As promised, I wanted to review some of the articles in the current (March/April 2007) issue of the American Journal of Occupational Therapy. Since I have gotten some emails and blog comments asking me about the Hall & Case-Smith study (see reference below) I thought that this would be a good article to begin discussing.
I was excited to see a study on this intervention technique because I have been hesitant in the past to fully embrace 'therapeutic listening.' In my opinion there is not adequate evidence to support promoting this technique so I was interested to see if this new study would change my mind.
I believe that the authors should be commended for initiating this inquiry, but I also believe that their conclusions are far overstated. There are several fundamental flaws in their design that are concerning.
To begin with, they used a convenience sample. This might have been appropriate if the study was truly exploratory, but there have been several other studies already completed on this topic. Convenience sampling limits our ability to state that the subject pool is truly representative of the entire population.
Sampling size is also an issue - it was based on expected effects for the Sensory Profile but the authors don't discuss the sample size that would be needed for the other tests used. This information is needed in order to determine if the sample size is adequate to draw conclusions about visual motor or handwriting skills.
Finally, their study population was very heterogenous - so even if there were effects within this group it is difficult to know what was causing any of the effects based on the sampling methods used. In sum there are many problems with the sampling procedures reported that severely limit the study.
Another problem is that the measurement methodology relies strongly on parental report. This is also a confounding factor because you are asking the parents if there are changes in behavior, but the parents have invested heavily in the intervention program - and in fact they may be expecting changes based on whatever communication occured in the process of educating them about the 'sensory diet.' It would have been much more useful to have the teachers fill out the Sensory Profile (school version) because they are completely separate from the intervention and could give an opinion that has much less opportunity for bias.
Another problem with measurement is that some of the evaluation tools were administered three times within the three month study period - opening the question of practice effect of these particular tests.
It is also unclear how many of these children were participating in other therapy programs and what the nature of those programs was. The presence or absence of other parallel intervention programs is a significant confounding factor.
All of these factors could be partially acceptable if the study was truly exploratory and if the conclusions were stated more modestly. Instead, this is just another in a long line of studies with significant limitations. What makes it particularly difficult to accept is that the authors conclude that "The present study produced encouraging findings to support the use of therapeutic listening as part of an overall sensory-integrative approach to occupational therapy in elementary school-age children" (p. 215). I don't think that the study accomplished this at all. Rather, I believe that the study is very limited, they relied primarily on the potentially biased self-report of parents that participated in the program, the conclusions overstate the results, and there is very little real conversation about the severity of the problems in sampling and measurement methods.
So for me the jury remains 'out' on therapeutic listening. I don't see any reason to encourage parents to invest heavily in these programs at this time. If anyone wants to rebut my conclusions I will happily post their comments, in full, on this blog.
Hall, J. & Case-Smith, J. (2007) The effect of sound-based intervention on children with sensory processing disorders and visual motor delays. American Journal of Occupational Therapy, 61, 209-215.