AI-DRIVEN DECISION SUPPORT SYSTEMS IN PUBLIC HEALTH ADMINISTRATION

Abstract

Background: The evolution of Artificial Intelligence (AI) has made it possible to create decision support systems to aid decision-making on timely, verifiable, and fact-based issues of immense importance and accuracy for public health administrators. Even with such interest, there are gaps in scholarly literature concerning the adoption, reliability, validity, and demographics of such systems. Objective: This research aims to analyze the adoption rate, the use of, and the reliability and validity of the decision support systems within the domain of public health, while assessing the demographic and professional differences in their use, and determining the most important predictors for their future adoption. Methods: This research was conducted with the use of quantitative cross-sectional methods to analyze the responses of 314 individuals in the public health domain, holding such positions as policy development, public health practice, healthcare provision, IT and AI analytics, and health data analysis. The participants answered a 20-item questionnaire, designed on a 5-point Likert scale that describes systems as a domain of public health and issues on supportive technologies embedded in public health (usefulness, ease of use, trust, organizational impact, and future readiness). Descriptive statistical framework systems, such as normality check, reliability Prism analysis (Cronbach’s Alpha) for community organization, validity (KMO and Bartlett’s test), and demographic computing (Independent Samples t-test, One-Way ANOVA, Kruskal–Wallis, Chi-Square test, Pearson correlation, and regression analysis) were employed. Results: The findings confirmed that the dataset was normally distributed with strong reliability (Cronbach’s Alpha = 0.87) and validity (KMO = 0.82; Bartlett’s χ² = 245.67, p < 0.001).  Group comparison tests showed differences between the groups across gender, roles, and education levels. The inter-item correlation was strong and positive, while regression analysis showed that the most significant predictor of future adoption was trust and transparency (β = 0.42, p < 0.001), followed by perceived usefulness and organizational impact.

Conclusion: AI DS DSSs are positively perceived and are expected to enhance the effectiveness of public health administration. The major barriers to adoption are demographic differences, trust, and lack of transparency. Demographic and trust barriers can be overcome with training and role-based approaches aimed at trust building, and they are essential for the successful adoption, sustainable future integration, and public health system performance of AI.