Note: Most Internet Explorer 8 users encounter issues playing the presentation videos. Please update your browser or use a different one if available.

MCHAT-R Screener Predictive Validity in a High-Risk Infant Sibling Population

Thursday, 2 May 2013: 09:00-13:00
Banquet Hall (Kursaal Centre)
H. Noble1, A. S. Weitlauf2, W. A. Loring3, A. Nicholson3, J. A. Crittendon2, C. R. Newsom4 and Z. Warren2, (1)TRIAD, Vanderbilt Kennedy Center, Nashville, TN, (2)Vanderbilt Kennedy Center, Nashville, TN, (3)Vanderbilt University, Nashville, TN, (4)Pediatrics, Psychiatry, & Psychology, Vanderbilt University, Nashville, TN
MCHAT-R Screener Predictive Validity in a High-Risk Infant Sibling Population

Background: Recent work suggests that the risk of recurrence of ASD in younger siblings of diagnosed children may be as high as 18.7% (Ozonoff et al., 2011).  Given the elevated prevalence rates in siblings and potential genetic contributions to ASD symptomatology (Szatmari et al., 2007), effectively screening infant siblings is of high importance to clinicians and families. However, the neurodevelopmental complexity of this population (e.g., elevated rates of non-ASD concerns, broader phenotype concerns, patterns of resiliency despite early delays)  effectively screening siblings regarding ASD specific risk is a challenging process

Objectives: The purpose of this study was to determine if a screening tool alone is sufficient to detect an “at risk” diagnosis for ASDs in Sibs-ASD. We did this by evaluating children who passed the MCHAT-R within the context of a comprehensive psychological evaluation.

Methods: Participants included 37 infant siblings of children diagnosed with an autism spectrum disorder (Sibs-ASD). Potential participants in a larger study who passed a screening with the M-CHAT-R between 16 and 30 months of age were invited to participate in a cost-free developmental evaluation between ages 16-42 months. This evaluation included a thorough developmental history, a DSM-IV structured clinical interview, and cognitive (Mullen Scales of Early Learning), social-communication (Autism Diagnostic Observation Schedule), and adaptive behavior (Vineland Adaptive Behavior Scales – II, Communication and Social domains) testing.

Results: Approximately 38% of the sample (14/38) received a neurdodevelopmental diagnosis at follow-up. Specifically, 19% (n=7) of Sibs-ASD who passed the MCHAT-R received an ASD diagnosis. An additional 19% (n=7) received other diagnoses, including developmental or language delay (n=5) and “at risk” for ASD (n=2). Post-hoc Tukey tests of one way ANOVAs revealed that participants who received an ASD diagnosis had significantly higher Calibrated Severity Scores (Gotham et al., 2010) than participants with no diagnosis (p < .05). No Diagnosis participants had significantly higher scores on Mullen Receptive and Expressive language than other participant groups (p < .05). No Diagnosis participants scored higher (p < .05) on measures of nonverbal problem solving than Other Diagnosis but not ASD participants. No differences emerged between groups on Vineland Communication or Social domains.  

Conclusions: Siblings of children with ASD are a complex population and therefore may require a more stringent diagnostic process for determining if they are actually “at risk” for an autism spectrum disorder. While some siblings with ASD will likely be captured a relatively small, but clinically meaningful percentage will not be detected via use of this instrument alone.  Given the high recurrence rate of ASD and other developmental concerns, clinicians and practice organizations my need to advocate for more in depth evaluation of siblings of children with ASD as standard practice rather than a follow-up to simple screening.

| More