Psychosocial Distress Screening in Key Populations Living with HIV Starting ART in Nigeria
Domain:
healthcareRecord type:
paper
Creator:
EroAbiWinKen
Publisher:
Host:
Background: Psychosocial distress (PSD) is an important determinant of HIV outcomes among key populations (KPs), yet programme evidence from Nigeria remains limited on how distress burden varies across geography and service-delivery pathways. We aimed to describe psychosocial distress screening positivity among key populations newly initiating antiretroviral therapy in Nigeria and to examine how geography, care pathways, and individual-level characteristics contributed to its distribution.
Methods: We did a cross-sectional analysis of routinely collected data from 15 One-Stop Shop (OSS) facilities in four Nigerian states between Oct 1, 2021, and June 28, 2024. We first described screening uptake in the full programme cohort and analysed psychologist-administered screens among adults newly initiated on antiretroviral therapy using prespecified CatBoost models to distinguish geographic and service-delivery patterning from baseline explanatory correlates. Performance was assessed with ROC-AUC, precision-recall AUC, and Brier score, and feature contributions were examined with SHAP values.
Findings: Among 69,527 people living with HIV enrolled in care, 15,577 (22.4%) completed mental-health screening (MHS). Of the 3,657 participants who underwent psychologist-administered screening, 1,143 (31.3%) screened positive for psychosocial distress. The full predictive model showed excellent discrimination and calibration (ROC-AUC 0.95 [95% CI 0.94-0.97]; PR-AUC 0.91 [0.86-0.95]; Brier score 0.07 [0.06-0.08]). A model including only LGA of residence also performed strongly (ROC-AUC 0.90), whereas the baseline-only explanatory model showed more modest discrimination (ROC-AUC 0.75 [0.71-0.79]), indicating that PSD screening positivity was strongly shaped by geographic and service-delivery context.
Interpretation: PSD was common among KPs initiating ART in this large Nigerian HIV programme and was strongly patterned by geographic and service-delivery context. These findings support stronger integration and standardisation of MHS and referral within OSS services and suggest that routine programme data can help identify where psychosocial needs are most concentrated. The findings are most directly applicable to programme implementation and targeted service planning.
Methods: We did a cross-sectional analysis of routinely collected data from 15 One-Stop Shop (OSS) facilities in four Nigerian states between Oct 1, 2021, and June 28, 2024. We first described screening uptake in the full programme cohort and analysed psychologist-administered screens among adults newly initiated on antiretroviral therapy using prespecified CatBoost models to distinguish geographic and service-delivery patterning from baseline explanatory correlates. Performance was assessed with ROC-AUC, precision-recall AUC, and Brier score, and feature contributions were examined with SHAP values.
Findings: Among 69,527 people living with HIV enrolled in care, 15,577 (22.4%) completed mental-health screening (MHS). Of the 3,657 participants who underwent psychologist-administered screening, 1,143 (31.3%) screened positive for psychosocial distress. The full predictive model showed excellent discrimination and calibration (ROC-AUC 0.95 [95% CI 0.94-0.97]; PR-AUC 0.91 [0.86-0.95]; Brier score 0.07 [0.06-0.08]). A model including only LGA of residence also performed strongly (ROC-AUC 0.90), whereas the baseline-only explanatory model showed more modest discrimination (ROC-AUC 0.75 [0.71-0.79]), indicating that PSD screening positivity was strongly shaped by geographic and service-delivery context.
Interpretation: PSD was common among KPs initiating ART in this large Nigerian HIV programme and was strongly patterned by geographic and service-delivery context. These findings support stronger integration and standardisation of MHS and referral within OSS services and suggest that routine programme data can help identify where psychosocial needs are most concentrated. The findings are most directly applicable to programme implementation and targeted service planning.
Similar
Prevalence and factors associated with psychological distress among key populations in NigeriaThe bottleneck is the programme: "when, not whom" in the retrospective acceptability of community-based ART initiation among criminalised key populations living with HIV in CameroonTemporal relationships between distress and pain in people living with HIVNavigating technology-facilitated abuse: A participatory action research study with young adults living with HIV and key populations in KenyaSocio-behavioral research INFLUENCE OF RELIGIOSITY AND LOCUS OF CONTROL ON ADHERENCE TO ANTIRETROVIRAL DRUGS AMONG KEY POPULATIONS LIVING WITH HIV IN NIGERIAStigma Toward People Living with HIV and Key Populations Among Healthcare Workers in Northwest Cameroon: A Cross-Sectional Study