Increased rates of chronic physical health conditions across all organ systems in autistic adolescents and adults.

The sample was a convenience survey, not a random population sample, and was biased toward white participants, participants assigned female at birth, UK residents, university-educated participants, overweight or obese participants, and participants who did not drink or smoke. Autism diagnosis and health conditions were self-reported, and some participants may have endorsed diagnoses without formal assessment. Only people with a professional autism diagnosis were included, so autistic people awaiting diagnosis or self-diagnosing were not represented. The survey was available only in English and required completion of full family and medical histories, so people without internet access, those unable to complete a long survey, and those with intellectual disability that limited participation were likely missed. The survey asked about conditions ever had across the whole lifespan, not when they occurred or whether they are current, and it excluded psychiatric conditions, infections and traumatic injuries. Cardiovascular, respiratory, diabetes and cancer results were not analysed in these regression models because they were reported elsewhere. Network analysis is preliminary, uses a model still in development, compares samples of different sizes, and groups conditions by organ system, which may misclassify conditions that affect multiple systems. Only 13% of autistic participants and 21% of non-autistic participants reported no physical health conditions, which may reflect sampling bias. Lifestyle variables were measured broadly, so smoking and alcohol patterns may not be fully captured, and social determinants of health were not measured in detail.

Funding was provided by the Autism Centre of Excellence at Cambridge, Rosetrees Trust, Cambridgeshire and Peterborough NHS Foundation Trust, Queen Anne's Gate Foundation, Corbin Charitable Trust, Wellcome Trust, Templeton World Charity Foundation, NIHR Cambridge Biomedical Research Centre, NIHR Applied Research Collaboration East of England, Innovative Medicines Initiative 2 Joint Undertaking and SFARI. The supplied text does not report author conflicts of interest.

Do not read the higher reported rates as proof that autism causes physical illness or that every autistic client has these conditions. The study is a cross-sectional self-report survey with sampling biases, so it cannot show cause and may not represent all autistic people.