This was an analytical cross-sectional observational study conducted from January to December 2025 at Kalembelembe Pediatric Hospital in Kinshasa, Democratic Republic of the Congo (DRC). The study aimed to identify the psychosocial and structural factors associated with viral non-suppression among adolescents and young people living with HIV receiving antiretroviral therapy (ART).

The study was conducted at Kalembelembe Pediatric Hospital, a national referral center for the care of children, adolescents, and young people living with HIV in the Democratic Republic of the Congo. The hospital provides comprehensive HIV care, including clinical follow-up, ART delivery, routine viral load monitoring, and psychosocial support delivered by a multidisciplinary healthcare team.

The hospital database contained records of 6032 children, adolescents, and young people living with HIV registered between January 2015 and December 2025. During the study period, 842 patients remained actively enrolled in HIV care. After applying the age eligibility criterion (10 - 24 years), 521 medical records were reviewed for eligibility. Of these, 124 were excluded, mainly because of the absence of a recent viral load result, incomplete medical records, or transfer to another healthcare facility. Ultimately, 397 adolescents and young people met the inclusion criteria and were enrolled consecutively using an exhaustive recruitment approach. The participant selection process is illustrated in Figure 1.

Figure 1.Flow diagram of study participants.

Inclusion Criteria

Participants were eligible if they:

were aged 10 - 24 years;were living with HIV;had been receiving ART for at least six months;had a documented viral load result obtained within the 12 months preceding data collection; andprovided written informed consent or, for participants younger than 18 years, had written informed consent provided by a parent or legal guardian.

Exclusion Criteria

The following participants were excluded:

patients without a documented viral load result during the 12 months preceding data collection;patients with incomplete or unusable medical records;patients recently transferred to or from another HIV care facility; andadolescents who were absent during the data collection period or who had cognitive impairment preventing questionnaire administration.

All adolescents and young people who met the eligibility criteria during the study period were included. An exhaustive recruitment strategy was adopted to enroll every eligible patient receiving care at Kalembelembe Pediatric Hospital. No sampling procedure was performed. This approach minimized selection bias and ensured that the study population accurately reflected the target population.

2.6.1. Dependent Variable

The primary outcome was viral non-suppression, defined as a viral load of ≥1000 copies/mL, in accordance with the World Health Organization (WHO) recommendations. The outcome was based on the most recent viral load measurement documented within the 12 months preceding data collection.

2.6.2. Independent Variables

Independent variables included:

Sociodemographic characteristics: age, sex, educational level, and family situation;Clinical characteristics: duration of ART, treatment interruptions, missed clinic appointments, and clinical suspicion of antiretroviral drug resistance;Psychosocial factors: family support, perceived HIV-related stigma, HIV status disclosure, depressive symptoms, participation in peer support groups, and substance use; andStructural factors: accessibility of HIV care, financial difficulties, transportation costs, availability of antiretroviral drugs, and perceived quality of healthcare services.

Depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9) [9], with a score of ≥10 indicating moderate-to-severe depressive symptoms. Family support was assessed using an adapted version of the Multidimensional Scale of Perceived Social Support (MSPSS).

Data were collected from medical records, HIV clinic registers, and a structured questionnaire administered by trained research assistants. Clinical information, including the most recent viral load result, was extracted from patients’ medical records, whereas psychosocial information was obtained during routine follow-up visits through face-to-face interviews. Only viral load measurements performed within the 12 months preceding data collection were considered for analysis.

The questionnaire was developed based on the existing literature and adapted to the local context. It was reviewed by experts for content validity and pilot-tested before implementation. Data collectors received standardized training on study procedures before data collection. Completed questionnaires were reviewed daily to verify completeness and internal consistency before data entry.

Data were analyzed using IBM SPSS Statistics version 25 (IBM Corp., Armonk, NY, USA). Categorical variables were summarized using frequencies and percentages, whereas continuous variables were described using means and standard deviations (SD).

Associations between independent variables and viral non-suppression were assessed using the Pearson chi-square test or Fisher’s exact test, as appropriate. Crude odds ratios (ORs) and their 95% confidence intervals (95% CIs) were calculated.

Variables with a p-value < 0.20 in the bivariate analysis, together with variables considered clinically relevant, were entered into a multivariable logistic regression model to identify factors independently associated with viral non-suppression. Multicollinearity among independent variables was assessed using the Variance Inflation Factor (VIF), while model calibration was evaluated using the Hosmer–Lemeshow goodness-of-fit test [10]. Results are presented as adjusted odds ratios (aORs) with corresponding 95% confidence intervals. Statistical significance was set at p < 0.05. Analyses were restricted to complete cases, and no imputation of missing data was performed.

The study protocol was approved by the appropriate Institutional Ethics Committee and authorized by the administration of Kalembelembe Pediatric Hospital. Written informed consent was obtained from all adult participants and from the parents or legal guardians of participants younger than 18 years, together with participants’ assent where applicable. Participant anonymity and the confidentiality of all collected data were maintained throughout the study.

