Challenges of AI for cost-effectiveness in health: a hybrid MCDM model

This study aims to elucidate the role of artificial intelligence (AI) in achieving cost-effectiveness within healthcare services, and to address the challenges that hinder its integration.

A hybrid multi-criteria decision-making model based on the DEMATEL and ARAS methods was designed. Data were collected with the participation of ten experts working in the fields of health management, health economics and health policy in Canada, India, Switzerland, USA and Türkiye. The causal relationships between the challenges were analyzed using DEMATEL analysis and the solutions to these challenges were evaluated using the ARAS method.

As a result of the DEMATEL analysis, the "Regulatory and ethical concerns (REC)" criterion reached the highest level of importance (13.94%), followed by "High initial costs and financial constraints (ICFC)" (13.65%) and "Lack of data standardization and biases (DBHQ)" (13.29%). In the ARAS evaluation, among the proposed solutions, "Carrying out comprehensive awareness-raising activities for patients and their relatives (SS3)" achieved the highest performance value (Ki = 0.93), followed by "Providing inclusive incentives for AI transformation (SS5)" (Ki = 0.87) and "Establishing a blockchain-based cybersecurity system (SS2)" (Ki = 0.85).

To achieve cost-effectiveness through AI in healthcare, it is imperative to establish a robust, integrated, and standardized data infrastructure, develop comprehensive ethical and regulatory frameworks, and implement patient education programs to foster trust and acceptance of AI technologies. This study provides policymakers and practitioners with evidence-based recommendations to facilitate the successful integration of AI in healthcare, ultimately enhancing cost-efficiency and service delivery.