International Meeting for Autism Research: Endorsement of Social Items on the M-CHAT Does Not Predict the Presence of Parental Concern about Toddler's Social Functioning

Endorsement of Social Items on the M-CHAT Does Not Predict the Presence of Parental Concern about Toddler's Social Functioning

Friday, May 13, 2011
Elizabeth Ballroom E-F and Lirenta Foyer Level 2 (Manchester Grand Hyatt)
3:00 PM
M. G. Arroyo1, D. L. Robins1 and D. A. Fein2, (1)Georgia State University, Atlanta, GA, (2)University of Connecticut, Storrs, CT

Studies show that prior to diagnosis, many parents of children with an autism spectrum disorder (ASD) report speech delay as a primary concern about their child’s development. Social concerns are reported infrequently and, if reported, are often secondary to language concerns. The current study investigates whether the number of social deficits endorsed on the Modified Checklist for Autism in Toddlers (M-CHAT; Robins et al., 1999) predicts parental endorsement of social concern.


It is predicted that parents who endorse a higher number of social deficits on the M-CHAT will be more likely to report social impairment on free response questions regarding concerns about their child’s development prior to a diagnostic evaluation.


Children who screened positive on the M-CHAT and received a diagnostic evaluation were recruited from a larger study (n=174). Of these children, 118 were diagnosed with an ASD (97 males; mean age at evaluation=26.48 months, SD=4.56, mean M-CHAT score=10.03, SD=4.01) and 56 children were diagnosed with another developmental delay, no diagnosis, or typically developing (44 males; mean age at evaluation=25.62 months, SD=4.21, mean M-CHAT score=7.70, SD=3.70). Age at evaluation was not significantly different between groups, t(172)=-1.177, p=.241, but M-CHAT score was significantly greater for the ASD group than the nonASD group, t(172)=-3.67, p<.001.

Prior to the diagnostic evaluation, all caregivers filled out a history questionnaire. Three open-ended items evaluated parent concern; parent responses were coded for the presence or absence of a social concern, such as “doesn’t look at me” or “ignores other children” (inter-rater reliability: Kappa=.82, p<.05).

Participants all received a score (maximum of 11) based on the number of M-CHAT social items endorsed (M-CHAT-Social); social items were determined by group consensus if they mapped onto DSM-IV social symptoms.


Of the parents indicating concern for their child’s development (n=139), 49 parents endorsed social delays. ROC analysis did not indicate an optimum cutoff point for M-CHAT-Social (AUC=.558), indicating that M-CHAT-Social score is not predictive of social concerns. Excluding parents who did not indicate any concerns, ROC analysis still failed to indicate an optimum cutoff point for M-CHAT-Social (AUC=.612). In addition, report of social deficits was not more likely for children diagnosed with ASD, χ2 (1, n=139)=1.189, p=.275, although overall M-CHAT-Social was higher for children with ASD, t(172)=-3.98, p<.001. Discriminant function analysis indicated that social concerns were more likely to be indicated if the parent endorsed the items “does your child respond to his/her name when you call” and “does your child take an interest in other children.”


A higher M-CHAT-Social score did not predict parent report of social impairment, indicating that these deficits may not be as salient to parents as language delay. Parents should be educated about social milestones for children so that when social deficits are present, parents bring them to the attention of healthcare providers, which would aid early detection. Future research could investigate the impact of various family factors such as birth order, parent age, and education on parent report of social concerns.

| More