Early Detection of ASD: A Meta-Analytic Study on the M-CHAT and Other Tools Used for Universal Toddler Screening

Background:

Great efforts are being focused on developing instruments for early detection. However, there is no agreement about the best instrument. In addition, some scientists and clinicians have questioned early universal screening, until an adequate instrument with good psychometric properties is available¹

The two versions of the Modified Checklist for Autism in Toddlers (M-CHAT and M-CHAT-R/F) are the most internationally used ASD screening instruments² and when including the follow-up interview procedure have a high positive predictive value ^3,4. However, no meta-analytic study has provided information on the M-CHAT and other tools.

Objectives:

This study aims to describe and synthesize the psychometric properties of the ASD screening tools, at an international level, applied in general population at early ages. The main objective is to analyze if there are statistically significant differences from the M-CHAT and M-CHAT-R/F and other screening tools.

Methods:

A systematic literature review of peer-reviewed publications was conducted to identify studies published from January 1992 through April 2015. CINHAL, ERIC, PsycINFO, PubMed and WOS databases were searched. Effect size will be calculated using the hierarchical summary ROC (HSROC) model.

A hierarchical Cluster Euclidean distance using HJ-Biplot scores-Linkage Ward has been calculated based on data sensitivity, effectiveness of each screening tools and several similarity coefficients which evaluate the concordance between the results of the screening tool (Jaccard, Sokal and Sneath, Ochaiai and Sokal, Michener).

Results:  

From the identified 1423 non-duplicate citations screened, 20 papers were selected for qualitative review (using Quadas-2). Fourteen studies were eligible for inclusion. The agreement between judges was adequate, kappa=0.643, p<.01.

As a preliminary finding we can differentiate two clusters; the first one shows a high concordance and is comprised of 11 studies. In this cluster are the majority of the studies done with the M-CHAT. In the second cluster we find studies with significant discrepancy between the results of the test and the reference standard.

Conclusions:  

Many researchers around the world are working to establish good ASD screening practices.. The results of this study will help us to have a better understanding about the effectiveness of the different procedures for the early detection.

References:

^1. Scottish Intercollegiate Guidelines Network. Assessment, diagnosis and clinical interventions for children and young people with autism spectrum disorders. 2007. Available from: http://www.sign.ac.uk/pdf/sign98.pdf.

^2. Canal-Bedia R, García-Primo P, Martín- Cilleros MV, et al. Modified Checklist for Autism in Toddlers: cross-cultural adaptation and validation in Spain. J Autism Dev Disord. 2011; 41(10):1342–1351

^3. Draft Evidence Review: Autism Spectrum Disorder in Young Children: Screening. U.S. Preventive Services Task Force. August 2015. http://www.uspreventiveservicestaskforce.org/Page/Document/draft-evidence-review106/autism-spectrum-disorder-in-young-children-screening

^4. Robins DL, Casagrande K, Barton M, Chen CM, Dumont-Mathieu T, Fein D. Validation of the modified checklist for Autism in toddlers, revised with follow-up (M-CHAT-R/F). Pediatrics.2014; 133(1):37-45

^5. Frutos E, Galindo M & Leiva V. An interactive bi-plot implementation in R for modeling genotype-by-environment interaction. Stochatic Environmental Research and Risk Assessment. 2014; 28, 1629-1641