Diagnostic Stability In Toddlers Diagnosed with Developmental Delay without Autism

Background:   While there has been increasing research and clinical interest in of the stability of diagnosis in young children with autism, less is known about the stability of diagnosis in children with developmental delay (without autism). Development of interventions for increasingly young children (e.g., Dawson et al., 2010; Green, Brennan, & Fein, 2002; McGee, Morrier, & Daly, 1999) and increasing public awareness of early risk for autism have led to greater numbers of referrals of very young children for differential diagnosis of autism versus developmental delay during toddlerhood. Further, recent work lends strong support for the idea that autism is detectable between 18-30 months in some children (Lord, 1995; Lord et al., 2006; Stone et al., 1999), but there is evidence that there may also be a subgroup of children whose core symptoms manifest closer to the third birthday. Indeed, Chawarska, Klin, Paul & Volkmar (2007) found that 1 of 4 children with developmental delay that they followed-up received a PDD diagnosis one year later.

Objectives:    The goal of the study was to examine the stability of diagnosis in young children with developmental delay.

Methods:    Twenty toddlers (4 female) referred for concerns regarding delays in development between 18-30 months received a diagnostic evaluation and then were followed-up as preschoolers. As toddlers, they displayed delays in at least one sub-domain of the Vineland and the Mullen, and had standard scores below 85 on the Mullen. Importantly, these children also were administered the Autism Diagnostic Observation Schedule (ADOS) and the toddler version of the Autism Diagnostic Interview-Revised (ADI-R), and they did not meet criteria for an autism spectrum disorder based on expert clinical judgment and combined scores from these measures.

Results:    At follow-up (mean age = 41.35 months, range = 30-59), three of the twenty children (15%) previously diagnosed with development delay received a diagnosis of an autism spectrum disorder (ASD) based on administration of the ADOS, ADI-R, and clinical judgment.  In terms of cognitive ability, at follow-up four children with initial diagnosed with developmental delay achieved Mullen composite scores in the average range. Baseline composite scores in the sample were significantly related to scores at follow-up, r(20)= .51, p = .02. 

Conclusions:   The current study suggests that some children initially diagnosed with developmental delay without autism will develop ASD symptoms during the preschool period and later qualify for a diagnosis of ASD.  Although cognitive ability as measured during toddlerhood was predictive of later cognitive ability during preschool, 20% of toddlers initially included in a group with developmental delays performed within the average range when retested as preschoolers.