People with autism spectrum disorder (ASD) had a higher risk of developing Parkinson’s disease early in life, a nationwide population-based study in Sweden has suggested.
After controlling for birth year, age and sex, the risk of Parkinson’s was four times higher in people with versus without autism (relative risk [RR] 4.43, 95% CI 2.92-6.72), reported Weiyao Yin, MD, PhD, of the Karolinska Institute in Stockholm, and co-authors, in JAMA Neurology.
The researchers said risk estimates were similar after adjusting for socioeconomic status, family history of mental illness or Parkinson’s, and age at autism diagnosis. Preterm or early-term birth, an autism risk factor, was not associated with the disease.
Depression correlated with Parkinson’s risk, independent of autism. Antipsychotic exposure reduced the link between autism and Parkinson’s (RR 2.00, 95% CI 1.27-3.14), but showed no interaction with autism on Parkinson’s risk.
The study was based on all people born in Sweden from 1974 to 1999. Parkinson’s disease in this cohort was very rare: 438 of 2.2m people without autism (0.02%) had a Parkinson’s diagnosis, compared with 24 of 51 954 people with autism (0.05%).
The findings add to research that suggests a potential biological link between autism and parkinsonism or Parkinson’s disease, including evidence of PARK2 copy number variation in people with autism.
It’s not clear what contributes to the relationship between autism and Parkinson’s, Yin noted.
“One hypothesis is that the brain’s dopamine system is affected in both conditions because dopamine plays a crucial role in social behaviour and motor control. Genetic correlation between the two conditions could also contribute to the association,” she told MedPage Today.
“Additionally, antipsychotic exposure partially explained the observed association between ASD and Parkinson’s disease in our study. However, even after adjusting for antipsychotic exposure, ASD was still independently associated with a twofold increased risk of Parkinson’s.
“Nevertheless, the possibility of residual confounding by other medications and environmental and behavioural factors associated with Parkinson’s disease remains.”
Yin and colleagues studied all individuals born in Sweden between 1974 and 1999, following them from age 20 until December 2022. Less than half (48.6%) of the cohort was female and the median age at study exit was 34.
Diagnoses were obtained from ICD codes in National Patient Register (NPR) data. Parkinson’s disease was defined as the first-ever diagnosis of Parkinson’s and other idiopathic or degenerative parkinsonian disorders.
“The quality of the NPR has been verified and validated (96%) for ASD,” the researchers wrote. The NPR’s accuracy in detecting parkinsonian disorders is 88%.
Restricting the analysis to idiopathic Parkinson’s disease showed a comparably strong association with autism (RR 3.53, 95% CI 1.86-6.70), Yin and co-authors said. The relationship between autism and Parkinson’s remained after adjusting for neuroleptics.
An important limitation of this analysis was that the maximum age at follow-up was under 50, the researchers acknowledged. “The limited follow-up period reduces the ability to observe the progression of Parkinson’s, an age-dependent degenerative disease, in individuals with ASD,” they wrote.
Future studies with longer follow-up may identify whether autism and late-onset Parkinson’s are connected, they added.
“Autism spectrum disorder is a lifelong condition, and more individuals with ASD now progress into middle and older adulthood,” Yin noted.
“This is the first large, longitudinal, population-based study of the future risk of Parkinson’s disease in people with ASD. We believe it could stimulate research that sheds light on the unclear causes of the disease and ASD.”
Study details
Risk of Parkinson disease in individuals with autism spectrum disorder
Weiyao Yin, Abraham Reichenberg, Michal Schnaider Beeri, et al.
Published in JAMA Neurology on 27 May 2025
Abstract
Importance
Recent research suggests a plausible biological link between autism spectrum disorder (ASD) and Parkinson disease (PD). Nonetheless, large longitudinal studies examining the risk of PD following ASD are lacking.
Objective
To examine the association between ASD and future PD risk.
Design, Setting, and Participants
A nationwide population-based prospective cohort study was performed using data from Swedish national registers. All individuals born in Sweden from 1974 to 1999 with follow-up from age 20 until 31 December 2022, and with complete covariate data were included. The analysis was completed in August 2024.
Exposures
Diagnoses of ASD as a time-varying exposure obtained from the National Patient Register.
Main Outcomes and Measures
Diagnoses of PD were obtained from the National Patient Register to 2022. The relative risk (RR) of PD was quantified using incidence rate ratios with 95% CIs from Poisson regression. Preterm birth, depression, antidepressant use, and antipsychotic exposure over time were potentially modifying life events.
Results
The study included 2 278 565 individuals (median [IQR] age at exit, 34 [29-42] years; 1 106 772 female [48.6%]), contributing 33 858 476 person-years. PD occurred in 438 of 2 226 611 individuals without ASD (0.02%; 1.3 cases/100 000 person-years) and 24 of 51 954 individuals with ASD (0.05%; 3.9 cases/100 000 person-years) (RR, 4.43 [95% CI, 2.92-6.72]). The risk estimates were similar after adjusting for sex, socioeconomic status, family history of mental illness, family history of PD, and age at ASD diagnosis. Preterm or early-term birth was not associated with and did not modify the PD risk. Depression and antidepressant use (present in 24 257 individuals with ASD [46.7%]) were associated with increased risk of PD (RR, 2.01 [95% CI, 1.40-2.88]), independent of ASD. Antipsychotic exposure (present in 16 387 individuals with ASD [31.5%]) reduced but did not fully attenuate the association (RR, 2.00 [95% CI, 1.27-3.14]) and showed no interaction with ASD on PD risk.
Conclusions and Relevance
ASD was associated with increased risk of PD, even after adjusting for depression or antidepressant use and antipsychotic exposure. These findings suggest a potential shared aetiology between neurodevelopmental disorders and PD, and a heightened awareness of long-term neurological conditions in individuals with ASD may be warranted.
Medpage Today article – Autism Tied to Early-Onset Parkinson’s Disease (Open access)
See more from MedicalBrief archives:
Autism diagnosis rates triple in past decade – US cross-sectional study
Brain changes in autism more sweeping than thought – UCLA study
Thousands of Parkinson’s patients initially misdiagnosed