A Pilot Randomized Control Trial of Motor-Based Treatments for Childhood Apraxia of Speech: Rapid Syllable Transition Treatment and Ultrasound Biofeedback.

Only 14 children took part. The study was too small to detect a difference between the two treatments even if one exists, and the authors describe the results as preliminary rather than definitive. There was no untreated control group, so the improvements cannot be separated from normal development, extra attention or practice. The authors argue that school-age children with apraxia do not tend to change quickly without therapy, but that is not the same as a control group. Children, parents and the treating clinicians all knew which treatment was being given. Only the people scoring the recordings were blind. Children varied a great deal in how much they improved, which the authors attribute to how heterogeneous childhood apraxia of speech is, and this variation further limits what a small study can show. One child in the ReST group missed the one-month follow-up, and his follow-up data were filled in by assuming a return to his starting scores. That is a conservative choice, but it is not a measurement. The two arms did not end up receiving the same number of practice trials: the ultrasound sessions produced more trials on average, with wide variation between children. The authors note that dose was not deliberately manipulated, so any effect of dose is tangled up with the difference between the treatments. No parent- or child-reported outcomes and no measures of participation were collected, so nothing here speaks to whether the changes matter in daily life; the authors also note there is no agreed measure of clinically significant change in this condition. Families agreed not to seek other speech therapy during the study or for four weeks afterwards, and not to do formal speech homework, but no data were collected to check whether they kept to this.

The study was funded by a grant from the Childhood Apraxia of Speech Association of North America (Apraxia-Kids). One author was also supported by a National Institutes of Health grant (R15DC016426). No other competing interests are reported.

Do not read the improvements as proof that either treatment works. There was no untreated comparison group, so ordinary development or the extra attention and practice could explain part of the change. Do not read the even result as proof that the two treatments are equally effective either: with 14 children the study was too small to show a difference if one exists, and one child's follow-up scores were filled in rather than measured.