For which younger siblings of children with autism spectrum disorder does parent-mediated intervention work?

The link between initial cumulative risk and the intervention effect is correlational, so measured or unmeasured pre-treatment variables could explain why the effect appeared only in the no additional risk subgroup. The researchers could not predict before the study which end of the cumulative risk scale would benefit, so they advise replication before clinical use of the significant findings. Fewer than 10% of families had more than one sibling with ASD, so cumulative risk in this sample mostly reflected sex and behavioral screening risk. The ASD diagnosis result was a post-hoc analysis and was not expected to have sufficient power because the outcome was dichotomized. Some indirect effects on social communication became nonsignificant when influential cases were deleted. The cumulative risk index produced 41% false positives and is not sufficient evidence of ASD or language delay by itself. Null findings for parenting stress and efficacy could reflect insensitive measurement rather than absence of effect.

The supplied metadata lists NIH extramural research support. The article text provided does not include an author conflict-of-interest declaration.

Do not conclude that ImPACT prevents autism or works for all high-risk siblings. The autism diagnosis result came from an after-the-fact analysis and occurred only in siblings with no additional cumulative risk. The intervention did not affect parenting stress or efficacy.