This finding likely showed the effectiveness of CEAC’s community-based, differentiated approach, which prioritizes individualized interventions over facility-based counseling alone.

In contrast, this study achieved lower VLS than a retrospective facility-based study in selected state of Nigeria, reported as 91% VLS following post EAC [16]. The main difference may be due to the variation in viral load suppression thresholds used. In their study, a viral load of < 1000 copies/mL was considered suppressed, whereas in our study, suppression was defined as a viral load count of < 50 copies/mL, in line with recent WHO and national ART guidelines [14].

In Ethiopia, this study achieved higher than both the national HIV impact assessment survey reported as 70.1% VLS rate [35] and in various regional states of retrospective facility-based studies ranges from 66.4 to 40.9% [22, 23, 24, 25]. This variation may be due to differences in EAC delivery models specifically, the community-based approach may have better addressed clients’ needs that were not fully met by facility-based EAC. Additionally, regional context and cultural differences could also contribute to the observed disparities.

However, our VLS result was lower than the national routine VLS estimate of 91% and the routine follow-up VLS level of 90.6% among PLHIV receiving ART in the Amhara region [36]. This discrepancy may be explained by the fact that our analysis targeted the level of re-suppression among individuals with an initially unsuppressed VL after six months of ART initiation, whereas the national estimates and Amara region’s result were based on routine VL monitoring across all PLHIV on ART. Over all our results aligned with national HIV treatment targets and highlight CEAC’s contribution toward achieving national and UNAIDS goals, particularly 95% viral suppression.

Our analysis identified key predictors of VLS. Regional differences were observed: clients in the Amhara and Gambella regions were significantly more likely to achieve suppression compared to those in SNNPR, with AORs of 1.65 and 1.81, respectively. These disparities may be influenced by differences in population characteristics. A previous retrospective facility-based studies have also reported lower suppression rates in Hawassa, SNNPR (40.9%) compared to similar study results in North Wollo, Amhara region and in Nekemte, Oromia regions [22, 23, 25].

Client age was identified as a significant predictor of VLS. Children aged 1–14 years had over twice the odds of suppression (AOR = 2.08; CI: 1.25–3.41) compared to adolescents aged 15–24 years. This finding is supported by research from Tanzania, which found better suppression among younger children due to increased caregiver involvement [37]. Recent studies have also reported higher odds of having sub-optimal adherence and detectable viremia among young PLHIV (compared to old age groups) receiving ART in Ethiopia [38]. This suggests the need for targeted adherence interventions for adolescents and youth who lack support systems.

Baseline VL was another critical predictor. Clients with VL below 10,000 copies/ml at CEAC enrollment were significantly more likely to suppress VL compared to those with a baseline VL of >10,000 copies/ml (AOR = 1.75, 95%CI: 1.41–2.17). Similar finding was reported in studies conducted in Uganda, Harare and North Wollo Ethiopia [21, 22, 28]. This highlighted the importance of early intervention before VLs escalated.

Clients had no unknown status contacts or completed elicitation prior to enrollment in CEAC were more likely to achieve VLS (AOR = 1.41, 95% CI: 1.10–1.83) compared to those who elicited contacts lately following CEAC. This suggested that early testing of partners or family members may promote better treatment outcomes by encouraging support, reducing stigma, and improving adherence. In contrast, delayed contact testing may reflect unwillingness to disclose further leading to lack of support in their HIV care journey, which may have resulted in lower viral suppression. This was supported by a study conducted in Debre Markos Town, Ethiopia, where only 28.6% of index clients conducted early contact testing and highlighted that early testing was associated with improved client health outcomes [39].

Another strong predictor of VLS was enrolled in community HIV care and support services in addition to CEAC. These clients were nearly six times more likely to achieve VLS (AOR = 5.77, 95% CI: 4.33–7.71). This was consistent with a sub-Saharan Africa systemic analysis, additional support provided clients achieved higher VLS rate [34], and studies in Kenya and Uganda, where participation in psychosocial groups and community associations was linked to improved treatment outcomes [30, 31].

