Artificial intelligence adoption by allied health professionals: a scoping review

Objective Although artificial intelligence (AI) can optimise care delivery, a gap exists in allied health workforce readiness to adopt AI in clinical practice. This review aims to identify how AI is currently utilised in clinical practice within physiotherapy, occupational therapy, speech pathology, podiatry, and dietetics. We also evaluated the benefits of and barriers to AI adoption.Data sources Medline, Embase, PsycINFO, and CINAHL databases were searched from 2014, followed by manual reference list screening of included articles.Review methods Scoping review of literature using the Arksey and O’Malley framework. Data from the selected professions were also analysed using large language models. ChatGPT Data Analyst GPT ® (Version GPT-5, OpenAI) helped categorisation of data.Results A rapid increase in allied health AI studies since 2022 showed that current use by allied health disciplines is predominantly in hospitals. Types of AI implemented included machine and deep learning, natural language processing, generative AI, expert systems, case-based reasoning and speech recognition and analysis. AI was used for prediction and prognostics, diagnosis and classification, therapy, real-time assessment, decision support and communication. Risks to AI implementation related to insufficient workforce education; inconsistent use in electronic medical records; concerns re patient privacy, and poor access to AI tools.Conclusion AI adoption is limited and variable across allied health disciplines. It is mainly used in diagnostics and reporting, and to a lesser extent in treatment. An urgent need exists to educate and train the allied health workforce to ensure competency in its use across the care pathway.