International Meeting for Autism Research: Regression and Autism: The Answer Depends on the Question

Regression and Autism: The Answer Depends on the Question

Friday, May 13, 2011
Elizabeth Ballroom E-F and Lirenta Foyer Level 2 (Manchester Grand Hyatt)
2:00 PM
J. Foley1, A. Diehl1, T. Smith2, S. L. Hyman3 and T. Musa1, (1)Pediatrics, University of Rochester, Rochester, NY, (2)University of Rochester, Rochester, NY, United States, (3)University of Rochester School of Medicine, Rochester, NY, United States

Estimates of the rates of regression in autism spectrum disorder (ASD) are typically based on retrospective parent reports. However, the reliability of such reports is currently unknown and may be affected by factors such as the assessment method. For example, although questionnaires allow greater standardization, interviews are more flexible and give more opportunities to clarify parent responses. Because of these differences, the two assessment methods may yield differing results.


We compared parents’ retrospective report of regression on a self-administered questionnaire and a clinician-administered interview given during consecutive study visits.


A chart review was conducted looking at parent reports of regression in their children (ages 2-17.5 years, n=161) who participated in the Rochester site of the Autism Treatment Network (ATN). Responses to regression questions in the Autism Diagnostic Interview–Revised (ADI-R, questions # 11-28) and the loss of skill question on the ATN Developmental and Medical History form were compared using descriptive statistics. Associations between reports of regression and child variables (age, IQ) were also analyzed.


Overall, parents were significantly more likely to report regression on the ATN questionnaire (n = 65) than on the ADI-R (n = 39), z = 2.98, p < .01. Similar patterns were observed in individual developmental domains, with 60 parents reporting loss of language on the ATN questionnaire compared to 30 parents on the ADI-R, z = 3.55, p < 0.01, and 30 reporting loss of motor skills on the ATN questionnaire compared to 11 on the ADI-R, z = 3.01, p < 0.01. The rate of intellectual disability (IQ < 70 or untestable) among children whose parents reported regression on both the ATN questionnaire and ADI-R (16 of 36, 44%) was not significantly different from the rate of intellectual disability in the rest of the sample (46 of 125, 37%), z = 0.64, NS.


Parent reporting is the basis for much clinical research, including the search for biomarkers and other biologic investigations that may be related to regression. Parents in this study were significantly more likely to report regression on a parent questionnaire than a clinical interview. Reports of regression were not associated with whether or not children had an intellectual disability. Only prospective studies can determine the accuracy of questionnaires and interviews for validly identifying regression in children with ASD.

| More