Using Data From the SC Autism and Developmental Disabilities Monitoring Project to Assess Perinatal and Neonatal Risk Factors for Autism

Background:  Data from the Centers for Disease Control and Prevention indicates that approximately 1 in 110 8-year-old children are affected with an Autism Spectrum Disorder (ASD).  Different studies have found a variety of perinatal and neonatal factors as risk factors for ASD, but the data is inconsistent and often lacks a comparison population. 

Objectives:  The purpose of this study was to link data from the South Carolina Autism and Developmental Disabilities Monitoring Project (SC ADDM) with the Perinatal Information System (PINS) at the Medical University of South Carolina (MUSC)  to allow exploration of perinatal and neonatal factors for a group of children with ASD and a control group without ASD. 

Methods:  Information was collected on 8-year old children (born in 1992, 1994, 1996, 1998, and 2000) residing in a 23-county area of South Carolina between 2000 and 2008 as part of an ongoing, population-based multiple source surveillance of Autism Spectrum Disorders in South Carolina.  At MUSC the Division of Pediatric Epidemiology oversees an integrated patient information system referred to as the Perinatal Information System (PINS).  The data contained in this system encompass the events and clinical experiences of maternal and child outcomes from the onset of pregnancy until discharge from developmental pediatrics care and include almost 4,000 different perinatal and neonatal variables. This data has been collected and maintained since 1992.  Data from the two databases will be linked.  For children with an ASD that are linked to the PINS database, an additional 3 controls from PINS will be picked and matched on age, gender, and race.  Differences in perinatal and neonatal risk factors will be compared between the groups and possible associations with ASD risk will be assessed.

Results:  To date, only 1 of the 5 study years has been linked.  In 2000, there were 2,463 births documented in the PINS dataset.  In 2008, there were 264 children with a confirmed surveillance case of ASD.  Approximately 12% of the children with ASD (n=31) from SC ADDM were located in the PINS database.  The group was 58% black, 42% white, 90% male and 10% female.  There was no difference in insurance status (private, Medicaid, self-pay) between the two groups.

Conclusions:  The initial findings indicate the feasibility of performing this data linkage.  Based on the current knowledge of prevalence, we would expect approximately 1% of the 2,463 births to be confirmed with ASD.  In fact, the 31 children represent 1.26% of the cohort with ASD.   Further there was a larger than expected percentage of African Americans found in this population.  The outcome allows   further exploration of this group that, to date, has little information available.  It is planned to complete the linkage of the remaining 4 study years (includes 609 children with ASD from SC ADDM) with PINS and expect approximately 73 more matches, to total 104 children linked between the 2 databases.  Variables to be explored include birth weight, past pregnancy outcome, parity, prenatal information, maternal pregnancy complications, maternal therapy, neonatal diagnosis, neonatal medications, and neonatal procedures.