Short-Term Effects of a Caregiver-Mediated Intervention for Preschoolers at Risk for Developmental Language Disorder: A Randomized Controlled Trial.

About 60% of children in both groups were also receiving community-based early intervention during the trial, so the intervention was added on top of usual care rather than replacing it, and some of the group progress may reflect that. Roughly 80% of families in the final sample were White and non-Hispanic, so the results may not carry over to the wider population of children receiving early language intervention. The tests used do not measure what the intervention actually taught: the vocabulary tests include few action words and none of the location or description words targeted, and the grammar tests emphasise later-developing tense and agreement features rather than early simple sentences. Test administration was shortened to reduce fatigue (stopping after a run of wrong or no answers), and the tense probes were only given to children who passed a phonology screener, so the grammar effects may be understated and do not cover the whole sample. When the grammar analysis was repeated using only children with complete data, the treatment effect was positive but no longer statistically significant. These are short-term results only, at the mid-points of the intervention; outcomes at the end of the full 18 months are not yet reported.

The work was funded by the National Institute on Deafness and Other Communication Disorders (grant U01DC017135, to Ann Kaiser, Megan Y. Roberts and Pamela A. Hadley). The supplied text lists no other funder and no industry involvement. Families were given books and toy sets every three months along with monetary incentives, with particularly strong incentives for control group families.

Do not conclude that the intervention caught these children up. At 42 months the treatment group's grammar scores were still far below age expectations in both groups, and only 7 of 35 treated children scored above the criterion on the third-person singular probe. Also do not read the vocabulary result as an all-round vocabulary gain: the combined vocabulary score showed no group difference, and only the receptive single-word test moved, not the expressive one.