Increased Risks of Congenital, Neurologic, and Endocrine Disorders Associated with Autism in Preschool Children: Cognitive Ability Differences CHUAN-YU CHEN,PHD, KUANG-HUNG CHEN, MSC,CHIEH-YU LIU,PHD, SU-LING HUANG, MSC, AND KEH-MING LIN, MD, MPH Objective To investigate the increased risk of congenital, neurologic, and endocrine disorders in autistic preschool children and to probe possible cognitive impairment-associated variation in such risks. Study design Using a population-based longitudinal study, a total of 3440 autistic children born in 1997-1999 and 33 391 age- and residential urbanicity–matched control subjects were identified from the National Health Insurance Research Database in Taiwan. Conditional logistic analyses were performed to estimate the strength of association stratified by the presence of cognitive impairment. Results Autistic children were found to have greatly elevated risks of congenital anomalies (eg, tuberous sclerosis: adjusted odds ratio [aOR]  3461) and neurologic disorders (eg, epilepsy: aOR  513) compared with their matched nonautistic peers. The increased risk of medical diseases for mentally retarded autism were approximately 1.6 to 9 times greater than those for isolated autism. Conclusions The observed cognitive impairment-related variation in the increased risk of congenital, neurological, and endocrine disorders with autism may provide some clinical and etiologic implications that warrant investigation in the future. (J Pediatr 2009;154:345-50) A utism is a diagnostic category comprising diverse clinical groups with manifested impairments in social interaction, deficits in verbal or nonverbal communication, and repetitive patterns of behaviors or interests. Most symptoms of autism emerge before 3 years of age, and the severe constellation of behavioral disturbances, along with other coexisting problems (eg, epilepsy), may place a lifelong burden on the affected individuals and their families. 1 Cumulative evidence suggests that the risk factors of autistic disorders may be heterogeneous across subgroups such as clinical presentation, 2-4 in which cognitive im- pairment is one of the clinical conditions commonly used to aid in the redefinition of phenotype and to facilitate understanding of the pathology of autistic disorders. 5,6 Meanwhile, clinical observation regarding higher prevalence of congenital, neuro- logic, and endocrine abnormalities among autistic cases has led to further investigation into possible roles of neurologic and autoimmune systems in the cause and pathology of autism. 7-13 Nonetheless, in few epidemiologic studies that have examined such associa- tions, the results have often been inconsistent, in part, because of clinical heterogeneity in autism and characteristics of certain medical problems (eg, rare incidence). 14,15 To bring clarity of this issue, the aims of this population-based study are to investigate the risk of congenital, neurologic, and endocrine disorders among autistic children, and to assess possible clinical heterogeneity, defined by the presence of cognitive impairment, in such risks in preschoolers. This evidence is important to generate insights into causal and pathologic processes, and even more important to offer potential therapeutic implications to symptom management. 16 METHODS This study was a population-based, matched case-control study of preschool chil- dren with the diagnosis of autism in Taiwan. The data were obtained from the 1997-2004 aOR Adjusted odds ratio ICD International Classification of Diseases NHIP National Health Insurance Program From the Division of Mental Health and Sub- stance Abuse Research, National Health Re- search Institutes (C.C., K.C., S.H., K.L.), Zhu- nan, the National Yang-Ming University, Institute of Public Health and Department of Social Medicine (C.C.), and the Depart- ment of Nursing, National Taipei College of Nursing (C.L.), Taipei, Taiwan. This work is completely supported by the National Health Research Institutes, which had no involvement in the study design, data collection, analysis, and interpretation, report writing, or the decision to submit the paper for publication. The authors de- clare no conflicts of interest. Submitted for publication Feb 19, 2008; last revision received Jun 4, 2008; accepted Sep 24, 2008. Reprint requests: Chuan-Yu Chen, PhD, National Health Research Institutes, Divi- sion of Mental Health and Substance Abuse Research, 35 Keyan Road, Zhunan, Miaoli County 350, Taiwan. 0022-3476/$ - see front matter Copyright © 2009 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.09.043 345