Artificial intelligence–driven remote depression detection as a socioeconomic intervention: implications for food and job security in North-West Nigeria

Abstract

Depression remains a major public health challenge with substantial socioeconomic implications, particularly in low-resource settings where early detection and treatment gaps persist. This study developed and evaluated an artificial intelligence (AI)–driven remote depression detection system integrated with a structured mental health intervention framework and socioeconomic impact assessment in North-West Nigeria. A total of 412 adults aged 18–60 from six rural and peri-urban communities participated in the study. Using a quasi-experimental design with treatment and control groups, data were collected over a 16-week period comprising baseline assessment, AI-driven monitoring, targeted intervention, and follow-up evaluation. Ethical approval was obtained from the Nigerian National Health Research Ethics Committee (NHREC), and all participants provided informed consent. A machine-learning model trained on multi-modal behavioural and survey data demonstrated strong predictive performance, achieving 88.4% accuracy (95% CI: 85.2–91.1), 0.86 precision, 0.91 recall, 0.88 F1-score, and an ROC-AUC of 0.92. Following AI-triggered intervention, mean depression scores decreased by 37% (p < 0.001), with 64% of high-risk participants transitioning to lower-risk categories. Socioeconomic outcomes improved significantly, including a 21% increase in productivity, 18% improvement in employment stability, 24% reduction in absenteeism, 14% rise in monthly income (p < 0.05), and 19% reduction in moderate-to-severe food insecurity. Difference-in-Differences analysis confirmed that the treatment group experienced a 32% greater reduction in depression and a 26% improvement in composite welfare relative to controls. These findings validate the impact pathway linking mental health improvement to enhanced productivity, income growth, and household welfare.  The study demonstrates that AI-driven remote depression detection, when combined with timely intervention, can yield significant psychological and economic benefits. The results position AI-enabled mental health screening as a scalable, cost-effective policy tool for strengthening human capital and reducing poverty in underserved regions.