Immune Comorbidities in Children with Autism Spectrum Disorders

Lisa A. Croen, PhD1, Lilian L. Tee, MPH1, Bruce Fireman, M.A.1, Albin Leong, MD2, Lisa F. Barcellos, PhD3, and Pilar Bernal, MD4. (1) Division of Research, Kaiser Permanente, 2000 Broadway, Oakland, CA 94612, (2) Pediatrics, Kaiser Permanente, 2025 Morse Avenue, Sacramento, CA, (3) Division of Epidemiology, UC Berkeley School of Public Health, B84 Hildebrand Hall MC#7356, Berkeley, CA 94720-7356, (4) Child Psychiatry, Kaiser Permanente, 175 Bernal Road, San Jose, CA

Background:  Immune system disturbances have been reported in several clinical studies of autism, and the presence of immune-related illnesses are frequently described by parents and providers.  However, population-based data on the prevalence of immune comorbidities among children with autism are lacking.
Objectives: To determine the prevalence of immune comorbidities in individuals with autism, investigate whether these conditions occur more often than expected, and explore the timing of onset in relation to the autism diagnosis.
Methods:  We conducted an observational study among the membership of Kaiser Permanente (KP) in Northern California.  Individuals born from 1980 to 2003 with a minimum of one year KP membership were eligible for inclusion.  All children with at least two autism diagnoses recorded in outpatient records between the ages of 3-18 were identified (n=5,565).  A comparison group of children without autism was randomly sampled at a ratio of 5 to 1, matched to case children on year of birth, sex, and length of KP membership (n=27,825).  The main outcomes - asthma, allergies, and autoimmune diseases - were identified from KP inpatient and outpatient databases.  Chi-square tests were used to evaluate case-control differences.

Results:  Overall, immune-related conditions were diagnosed with equal frequency among children with autism and controls (28.6% vs. 27.6%, P=0.11).  However, allergies and autoimmune diseases were diagnosed significantly more often among children with autism than controls (allergy: 20.6% vs. 17.7%, P<0.0001; autoimmune disease: 1% vs. 0.76%, P=0.04), and asthma was diagnosed significantly less often (13.7% vs. 15.9%, P<0.0001).  Psoriasis was the most frequently diagnosed autoimmune condition among children with autism, and it occurred over twice as often in cases as controls (0.34% vs. 0.15%, P<0.0018). 

Conclusions:  These results support previous observations that children with autism have somewhat elevated rates of specific immune-related comorbidities, and improve our understanding of the biomedical correlates of the autism phenotype.