A total of 397 adolescents and young people living with HIV were included in the study. The participant selection process is illustrated in Figure 1. The sociodemographic characteristics of the participants are summarized in Table 1. The 15 - 19-year age group was the most represented (44.3%), followed by those aged 20 - 24 years (32.5%) and 10 - 14 years (23.2%). The mean age was 17.9 ± 3.8 years. Female participants accounted for 57.4% of the study population. Most participants had attained secondary education (65.2%). Regarding family circumstances, 47.4% were living with a single parent, while 26.7% were orphans.

Table 1. Sociodemographic characteristics of the participants (n = 397).

The clinical characteristics of the participants are presented in Table 2. More than two-fifths of the adolescents and young people (41.6%) had been receiving antiretroviral therapy for more than five years, while 41.1% had been on treatment for two to five years and 17.4% for less than two years.

A history of treatment interruption was reported by 30.5% of participants. Regarding clinical follow-up, 37.0% had not missed any clinic appointments during the previous 12 months, 41.3% had missed one or two appointments, and 21.7% had missed three or more appointments.

The psychosocial and structural characteristics of the participants are presented in Table 3. More than half of the participants (55.2%) reported experiencing HIV-related stigma, whereas 39.3% reported low family support. Moderate-to-severe

Table 2. Clinical characteristics of the participants (n = 397).

Table 3. Psychosocial and structural characteristics of the participants (n = 397).

depressive symptoms (PHQ-9 score ≥10) were observed in 33.2% of participants. Participation in a peer support group was reported by 34.8% of the adolescents and young people.

Regarding structural factors, 50.6% of participants reported financial difficulties related to HIV care, and 43.6% reported transportation challenges in accessing healthcare services. Previous interruptions in the supply of antiretroviral medications were reported by 22.9% of participants.

In addition, 27.2% of participants reported substance use, 53.9% belonged to households with low socioeconomic status, and 12.3% had a clinical suspicion of antiretroviral drug resistance.

The prevalence of viral non-suppression was 33.5% (133/397), whereas 66.5% (264/397) of participants achieved viral suppression (viral load < 1000 copies/mL), according to the World Health Organization recommendations (Table 4).

Table 4. Prevalence of viral suppression and viral non-suppression (N = 397).

The results of the bivariate analysis are presented in Table 5. Treatment interruptions (OR = 2.68; 95% CI: 1.71 - 4.20; p < 0.001), missed clinic appointments (OR = 2.31; 95% CI: 1.42 - 3.77; p = 0.001), low family support (OR = 1.94; 95% CI: 1.23 - 3.04; p = 0.004), and moderate-to-severe depressive symptoms (PHQ-9 ≥ 10) (OR = 2.03; 95% CI: 1.29 - 3.18; p = 0.002) were significantly associated with viral non-suppression.

Table 5. Bivariate analysis of factors associated with viral non-suppression.

Significant associations were also observed for perceived HIV-related stigma (OR = 1.63; 95% CI: 1.05 - 2.54; p = 0.028), financial difficulties related to HIV care (OR = 1.54; 95% CI: 1.00 - 2.37; p = 0.047), substance use (OR = 1.72; 95% CI: 1.03 - 2.87; p = 0.036), low socioeconomic status (OR = 1.59; 95% CI: 1.01 - 2.51; p = 0.043), and clinical suspicion of antiretroviral drug resistance (OR = 2.94; 95% CI: 1.51 - 5.70; p = 0.001). Participation in a peer support group was not significantly associated with viral non-suppression (p = 0.284).

Variables associated with viral non-suppression in the bivariate analysis, together with those considered clinically relevant, were included in a multivariable logistic regression model in Table 6. After adjustment for potential confounding factors, treatment interruptions (aOR = 2.47; 95% CI: 1.48 - 4.11; p < 0.001), missed clinic appointments (aOR = 1.96; 95% CI: 1.11 - 3.44; p = 0.020), low family support (aOR = 1.79; 95% CI: 1.06 - 3.01; p = 0.029), and moderate-to-severe depressive symptoms (PHQ-9 ≥ 10) (aOR = 1.88; 95% CI: 1.10 - 3.19; p = 0.021) remained independently associated with viral non-suppression.

The final model demonstrated good calibration (Hosmer–Lemeshow goodness-of-fit test: p = 0.64). None of the other variables retained an independent association with viral non-suppression after adjustment.

Table 6. Final multivariable logistic regression model of factors associated with viral non-suppression.

The variables low socioeconomic status and clinical suspicion of antiretroviral drug resistance were entered into the initial multivariable model but were not retained in the final model following the variable selection procedure.

This study has several strengths. These include the exhaustive inclusion of all eligible participants at a national HIV referral center, the use of an objective biological outcome (viral load) as the primary endpoint, the application of standardized instruments to assess moderate-to-severe depressive symptoms (PHQ-9) and perceived social support, and the use of multivariable logistic regression to control for potential confounding factors.

Nevertheless, several limitations should be acknowledged. First, the cross-sectional design precludes establishing causal relationships between the identified factors and viral non-suppression. Second, because the study was conducted at a single referral center, the generalizability of the findings to other settings may be limited. Third, the psychosocial variables were self-reported and therefore may be subject to recall bias and social desirability bias. Finally, the possibility of residual confounding due to unmeasured variables cannot be completely excluded.