Early Detection of Pervasive Developmental Disorders

University Of Connecticut Storrs, Storrs-Mansfield CT

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Autism and other pervasive developmental disorders (PDD) are severe disorders of development, often leading to life-long disability. Early detection and intervention can substantially improve prognosis; however, with available screening instruments early detection is difficult. Diagnosis and intervention often follow initial suspicion by as much as three years or more, and the nature of effective interventions is not well understood. The proposed study involves the validation of a parent checklist for 24-month-olds which is under current development. Pilot data on 1293 two year olds, 59 of whom were evaluated and 38 of whom were diagnosed as autistic or PDD, demonstrates excellent promise of effectiveness. Small grants from NAAR, the U.S. Department of Education, and the University of Connecticut Research Foundation, and an NIMH/NRSA individual predoctoral fellowship, have allowed the investigators to begin preparing for large-scale screening and to begin a follow-up study of the current participants. In the current application, 33,400 children will be screened, 30,000 from an unselected sample of pediatric visits, and 3,400 from high risk samples of siblings of children with autism and children referred for early intervention. A projected 140 with PDD (and approximately 35 false positives) will be evaluated at 24 months and again at 42 months. Evaluations will include psychiatric diagnoses and assessment of cognitive, language, and adaptive functioning; reliability of diagnoses (by ADOS and ADI-R) will be confirmed by a child psychiatrist. Analyses at both time points will use signal detection and discriminant function analyses to determine the optimal set of items to detect PDD, and will assess the sensitivity and specificity of the instrument. Data on interventions received during the 18 months between the initial and follow-up evaluations will also be collected. Using clinical status variables at 42 months as outcome measures, and controlling for child and family characteristics, intervention variables such as intensity, type, setting, and parent involvement will be studied for their relationship to outcome. The proposed study should greatly advance the development of an effective early screening tool for autism and related disorders. Although not a randomized trial of different interventions, the proposed research will also provide important information on the nature of effective interventions for children with autism and PDD. Planned future studies include replication in other states and other countries (e.g., Japan).