True versus false positives and negatives on the Modified Checklist for Autism in Toddlers

Abstract

Given the importance of early intervention services for children with autism spectrum disorders (ASD), early diagnosis of children is critical. At present, several ASD screeners exist for young children, with the Modified Checklist for Autism in Toddlers (M-CHAT) being one of the most widely researched. While the M-CHAT has good sensitivity and specificity, the presence of false positives and false negatives is inevitable. While false positives are not as disconcerting, since follow-up assessment will reveal the absence of ASD, false negatives can inadvertently delay much needed services to toddlers. The current study examined parental report for 552 toddlers who were referred for early intervention services due to being part of an at-risk population. The groups were divided into four groups based on their M-CHAT scores and actual diagnoses - a true positive group (n = 150), false positive group (n = 150), true negative group (n = 151), and false negative group (n = 101). All four groups were compared with respect to their total M-CHAT score and total critical items score on the M-CHAT. While the true positive and false positive groups significantly differed from the true negative and false negative groups on both of these dimensions, significant differences were also found between the true and false positive groups but not between the true and false negative groups. Furthermore, all 23 M-CHAT items were more commonly endorsed by individuals in the true positive group when compared to the false positive group, while differences in item endorsements between the true and false negative groups were scant. Based on these results, the true and false positive groups tended to be more easily discriminated from one another while the true and false negative groups presented themselves similarly. The implications of these findings are discussed. © 2012 Elsevier Ltd. All rights reserved.