Viral load suppression (VLS) remains suboptimal among adolescents and youth living with HIV (AYLWH) in Malawi. Depressive symptoms may influence adherence and HIV outcomes, but evidence among AYLWH in Malawi is limited. This study used Malawi Population-based HIV Impact Assessment (MPHIA) 2020–2021 data to assess the association between depressive symptoms and VLS in AYLWH aged 15–24 years.
This is a cross-sectional secondary analysis of MPHIA 2020–2021. VLS was defined as HIV RNA <200 copies/mL. Depressive symptoms were screened using the Patient Health Questionnaire-2 (PHQ-2), with a score of 3 or higher indicating depressive symptoms.
Survey-weighted logistic regression models were used to estimate associations between depressive symptoms and VLS. Multivariable models adjusted for: sex, education, residence, household wealth, alcohol use, participation in Teen Club activities, and awareness of HIV-positive status. All analyses and adjustments reported are those presented by the study authors.
The analytical sample comprised 201 participants; viral-load results were available for 200 individuals. Among those with viral-load data, 137 (68.5%) achieved viral load suppression (<200 copies/mL). Six participants (3.0%) screened positive for depressive symptoms by PHQ-2 (score ≥3).
In univariable analysis, depressive symptoms were not significantly associated with VLS (odds ratio [OR] 0.8, 95% confidence interval [CI] 0.1–4.6; p = 0.80).
In multivariable, survey-weighted logistic regression adjusting for the prespecified covariates, depressive symptoms remained not significantly associated with VLS (adjusted OR [aOR] 0.3, 95% CI 0.1–2.4; p = 0.28). The authors emphasize that the low prevalence of depressive symptoms and limitations of the PHQ-2 instrument reduced precision around the estimate.
Two variables were significantly associated with lower odds of achieving VLS in the adjusted models:
Urban residence: aOR 0.2, 95% CI 0.1–0.5; p = 0.002. According to the authors, urban residence was associated with reduced odds of VLS compared with rural residence.
Unawareness of HIV-positive status: aOR 0.1, 95% CI 0.02–0.1; p < 0.001. Participants who were unaware of their HIV-positive status had substantially lower odds of VLS.
These associations remained after adjustment for the other covariates listed above.
The analysis did not find a statistically significant association between self-reported depressive symptoms (PHQ-2 ≥3) and viral load suppression among AYLWH in Malawi. However, the study authors note key limitations that affect interpretation:
The prevalence of depressive symptoms in the sample was low (3.0%), which reduces statistical power to detect an association and yields imprecise effect estimates.
The PHQ-2 is a brief screening tool with known limitations for identifying depression; reliance on PHQ-2 alone may underdetect clinically relevant depressive disorders.
As a cross-sectional secondary analysis, causality cannot be inferred, and temporality between depressive symptoms and viral suppression cannot be established.
Despite the lack of a detected association in this sample, the findings draw attention to other important correlates of VLS, particularly awareness of HIV-positive status and urban residence, which were associated with lower odds of suppression.
The authors interpret the results as supportive of programmatic emphasis on strengthening HIV-status disclosure and addressing context-specific barriers to care among adolescents and youth living with HIV in Malawi.
The secondary analysis received no specific funding. The authors reported no competing interests. Ethical approval for the work was provided by the Human Research Ethics Committee of the Faculty of Health Sciences, University of Cape Town (HREC ref 671/2024). The de-identified participant-level data used are held by the Malawi Population-based HIV Impact Assessment (MPHIA) 2020–2021 and are available to researchers via the PHIA Data Manager subject to registration and data-access procedures. The authors stated they do not own the underlying participant-level dataset; the secondary-analysis protocol and R code are available from the corresponding author on reasonable request.
In this MPHIA 2020–2021 sample of AYLWH aged 15–24 years in Malawi, depressive symptoms (PHQ-2 ≥3) were uncommon and were not significantly associated with viral load suppression, but estimates were imprecise due to low prevalence and PHQ-2 limitations.
Unawareness of HIV-positive status and urban residence were independently associated with lower odds of VLS, highlighting targets for intervention: improving status disclosure and addressing urban-specific barriers to adherence and retention in care.
Further research using more sensitive measures of depression, larger samples with higher prevalence of depressive symptoms, and longitudinal designs would better clarify the role of mental health in VLS among AYLWH.