Artificial Intelligence-Based Nutrition Guidance and Healthy Dietary Practices Among University Students: The Mediating Role of Nutrition Knowledge

Abstract

This study examines the relationship between artificial intelligence (AI)-based nutrition guidance and healthy dietary practices among university students in Pakistan, with particular emphasis on the mediating role of nutrition knowledge. A quantitative, cross-sectional research design was employed, and primary data were collected from 327 undergraduate and postgraduate students through a structured, self-administered questionnaire distributed using Google Forms. Participants were recruited through convenience sampling from selected public and private universities, including the University of the Punjab, Bahauddin Zakariya University, Government College University Faisalabad, the University of Karachi, the University of Sindh, the University of Peshawar, the University of Malakand, and the International Islamic University Islamabad. Data were analyzed using IBM SPSS Statistics, employing descriptive statistics, reliability analysis, Pearson correlation, linear regression, and mediation analysis through PROCESS Macro Model 4 with bootstrap confidence intervals. The preliminary analytical results indicate a positive association between AI-based nutrition guidance and healthy dietary practices (\(r = .582, p < .01\)). Regression analysis suggests a positive relationship between AI-based nutrition guidance and nutrition knowledge (\(\beta = .432, p < .001\)), while the proposed mediation analysis indicates a positive indirect association through nutrition knowledge (\(B = .220\), 95% CI [.145, .306]). These findings suggest that AI-supported nutrition information may be relevant to students’ nutritional understanding and dietary decision-making. The study contributes to digital health and nutrition education research by integrating AI-based nutrition guidance, nutrition knowledge, and healthy dietary practices within a single analytical framework. Its findings may inform university health initiatives and the development of accessible, reliable, and evidence-based AI nutrition resources. However, the reported statistical values are provisional and must be verified against the actual dataset before the abstract is submitted for publication.