International Meeting for Autism Research: Longitudinal Change in Social Affect and Restricted and Repetitive Behavior Severity Using the ADOS

Longitudinal Change in Social Affect and Restricted and Repetitive Behavior Severity Using the ADOS

Friday, May 21, 2010: 4:15 PM
Grand Ballroom AB Level 5 (Philadelphia Marriott Downtown)
3:45 PM
V. Hus , University of Michigan Autism & Communication Disorders Center, Ann Arbor, MI
K. Gotham , University of Michigan Autism & Communication Disorders Center, Ann Arbor, MI
A. Pickles , Health Methodology Research Group, University of Manchester, Manchester, United Kingdom
C. Lord , University of Michigan, Ann Arbor, MI
Background: The development of a standardized ADOS severity metric (Gotham, Pickles, & Lord, 2009) provides an overall index to measure change in intervention research and will facilitate investigations of developmental trajectories of autism over time. Separate calibration of the Social Affect (SA) and Restricted and Repetitive Behavior (RRB) domains will further promote these efforts by allowing separate examination of these behaviors, which may have distinct trajectories. 

Objectives: To standardize ADOS SA and RRB domain scores, and then plot longitudinal trajectories of ASD severity separately for each standardized domain.

Methods: Standardized ADOS severity scores were calibrated separately for the SA and RRB domains using data from 1,807 cases. Mixed effects modeling was then used to identify separate domain severity trajectories for 345 individuals (total of 1,026 longitudinal cases).

Results: For some individuals, there is a discrepancy between SA and RRB severity scores, suggesting more severe symptoms in one domain compared to the other. Separate scores also allowed trajectories of severity to be analyzed for each domain. Several distinct patterns were identified. In general, SA and RRB severity show approximately consistent trajectories, with severity in both domains increasing or decreasing with time. However, in some cases, SA severity appears to decrease over time, whereas RRB severity remains relatively stable. In a few cases, SA and RRB trajectories are apparently opposite - i.e. as one increases, the other decreases.

Conclusions: If replicated, separately calibrated SA and RRB severity scores, as well as the identified trajectories of severity for each domain may be useful for stratifying samples in genetic and neurobiological studies. These scores may also inform clinical recommendations and intervention efforts. For individuals with uneven profiles (i.e. higher severity in one domain compared to the other), more targeted intervention programs may be identified. Using separate calibrated scores may also provide a more sensitive measure of intervention response, enabling change in one domain to be detected, even when behaviors in the second domain persist.