Corentin J Gosling 1 2 3, Laure Boisseleau 4, Marco Solmi 5 6 7 8 9, Micheal Sandbank 10, Lucie Jurek 5 11, Mikail Nourredine 5 12, Gabriella Porcu 4, Elisa Murgia 4, Joaquim Radua 13 14, Paolo Fusar-Poli 14 15 16 17, Klara Kovarski 18 19, Serge Caparos 20 21, Ariane Cartigny 22 23, Samuele Cortese 5 24 25 26, Richard Delorme 22 27 28
Affiliations Expand
PMID: 40877399
DOI: 10.1038/s41562-025-02256-9
Abstract
The use of complementary, alternative and integrative medicine (CAIM) is highly prevalent among autistic individuals, with up to 90% reporting having used CAIM at least once in their lifetime. However, the evidence base for the effects of CAIM for autism remains uncertain. Here, to fill this gap, we conducted an umbrella review of meta-analyses exploring the effects of CAIM in autism across the lifespan and developed a web platform to disseminate the generated results. Five databases were searched (up to 31 December 2023) for systematic reviews with meta-analyses exploring the effects of CAIM in autism. Independent pairs of investigators identified eligible papers and extracted relevant data. Included meta-analyses were reestimated using a consistent statistical approach, and their methodological quality was assessed with AMSTAR-2. The certainty of evidence generated by each meta-analysis was appraised using an algorithmic version of the GRADE framework. This process led to the identification of 53 meta-analytic reports, enabling us to conduct 248 meta-analyses exploring the effects of 19 CAIMs in autism. We found no high-quality evidence to support the efficacy of any CAIM for core or associated symptoms of autism. Although several CAIMs showed promising results, they were supported by very low-quality evidence. The safety of CAIMs has rarely been evaluated, making it a crucial area for future research. To support evidence-based consideration of CAIM interventions for autism, we developed an interactive platform that facilitates access to and interpretation of the present results ( https://ebiact-database.com ).
