Health systems are increasingly called to be equitable, intelligent, and person-centred. Yet the models used to evaluate them remain stubbornly reductionist. In our collaborative research across cardiac rehabilitation, aged care, digital health, and AI-enabled triage, we have consistently found that traditional evaluation tools, particularly utility-based patient-reported outcome measures (PROMs), fail to reflect the multilevel, relational, and lived experience of care. We argue that to meet the needs of diverse populations and complex systems, evaluation must evolve toward biographical, participatory, and predictive models that work across individual, organisational, and policy levels.

From validated PROMs to constrained insight

In our systematic review of utility-based patient-reported outcome measures (PROMs) in cardiac rehabilitation (Bulamu et al., 2024), we identified nine PROMs, with only three [the EuroQol 5-Dimension 5-Level (EQ-5D-5L), 12-Item Short Form Survey (SF-12), and MacNew Heart Disease Health-Related Quality of Life Questionnaire (MacNew)] showing sufficient evidence for responsiveness and hypothesis testing. These findings were limited to German-language versions and populations, restricting generalisability. Moreover, while the tools were mapped to the International Classification of Functioning, Disability and Health (ICF) and the International Consortium for Health Outcomes Measurement (ICHOM) domains for cardiovascular disease, their capacity to reflect social, cultural, or relational dimensions of care was limited. As our results highlighted, PROMs remain indispensable in cost-effectiveness analysis but insufficient for understanding how care unfolds, adapts, or is experienced.

Care is more than an outcome: The CLCT and Care Biography

The Caring Life Course Theory (CLCT) is used to address this, a framework that reconceptualises care through several interrelated constructs, e.g., care capability, care transitions, care networks, and care trajectories (Pinero de Plaza et al., 2024a). In our study of 15 cardiac rehabilitation programs across rural Australia, we applied CLCT to demonstrate how care participation and engagement were shaped more by life transitions and social support than clinical indicators alone. This work calls for an evaluative shift from viewing care as a discrete intervention to understanding it as part of a broader, biographical system.

The concept of Care Biography further supports this shift. Defined and examined in our concept analysis (Tieu et al., 2024), care biography reflects the cumulative life events and interpersonal relationships that shape how individuals experience and engage with care. As a research and practice tool, it helps clinicians and systems understand people not as passive recipients of services but as active agents whose experiences of care are profoundly social and temporal.

Modelling complexity: Fundamental care as a structural system

We advanced this perspective through a pilot structural equation modelling (SEM) study grounded in the Fundamentals of Care Framework (Pinero de Plaza et al., 2025). We used a partial least squares model (cPLS-SEM) to analyse patient-reported experience data across healthcare settings in Spain and Australia. The results revealed that the Context of Care significantly predicted the clinician–patient Relationship, which in turn fully mediated the Integration of Care, highlighting how systemic and relational factors interact to produce person-centred outcomes in diverse healthcare systems.

Importantly, this work operationalises care relationships and shows how fundamental elements of care, long considered intangible, can be statistically modelled and meaningfully interpreted across settings.

From measurement to meaning: PROLIFERATE and PROLIFERATE_AI

To support real-world evaluation and transformation, we developed PROLIFERATE, a multilevel participatory framework that captures stakeholder priorities, perceptions, and behavioural patterns while implementing health innovations (Pinero de Plaza et al., 2023). PROLIFERATE enables mapping readiness for change across stakeholder roles (e.g., policymakers, clinicians, patients), measuring dimensions such as comprehension, emotion, motivation, barriers, and optimisation strategies.

In evaluating digital and AI-based innovations, including the RAPIDx AI clinical decision support tool, we extended this model into PROLIFERATE_AI, a human-centred machine learning-enabled method (Pinero de Plaza et al., 2025b). Our work across 12 Australian emergency departments revealed that user trust, emotional response, and local leadership significantly shaped AI adoption, factors often missed by technical validations. PROLIFERATE_AI allowed us to simulate future implementation patterns based on stakeholder data, demonstrating how predictive modelling can be ethical and practical when grounded in participatory logic.

Responding to complexity and post-truth challenges

These findings speak to the expanded debate about the role of knowledge, complexity, and co-design in health systems reform. In our studies of transdisciplinary collaboration (Lawless et al., 2024; Archibald et al., 2023) and post-truth policy dynamics (Tieu et al., 2023), we showed how scientific knowledge often competes with political and cultural narratives. In this context, overly narrow evaluation methods can distort policy decisions or marginalise community priorities.

We argue that a new evaluative paradigm is needed, one that is:

• Multilevel, linking individual, organisational, and policy insights

• Biographical, informed by people’s care histories and networks

• Predictive, capable of informing future practice through real-time feedback

• Participatory, co-designed with those affected by interventions

• Transdisciplinary, integrating care theory, systems science, implementation research, and health informatics

A call for meaningful metrics

If health systems are to become adaptive and person-centred, we must evolve beyond what is easy to measure. Traditional PROMs, while necessary, are insufficient. The next generation of evaluation must honour complexity and care simultaneously. Our work demonstrates that such models are not only conceptually robust but also operational, scalable, and already in use.

As researchers, we call on funders, policymakers, and health leaders to support evaluation methods that reflect real-world conditions of care, which can navigate complexity without erasing it and foreground human experience without sacrificing analytical rigour.

References

Bulamu et al. (2024). Measurement properties of utility-based health-related quality of life measures in cardiac rehabilitation and secondary prevention programs: A systematic review. Quality of Life Research, 33, 2299–2320. https://doi.org/10.1007/s11136-024-03657-5

Pinero de Plaza et al. (2024a). The Caring Life Course Theory: Opening new frontiers in care—A cardiac rehabilitation example. Journal of Advanced Nursing. https://doi.org/10.1111/jan.16312

Tieu et al. (2024). Care biography: A concept analysis. Nursing Philosophy. https://doi.org/10.1111/nup.12489

Pinero de Plaza et al. (2025). Piloting Structural Equation Modeling for Fundamental Care Decision-Making Across Healthcare Settings. Research Square Preprint. https://doi.org/10.21203/rs.3.rs-6365026/v1

Pinero de Plaza et al. (2023). Co-designing, measuring, and optimizing innovations and solutions within complex adaptive health systems: The PROLIFERATE method. Frontiers in Health Services, 3, 1154614. https://doi.org/10.3389/frhs.2023.1154614

Pinero de Plaza et al. (2025b). Human-centred AI for emergency cardiac care: Evaluating RAPIDx AI with PROLIFERATE_AI. International Journal of Medical Informatics, 196, 105810. https://doi.org/10.1016/j.ijmedinf.2025.105810

Lawless et al. (2024). From promise to practice: How health researchers understand and promote transdisciplinary collaboration. Qualitative Health Research, 34(2). https://doi.org/10.1177/10497323241235882

Archibald et al. (2023). How transdisciplinary research teams learn to do knowledge translation. Health Research Policy and Systems, 21, 20. https://doi.org/10.1186/s12961-023-00967-x

Tieu et al. (2023). Wicked problems in a post-truth political economy: A dilemma for knowledge translation. Humanities and Social Sciences Communications, 10, 83. https://doi.org/10.1057/s41599-023-01789-6