Non-verbal IQ and change in restricted and repetitive behavior throughout childhood in autism: a longitudinal study using the Autism Diagnostic Interview-Revised.

The study used parent-reported ADI-R answers, so a behavior could have been present but not reported, or reported differently at different ages. The analysis looked only at whether each behavior was present or absent, not how often it occurred or how severe it was. Fewer children were assessed at age 11 than at diagnosis, so later childhood patterns depend on which families stayed in the study. Only 28 girls were included, and sex was not analysed, so the findings may not apply equally to autistic girls. Non-verbal IQ was estimated from one WISC-IV index at age 8, and conventional IQ testing may underestimate ability in autistic children, especially those with limited speech. The study was observational and did not assign or test any therapy, so it cannot show that a treatment caused behavior change. Some item-level findings appeared only with less strict multiple-comparison corrections, so they are less certain than the strictly corrected findings. The paper did not analyse adaptive abilities or other developmental indicators alongside age and IQ.

The supplied text does not include an author conflicts-of-interest statement. The article metadata labels the work as supported by non-U.S. government research.

Do not read the age-related decreases or increases as evidence that therapy caused them. This was an observational study of parent-reported behavior, and no treatment was assigned or tested.