Making patient-reported outcomes reusable at scale

AI Summary

PROs are essential for patient-centred care, digital health and AI but remain hard to reuse because of inconsistent implementation and measurement heterogeneity.
The central problem is absence of a coherent, end-to-end approach linking measurement, capture, standards and analytic reuse across disciplines and governance.
Adopt a practical four-step FAIR pathway: define outcome domains and questionnaires; capture structured, computable PROs; store analysable data; enable cross-questionnaire linking.

TL;DR (AI Generated)

Yet, PROs remain difficult to reuse at scale: their implementation is inconsistent across health-care settings, measurement approaches are highly heterogeneous, and PROs are rarely available as large, standardised, analysable datasets suitable for multisite analyses and advanced analytics, including artificial intelligence models.

A shared FAIR-PROs approach would allow PROs to become a routine, reusable component of learning health systems and a driver of both person-centred care and data-driven innovation.

Basic summary

Lancet Digit Health. 2026 Nov 13:101089. doi: 10.1016/j.landig.2026.101089. Online ahead of print.

ABSTRACT

Patient-reported outcomes (PROs) are essential to understanding how people live with illness, treatment, and care, and are increasingly promoted for routine use in clinical practice, quality improvement, and value-based health care. PROs are also increasingly relevant in digital health and artificial intelligence applications, both as inputs to predictive models and as outputs reflecting patient-centred aspects of health. Yet, PROs remain difficult to reuse at scale: their implementation is inconsistent across health-care settings, measurement approaches are highly heterogeneous, and PROs are rarely available as large, standardised, analysable datasets suitable for multisite analyses and advanced analytics, including artificial intelligence models. In this Viewpoint, we argue that the central challenge is not a lack of initiatives, but the absence of a coherent, end-to-end approach that connects measurement, routine capture, data standards, and analytic reuse. We propose a practical four-step pathway towards Findable, Accessible, Interoperable, and Reusable (FAIR) PROs. This pathway focuses on (1) defining the outcome domains that matter and selecting high-quality questionnaires; (2) capturing PROs consistently at the point of care in structured, computable formats; (3) storing them in analysable, shareable data models; and (4) establishing transparent cross-questionnaire linking to enable comparable scores. None of these steps poses major technical challenges in isolation; the primary difficulty is that these efforts are often developed in silos, with minimal coordination across disciplines, from questionnaire development and validation to clinical workflows, informatics standards, analytic infrastructures, and governance. A shared FAIR-PROs approach would allow PROs to become a routine, reusable component of learning health systems and a driver of both person-centred care and data-driven innovation.

PMID:43026136 | DOI:10.1016/j.landig.2026.101089

Document this CPD