The International Classification of Functioning, Disability and Health at 25: From conceptual backbone to computable language in childhood‐onset disability

Twenty-five years ago, the World Health Organization's International Classification of Functioning, Disability and Health (ICF) was endorsed by 191 member states at the 54th World Health Assembly.1 The occasion marked a paradigm shift in how the world understood disability – away from a deficit located within the individual, and toward a dynamic interaction between health conditions, a person's functioning, and the physical, social, and attitudinal environment in which a person lives. The Developmental Medicine & Child Neurology (DMCN) community embraced this shift rapidly and productively. Within years of the ICF's endorsement, researchers in paediatric neurorehabilitation were applying its framework to systematic reviews, outcome measurement, and clinical assessment. Work published in DMCN over the past two decades made clear that developmental change and context are difficult to capture with impairment-led scales alone. Reviews of generic patient-reported outcome measures exposed a recurring conceptual conflation of ‘quality of life’ with ‘health status’, and demonstrated that many instruments over-represent body functions and daily activities while systematically under-representing participation and environmental factors. The development of ICF Core Sets for cerebral palsy, autism spectrum disorder, and attention-deficit/hyperactivity disorder then translated the breadth of the classification into clinically relevant, condition-specific category selections spanning all ICF components.2, 3 Taken together, this body of work has established the ICF as the conceptual backbone of childhood disability science. At 25 years, the field is evolving in ways that both affirm and extend this foundation. The growing emphasis on social and societal dimensions of disability, on the voice of children and their families, on cultural diversity and socioeconomic determinants, and on lifelong trajectories4 – all themes increasingly prominent in DMCN – sits comfortably within the ICF's biopsychosocial architecture. The model always had room for these perspectives; the field has steadily grown into them. What the field now needs to grow into is the ICF's use as a semantic interoperability standard and coding system in the context of rapidly digitalizing health care. This is not a technical footnote – it is the next frontier. Electronic health and social care record system, artificial intelligence (AI)-enabled clinical decision support tools, and health information exchange platforms are becoming the primary infrastructure through which clinical knowledge is generated, shared, and acted upon. If the ICF is embedded within this infrastructure not merely as a conceptual reference but as a computable, coded language operating alongside the International Classification of Diseases, 11th Revision (ICD-11), its influence on real-world care can be even more transformative. If it is not, that influence on real-world care will remain constrained and suboptimal. The ICF 2025 release provides the foundational architecture and digital tooling to make this vision achievable (https://icd.who.int/browse/2026-01/icf/en). In terms of content, it incorporates all accepted cumulative update proposals since 2001 – including those derived from the ICF Children and Youth Version (ICF-CY), giving the ICF full lifespan coverage – and has been integrated together with ICD-11 and the International Classification of Health Interventions (ICHI) into the WHO Family of International Classifications (WHO-FIC) Foundation. For end-users, an updated ICF Browser and ICF Coding Tool are now available in 15 languages, with translations into a further 18 underway. For software developers and system integrators, the ICF has been incorporated into the ICD-11 API, enabling online and offline deployment within electronic health and social care record systems. Ongoing enrichment of the ICF Index is also expanding its natural language processing capacity, opening new possibilities for computer and AI-assisted coding and retrieval. These developments offer concrete opportunities for the childhood disability research and clinical community. Contributing terminology proposals to enrich the ICF Index, refining ICF Core Sets for cerebral palsy and related conditions to address not only category selection and quantification but structured data representation, and advocating for the co-deployment of the ICD-11, ICF, and ICHI as complementary interoperable standards within clinical software – all of these are within reach for this community and the field at large, and all of them matter. The ICF was designed to give the world a common language for health and functioning. Twenty-five years on, the task is to make that language computable – to embed it as a semantic interoperability standard within digital health and social care information systems, and to ensure it anchors the development of large language models and AI-enabled tools in health, rehabilitation, and social care. Not required.