Sex Differences in Parent Report of Adaptive Behavior of Children at Risk for Autism Based on the Modified Checklist for Autism in Toddlers (M-CHAT)

Background:   The Positive Predictive Value (PPV) of the Modified Checklist for Autism in Toddlers (M-CHAT), a parent report early screening tool, is higher for males than for females (Ludwig et al., IMFAR 2011), and parents endorse different items on the M-CHAT based on sex (Ludwig et al., 2012). However, sex differences have not emerged on direct measures of ASD symptoms and cognitive skills in children who screen positive on this screening tool (Ludwig et al., IMFAR 2013).

Objectives:  The current study examined sex differences in parent report of adaptive behavior in toddlers who received an ASD evaluation based on M-CHAT screen positive status in order to explore how parent report may contribute to the differential PPV of the M-CHAT in boys and girls. 

Methods:   The sample included 250 males and 106 females (mean age=25.3 months, SD=4.55) who were evaluated based on screen positive (at risk) status on the M-CHAT, a parent questionnaire administered at pediatric well-visits. Evaluations included autism diagnostic measures (ADOS, ADI-R, CARS), cognitive testing (Mullen Scales of Early Learning; MSEL), as well as parent report of history and adaptive behavior (Vineland Adaptive Behavior Scales; VABS). One hundred seventy-six toddlers were diagnosed with ASD at the time of evaluation (139 males, 37 females) and 180 toddlers were not (111 males, 69 females); 71% of toddlers not diagnosed with ASD were diagnosed with another developmental delay. 

Results:   Four ANOVAs were conducted with sex and ASD status (ASD vs. non-ASD) as fixed factors and the VABS domain scores as dependent factors. There was no interaction or main effect of sex for the Daily Living Skills or Socialization domain scores; however toddlers diagnosed with ASD demonstrated weaker skills in these areas (p’s<.05). Although children diagnosed with ASD also demonstrated weaker parent reported Communication and Motor Skills than toddlers without ASD, there was a significant interaction between sex and ASD status such that girls diagnosed with ASD were rated lower on these domains than boys diagnosed with ASD (p’s<.05).

Conclusions: This pattern of findings suggests that there are differences in parent report of adaptive skills in boys and girls considered at risk for ASD based on the M-CHAT. Generally, toddlers with ASD demonstrated weaker parent-reported adaptive behavior than toddlers without ASD, but girls with ASD were rated lower on Communication and Motor Skills than boys with ASD. These findings are in contrast to recent findings that demonstrated no sex differences on direct measures of language and communication (MSEL) in the same sample (Ludwig et al., IMFAR 2013). These findings may suggest that parents of girls with ASD may under-report their child’s everyday adaptive skills, or that our current direct observation measures may not capture the full range of behavioral sex differences in children with ASD. In addition, these data also suggest that parent perception of child behavior may differ based on sex, which could influence the way in which parents fill out the M-CHAT.