In this study, clients enrolled in community HIV care and support services were provided with additional community-based services based on clients’ need including linked to support groups, enrolled in village saving and loan associations (VSLA), community-based management of depression, screening and referrals, and disclosure and food support [33].

The type of health facility where clients received their HIV care and treatment service including ART refilling also influenced the outcomes of VLS. Clients received HIV care and treatment at health centers level showed higher VLS (AOR = 1.40, 95%CI: 1.13–1.74) than those received the HIV care and treatment service at hospitals level. While no prior studies in Ethiopia to compare directly the VLS outcomes by health facility type, this variation may be due to health centers likely had lower client volumes and closer to the community than hospitals, this may allowed clients to have more frequent visit for consultation, and HCWs would have more time to provide individualized support.

Finally, clients who initially reported personal adherence challenges, but resolved during CEAC, were significantly more likely to achieve VLS (AOR = 1.86, 95% CI: 1.48–2.34). A study from Uganda similarly found that family support improved treatment outcomes (AOR = 1.98; 95% CI: 1.34–3.44) [30]. These findings emphasize the importance of addressing client-specific barriers as part of community-based adherence interventions.

Strengths and limitations of the study

A key strength of this study is its use of routinely collected program data, sourced directly from primary client records. These data were regularly audited for accuracy, completeness and consistency, enhancing data quality and ensuring the results reflect real-world program implementation. Moreover, this study was a multi-center national study conducted in a large and diverse group of PLHIV which facilitates the generalizability of results to all PLHIV receiving ART in Ethiopia. As such, the study offers practical insights into improving VLS, particularly in the context of achieving the third “95” target through differentiated CEAC services. And the findings may serve as a valuable reference for future research and evaluations.

While this study offers valuable insights, certain limitations should be acknowledged. As a retrospective analysis, it relied on routinely collected program data, which limited the inclusion of additional variables such as marital status, religion, income, and some clinical indicators like current CD4 count. Furthermore, the study did not assess the quality of CEAC sessions or explored other contextual and behavioral factors that might influence viral load re-suppression. To mitigate this, relevant literature and program documents were reviewed to identify explanations for the observed differences in suppression within the scope of the available data. Hence, future research, particularly comparative with qualitative explorations, recommended to better understand the impact of service quality and other potential determinants of viral load re-suppression not studied here.

Conclusion and recommendation

This study found that the CEAC service demonstrated encouraging results in supporting VLS among PLHIV who had HVL and receiving ART, with most clients achieving VL suppression following three consecutive CEAC sessions. Targeted community-based interventions showed potential in addressing individual barriers to viral suppression that were challenging to health facility to improve client health outcome. We recommend continued efforts to scale up and integrate CEAC with ongoing health facility EAC services to synergistically improve viral re-suppression and accelerate epidemic control.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Author 1 was responsible for the conceptualization and design of the study, performed the primary data analysis, drafted the manuscript, and led the overall development and finalization of the paper. Author 2 and 3 provided overarching guidance throughout the research process, critically reviewed the manuscript at multiple stages, Author 5, 7, 9 & 11 contributed significantly to the critical revision of the manuscript, providing detailed feedback and suggestions to improve content and clarity. Authors 4, 6, 8, 10, 12, 13, 14, 15 & 16 participated in the manuscript review process, offering comments and recommendations that helped shape the final version. Finally, all authors read and approved the final manuscript.

Funding

All authors declare that we have not received any research funding from any organization to conduct this study we conducted by utilizing available program data.

Data availability

The data that support the findings of this study are available from the corresponding author.

Declarations

Ethics approval and consent to participate

All participants of this study provided verbal consent as part of the standard operating procedure for enrollment into the CEAC services and permission to use their data for program monitoring, donor reporting, and implementation research.

Competing interests

The authors declare no competing interests.