Accuracy of Parental Report As Compared to Standardized Assessment: A Comparison of the Ages and Stages to the Mullen and ADOS

Background: The need for early detection of developmental delays in young children is critical. Early intervention is key for children with ASD where it has been shown to alter developmental trajectories and improve lifetime outcomes (Orinstein et al., 2014; Zwaigenbaum et al., 2015).  There is a push for effective, parental self-report screening measures to help identify developmental delays and ASD. One of the more commonly used measures is the Ages and Stages Questionnaire (ASQ).  However, previous research has revealed mixed findings regarding the ASQ as a developmental screener with some studies showing poor agreement with the Bayley Developmental Scales (Veldhuizen et al., 2014) and others, especially across cultures, finding the ASQ to be a strong predictor of developmental delays (Juneja et al., 2011).  Additionally, there is a need to evaluate the use of this measure for children with ASD. 

Objectives: The goal of this study is to assess the relationship between the ASQ and standardized assessments for children with ASD.

Methods: 51 children (47 males) who received an initial ASD diagnosis following evaluations conducted as part of a research study. All children were between the ages of 16-43 months (mean = 29.71) and referred based on parent concerns and/or recommendation from pediatrician or early interventionist. Parents completed the ASQ and children were evaluated with the Mullen Scales of Early Learning and the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2).

Results: ASQ scales were significantly correlated with T-scores from the Mullen, with significant associations between the ASQ problem solving scale and Mullen visual reception, fine motor, expressive, and receptive language domains  (r’s=.53, .49, .46, and .64 respectively, all p’s<.01). The ASQ fine motor scale was associated with Mullen visual reception, fine motor, and expressive language domains  (r’s=.29, .28, and .35 respectively, all p’s<.05). The ASQ communication scale was associated with Mullen visual reception, fine motor, receptive, and expressive language domains (r=.54, .32, .68 and.70 respectively, all p’s<.05).  ASQ social-emotional scale was associated with Mullen visual reception, fine motor, receptive and expressive language domains (r=.42, .33, .36, .45 respectively, all p’s<.01). The ASQ communication scale was negatively correlated with the ADOS social affect score (r=-0.39, p<.05) and ADOS total score (r=-.83, p<.05).  

Conclusions: Findings reveal that the ASQ is a good predictor of delays in development for children with ASD. Results provide support for the use of parent-report measures in identifying children requiring referrals for evaluations and parent perceptions regarding their children’s skill level in the specific domains of nonverbal problem solving, fine motor, and communication skills. The social-emotional scale was also associated with Mullen performance; however, this scale did not predict specific autism symptomatology on the ADOS. Instead, the communication scale of the ASQ is indicative of social and communication ASD red flags. Practitioners are encouraged to pay close attention to the communication scale when considering whether a referral for an ASD specific evaluation is necessary. Overall, the ASQ appears useful to identify developmental delays but an autism specific screener is also indicated given the more limited associations with the ADOS.