The quantitative checklist for autism in todders

Conceptualisation of Autism Spectrum Conditions (ASC) are moving away from a categorical to a dimensional and quantitative approach. This thesis reports on preliminary validation data of a quantitative screening measure, the Quantitative Checklist for Autism in Toddlers (Q-CHAT). Chapter 1 reviews issues surrounding diagnosis and screening for ASC. Chapter 2 reports a systematic review of currently available screening instruments for ASC in children less than four years of age. No single screening instrument was found to have acceptable levels of sensitivity, specificity and positive predictive value in either a population or a clinic screening context. Chapter 3 reviews the first attempt in the UK at population screening for ASC (using the Checklist for Autism in Toddlers (CHAT)), and retrospectively applies an alternative scoring algorithm. Sensitivity improved at a small cost to specificity. Chapter 4 describes the development of the Q-CHAT and reports on a pilot study. Children with ASC scored higher on the Q-CHAT than unselected 18-24 month old toddlers. Chapter 5 describe the methods for the main validation data collection phase using the Q-CHAT. Chapter 6 reports the results from the screening and diagnostic phases of the study. Q-CHAT score approximated a normal distribution. Boys scored higher than girls. Six out of eight children who showed impairments consistent with ASC scored above the preliminary cut-point. Concurrent validity of the Q-CHAT was acceptable. Chapter 7 revealed adequate test retest reliability of the Q-CHAT. Chapter 8 investigated the dimensionality of the Q-CHAT using exploratory and confirmatory factor analysis techniques. There was no evidence that autistic traits in toddlers lie on a single dimension. Overall, results demonstrate that autistic traits in toddlers can be measured quantitatively. The Q-CHAT should be further investigated and refined as a screening instrument for ASC.EThOS - Electronic Theses Online ServiceGBUnited Kingdo