From the Journals

Autism screening tests fall short

View on the News

Is the screen at fault?

The investigators’ conclusions that more sensitive autism spectrum disorders (ASD) screening tools are needed may need to be tempered.

The study authors conclude that, even among children who screen negative on the M-CHAT, those with ASD frequently display early signs and symptoms that, with more sensitive screening instruments, may enhance early detection.

While such a conclusion is logical, the study data indicate that the M-CHAT had a sensitivity of 23% in this population, which is dramatically lower than the sensitivity reported in other studies. If the sensitivity of the updated M-CHAT-R screening tool is truly 91%, as has been claimed, it is hard to argue that more sensitive screening tools are needed.

As the M-CHAT-R was developed to improve usability and decrease the false-positive rate of the M-CHAT, the difference in screening tools is unlikely to account for the low sensitivity of the M-CHAT in this population. Possibilities that may factor into the low sensitivity include timing (the M-CHATs were evaluated only at 18 months), differential follow-up across false- and true-negative screens, or population differences in the Norwegian study and previous work.

These criticisms withstanding, the current study does send a clear warning that the M-CHAT likely does not equally identify all manifestations, or clinical phenotypes, of ASD. The findings lend credence to the concern of the USPSTF (U.S. Preventive Services Task Force) that “clinical and convenience samples do not adequately demonstrate the psychometric properties of screeners in practice.” In this study, the researchers reinforce the notion that screened and clinical populations may be systematically different and that more research is needed to understand such differences.

This commentary is edited from an accompanying editorial in Pediatrics (2018;141[6]:e20180965) by Sarabeth Broder-Fingert, MD, MPH; Emily Feinberg, ScD; and Michael Silverstein, MD, MPH, of the Boston University/Boston Medical Center. Dr Silverstein is a member of the U.S. Preventive Services Task Force but speaks here for himself alone.


 

FROM PEDIATRICS

Children whose autism was not detected by the Modified Checklist for Autism in Toddlers (M-CHAT) at 18 months old were more likely to have delays in social, communication, and fine and gross motor skills at the time of the screen, compared with other children who had negative results, according to findings from a retrospective analysis of 68,197 screen-negative cases in the Norwegian Mother and Child Cohort Study.

Parents of children with false-negative M-CHAT results rated their children’s gross and fine motor skills and social and communication skills at 18 months as less developed than did parents of children with true-negative screens. For girls who had false-negative results and were later diagnosed with autism, the delays were more pronounced, compared with girls with true-negative results. Also, girls later diagnosed with autism were rated as less shy than girls with true-negative scores. Shyness was more common in boys later diagnosed with autism than in boys with true-negative scores.

“When trying to determine if a young child is exhibiting autism symptoms, clinicians should not rely solely on a single instrument but consider parental concerns and draw on other developmental surveillance instruments, as well as their clinical judgment. ... The clinicians also need to be particularly wary about discounting symptoms of social difficulties in girls because they may be masked by limited shyness or social inhibition,” wrote Roald A. Øien, MA, of the University of Tromsø (Norway) and Yale University in New Haven, Conn., and his associates in Pediatrics.

The researchers noted that the study was based on use of a previous M-CHAT version. The findings may not be relevant to the updated M-CHAT-R/F, which has 20 questions, new cutoffs, and a recommended follow-up interview.

Of the Norwegian children who were at least 40 months old at the time of the study, 67,969 had true-negative M-CHAT screens, and 228 had false-negative screens based on later diagnoses reported in the Autism Birth Cohort, a substudy of the Norwegian Mother and Child Cohort Study.

The 18-month-olds had been assessed with the M-CHAT, selected items from the Ages and Stages Questionnaire and the Emotionality Activity Sociability Temperament Survey. Of the 23 pass-fail M-CHAT items, 6 are highly predictive of a later ASD diagnosis; a positive screen is failure of at least 2 of those 6 items.

Both boys and girls with false negatives were less social and had lower communication and gross motor skills, compared with their true-negative counterparts, but these differences were greater between false-negative and true-negative girls. Fine motor skills were also significantly lower in those with false negatives than in those with true negatives, but the magnitude was no different between girls and boys.

Overall, boys had more advanced gross motor skills and higher activity levels than girls, independent of true- or false-negative status.

In post hoc analyses, boys with false-negative results were rated as more shy than boys with true-negative results. Girls with false negatives were rated as less shy than girls with true negatives and boys with false negatives. Ratings of emotionality and activity did not differ among the children with true or false negatives.

The authors speculated that girls with false negatives may have “somewhat lower levels of social fearfulness or lower inhibitory control, compared with boys.”

The authors also suggested possible reasons for the false negatives, including parents’ difficulty in matching behaviors described in the M-CHAT with their children’s behaviors and the lack of graded responses on the M-CHAT, which may influence parents’ responses.

By comparison, the Ages and Stages Questionnaire “gives parents the opportunity to express that the children exhibit skills occasionally albeit inconsistently, which may allow them to express their concerns and perceptions in a more graded manner,” the authors wrote.

Another possible reason for false negatives, the authors suggested, is that symptoms in those with autism spectrum disorder may manifest differently in early childhood, partly depending on the level of the child’s verbal and nonverbal skills.

“We believe that our results contribute, at a fundamental level, to our understanding of early screening for ASD, and we highlight the discrepancy between hard cutoff criteria for autism and the social-communicative, developmental, and temperamental signatures of emerging or subthreshold autism phenotypes,” the authors wrote. They noted a need for screens that take into account temperament and verbal and nonverbal skill levels.

The research was funded by the Norwegian Ministry of Health and Care Services, the Norwegian Ministry of Education and Research, the Research Council of Norway and Functional Genomics in Norway, the National Institute of Neurological Disorders and Stroke and the National Institute of Environmental Health Sciences. Dr Hornig coinvented an intestinal microbiome biomarker for autism which has patents assigned to Columbia University.

SOURCE: Øien RA et al. Pediatrics. 2018;141(6):e20173596.

Recommended Reading

Adolescent suicidal ideation and attempts are on the rise
MDedge Family Medicine
ADHD, asthma Rxs up
MDedge Family Medicine
ECT cost effective in treatment-resistant depression
MDedge Family Medicine
Two more and counting: Suicide in medical trainees
MDedge Family Medicine
Heart rate variability may be risk factor for depression, not a consequence
MDedge Family Medicine
Simple postural exercises may reduce depressive symptoms
MDedge Family Medicine
Drug-related deaths continue to rise in United States
MDedge Family Medicine
Behavioral sleep intervention linked to sleep improvement in infants
MDedge Family Medicine
Sulforaphane for autism? Maybe
MDedge Family Medicine
VIDEO: Pills alone not the answer for pain management
MDedge Family Medicine