Effect of AI-Assisted Mental Health Screening Tools in Educational Institutions

Abstract

A mental health crisis among university and college students is a growing issue in personal health, and the rates of anxiety, depression, and psychological distress remain high and regularly reported. In college students, the utilisation of mental health services has increased over the past ten years, and increases in outpatient, inpatient, and emergency units have been observed (Lipson et al., 2019). The implementation of mental health services in education through artificial intelligence (AI) is a prospective opportunity for scalable, timely, and accessible early intervention. The research question addressed in the current study is as follows: What is the impact of AI-aided mental health screening tools implemented in three universities of the United Kingdom over 12 months, and how may they influence the overall mental health outcomes of students? Quantitative data were gathered using validated measures such as the Patient Health Questionnaire-9 (PHQ-9), the Generalised Anxiety Disorder scale (GAD-7), and the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS) through a concurrent mixed-methods research design involving 342 undergraduate and postgraduate students. Qualitative data were collected through semi-structured interviews with 24 mental health professionals and academic employees. The AI screening instruments showed a mean accuracy of 87.3% and statistically significant improvements in all measured wellbeing outcomes after implementation (p < .001). The most common qualitative themes highlighted benefits related to accessibility and reduced stigma, while also identifying concerns regarding data privacy, algorithm bias, and the inability to recreate the therapeutic relationship between a human and an algorithm. The results indicate that AI-supported technologies can significantly contribute to conventional mental-health support frameworks when supported by ethical and adequate governance frameworks.