USE OF AI CHATBOTS IN INFLUENCING MENTAL HEALTH STATUS AMONG UNIVERSITY UNDERGRADUATES IN BENIN CITY, EDO STATE
Faculty
Year of Publication
Publication Type
Abstract
BACKGROUND
Mental health disorders among university students are an increasing public health concern globally, particularly in low- and middle-income countries where access to formal mental health services remains limited. Simultaneously, artificial intelligence (AI) chatbots are increasingly being integrated into students’ academic and social activities, with emerging interest in their potential role in mental health support. This study aimed to assess the knowledge, attitudes, uptake, utilization, factors influencing use, and mental health status associated with AI mental health chatbot use among undergraduate students of the University of Benin, Benin City, Edo State, Nigeria.
METHODS
An analytical cross-sectional study was conducted among 436 undergraduate students of the University of Benin, Benin City, Edo State, Nigeria, using a pretested self-administered structured questionnaire. Respondents were selected using a multistage sampling technique. Data collected were analyzed using IBM SPSS version 25.0. Statistical significance was set at p < 0.050 at 95% confidence interval.
RESULTS
The mean age of respondents studied was 21.84 ± 3.97 years. Nearly all respondents demonstrated awareness of AI chatbots, while approximately nine-tenths had good overall knowledge. However, awareness of clinically validated mental health–specific chatbots such as Woebot and Wysa was very low. About three-quarters of respondents demonstrated positive attitudes towards AI mental health chatbots. Uptake of AI chatbots was near-universal (96.6%), driven predominantly by general-purpose platforms such as ChatGPT for academic purposes. Uptake of clinically validated mental health-specific chatbots such as Woebot and Wysa was negligible. Only a small proportion reported using AI chatbots specifically for emotional support or mental health-related purposes. Ethnicity and level of study were identified as significant predictors of good knowledge of AI chatbots. Respondents with good knowledge had significantly higher odds of positive attitudes towards AI mental health chatbots (OR = 4.003; CI = 1.940–8.258; p < 0.001). Peer influence, anonymity, affordability, and privacy concerns significantly influenced AI chatbot utilization. High utilization was significantly associated with academic level and religion. Nearly three-fifths (59.9%) of respondents screened positive for depression. High AI chatbot utilization (OR = 1.753; CI = 1.083–2.836; p = 0.022) and high dependency (OR = 2.173; CI = 1.039–4.542; p = 0.039) were identified as significant predictors of depression.
CONCLUSION
Despite high awareness, positive attitudes, and near-universal uptake of AI chatbots among undergraduate students, awareness of clinically validated mental health–specific platforms remain low. Depression was highly prevalent among respondents, and high AI chatbot utilization and dependency were significantly associated with depressive symptoms. There is need for targeted digital mental health literacy programmes, institutional regulation of AI mental health tools, and integration of safe, evidence-based digital mental health interventions within university settings.
Mental health disorders among university students are an increasing public health concern globally, particularly in low- and middle-income countries where access to formal mental health services remains limited. Simultaneously, artificial intelligence (AI) chatbots are increasingly being integrated into students’ academic and social activities, with emerging interest in their potential role in mental health support. This study aimed to assess the knowledge, attitudes, uptake, utilization, factors influencing use, and mental health status associated with AI mental health chatbot use among undergraduate students of the University of Benin, Benin City, Edo State, Nigeria.
METHODS
An analytical cross-sectional study was conducted among 436 undergraduate students of the University of Benin, Benin City, Edo State, Nigeria, using a pretested self-administered structured questionnaire. Respondents were selected using a multistage sampling technique. Data collected were analyzed using IBM SPSS version 25.0. Statistical significance was set at p < 0.050 at 95% confidence interval.
RESULTS
The mean age of respondents studied was 21.84 ± 3.97 years. Nearly all respondents demonstrated awareness of AI chatbots, while approximately nine-tenths had good overall knowledge. However, awareness of clinically validated mental health–specific chatbots such as Woebot and Wysa was very low. About three-quarters of respondents demonstrated positive attitudes towards AI mental health chatbots. Uptake of AI chatbots was near-universal (96.6%), driven predominantly by general-purpose platforms such as ChatGPT for academic purposes. Uptake of clinically validated mental health-specific chatbots such as Woebot and Wysa was negligible. Only a small proportion reported using AI chatbots specifically for emotional support or mental health-related purposes. Ethnicity and level of study were identified as significant predictors of good knowledge of AI chatbots. Respondents with good knowledge had significantly higher odds of positive attitudes towards AI mental health chatbots (OR = 4.003; CI = 1.940–8.258; p < 0.001). Peer influence, anonymity, affordability, and privacy concerns significantly influenced AI chatbot utilization. High utilization was significantly associated with academic level and religion. Nearly three-fifths (59.9%) of respondents screened positive for depression. High AI chatbot utilization (OR = 1.753; CI = 1.083–2.836; p = 0.022) and high dependency (OR = 2.173; CI = 1.039–4.542; p = 0.039) were identified as significant predictors of depression.
CONCLUSION
Despite high awareness, positive attitudes, and near-universal uptake of AI chatbots among undergraduate students, awareness of clinically validated mental health–specific platforms remain low. Depression was highly prevalent among respondents, and high AI chatbot utilization and dependency were significantly associated with depressive symptoms. There is need for targeted digital mental health literacy programmes, institutional regulation of AI mental health tools, and integration of safe, evidence-based digital mental health interventions within university settings.
Supervisor(s)
co-supervisor