Artificial intelligence is helping to analyse and improve healthcare systems - from waiting times to treatment outcomes and costs. However, some of the most important aspects of care remain harder to measure - such as whether people feel heard, understood and genuinely cared for.
AWARE_AI, an innovative AI program designed at Flinders University, uses machine learning to investigate how human experiences of healthcare are connected to wellbeing.
The program offers a new way to explore whether conditions in which care is delivered, the relationships between patients and healthcare professionals, and the way care is brought together are associated with people's wellbeing.
Lead researcher Dr Maria Alejandra Piñero de Plaza, from the Caring Futures Institute at Flinders University, says the goal is to make the "human dimensions of healthcare more visible, measurable and better understood".
"Healthcare is fundamentally human, but many of the things we routinely measure tell us more about how services perform than how people experience being cared for," says Dr Piñero de Plaza, from the College of Health and Enablement.
"Waiting times, clinical results and costs are important, but they cannot tell us the whole story. We also need to understand whether people feel listened to, whether their needs are recognised and how these experiences relate to their wellbeing."
The latest research, published in the Journal of Advanced Nursing and conducted with experts in Australia, Denmark, Spain and Colombia, builds on long-running investigations into the 'Fundamentals of Care', an established framework explaining how good healthcare depends on three connected elements: a supportive care environment, trusting relationships between patients and clinicians, and care that brings together people's physical, emotional and social needs.
Earlier research in Australia and two larger studies in Spain, each involving more than 1,000 participants, provided evidence about how these aspects of care are connected.
AWARE_AI takes the next step, examining how these relationships extend to people's self-reported wellbeing.
Using survey responses from 36 Australian adults, the researchers found that more positive experiences of integrated, person-centred care were associated with more positive assessments of wellbeing, particularly how satisfied people felt with their lives.
"We already had evidence that the healthcare environment, the relationship between patients and clinicians, and the way care is integrated are closely connected," Dr Piñero de Plaza says. "What makes this study exciting is that we can now begin to investigate how these connections extend beyond the experience of receiving care to how people evaluate their lives.
"That opens an important question for healthcare: Could understanding and improving the human experience of care also help us understand changes in people's wellbeing?"
Unlike AI systems designed to diagnose diseases or predict medical emergencies, AWARE_AI uses established statistical and machine-learning methods to investigate relationships within complex healthcare experiences.
Its analysis is designed to be transparent and reproducible, allowing researchers to examine how findings were obtained and test them in future studies.
The small Australian study demonstrated that the approach was feasible and reproducible, although it could not reliably predict an individual person's wellbeing. The findings show associations, not proof that one aspect of care causes another or directly improves wellbeing.
Study co-author Dr Fernando Marmolejo Ramos says recognising these limits is part of using AI responsibly.
"Artificial intelligence can help us identify relationships that deserve closer investigation, but finding a pattern is not the same as proving what causes it," says Dr Ramos, Senior Lecturer in Psychology at Flinders University.
"By making the analysis transparent and testing what the methods can and cannot do, we can build stronger evidence for future research."
The next step is to test AWARE_AI in larger studies that follow people over time, examining whether changes in their experiences of care are accompanied by changes in wellbeing.
Dr Piñero de Plaza believes this could eventually help healthcare services evaluate important aspects of care alongside the clinical and operational measures they already use.
"Imagine being able to assess not only whether patients received the right treatment, but also whether they felt understood, whether their needs were met and how those experiences related to their wellbeing," she says.
"We are not trying to make healthcare more human. Healthcare is already human. We are trying to make its human dimensions more visible, measurable and better understood.
"Artificial intelligence should help us understand and improve the care people receive, not replace the relationships that make care possible."
The paper - 'AWARE_AI, a machine-learning-based reproducible workflow modelling the Fundamentals of Care framework and patient-reported wellbeing: a cross-sectional feasibility study' (2026) by Maria Alejandra Piñero de Plaza, Alison Kitson (Aalborg University, Denmark), Ana María Porcel-Gálvez (University of Seville, Spain), Renuka Visvanathan (Adelaide University), Alline Beleigoli (Flinders), Carlos Javier Barrera Causil (Instituto Tecnológico de Colombia, Bogotá), Fernando Marmolejo Ramos (Flinders University) and Regina Allande-Cussó (Flinders and University of Seville) - has been published in the Journal of Advanced Nursing (Wiley). DOI: 10.1111/jan.70800.
Acknowledgements: The study was supported by the Caring Futures Institute at Flinders University.
