Role of Artificial Intelligence in Advancing Value-Based Healthcare – Opportunities and Challenges: A Scoping Review

Abstract

Background: Healthcare systems worldwide face an escalating burden of increasing healthcare costs and inconsistent clinical outcomes. Value-based healthcare (VBHC) emerged to bring a fundamental shift in the healthcare sector, focusing on improved patient outcomes and cost-effectiveness to maximise the overall quality of care. The advent of artificial intelligence (AI), including machine learning (ML), deep learning and natural language processing (NLP), offers opportunities to advance VBHC by improving diagnostic accuracy, reducing wasteful expenditure, and personalising care delivery.

Objectives: This scoping review aims to: (i) identify the opportunities associated with AI technology adoption for advancing VBHC, (ii) ascertain the challenges and barriers to AI adoption in VBHC initiatives, and (iii) suggest future directions for overcoming these challenges.

Methods: A scoping review was conducted as per the Joanna Briggs Institute framework and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines. Articles published between 2020 and 2025 were selected for the study from PubMed, Directory of Open Access Journals, Scopus and Web of Science.

Results: The results emphasized the role of AI to advance the VBHC across the three attributes. In the monetary value domain, ML-based predictive modelling, hybrid deep learning models and AI-powered administrative automations showcased noticeable cost savings and resource optimization. In the patient-centred care domain, AI-enabled personalised medicine, integration of patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs), and enhanced patient engagement through virtual health technologies. In the quality of care domain, clinical decision support systems, disease prediction algorithms and AI-powered diagnostic tools improve the accuracy of diagnostics and patient outcomes. However, adoption is obstructed by data fragmentation, workforce resistance, immature governance frameworks, and algorithmic opacity.

Conclusions: Artificial intelligence contains noticeable potential to advance VBHC delivery and implementation. However, to realise this potential, there is an increased need to stress the importance of explainable AI frameworks, investment in AI literacy, effective data governance and national-level regulatory frameworks. Future research should focus on empirical studies, integration of patient-generated data into the ML model development AI equity validation.