2007 - 2013
The NDTA Research Committee is pleased to present the following summaries of research articles relating to NDT that were published in peer-reviewed journals from 2007 to 2013. While an in-depth critical appraisal of the studies was beyond the scope of this project, we noted multiple methodological issues with study design and statistical analysis in many of these studies. These included issues with internal validity, generalizability of results, choice of statistical tests, choice of outcome measures, and fidelity of treatment.
When study procedures do not control well for threats to internal validity, there is a risk that something other than the study intervention was responsible for the results. Examples of these kinds of issues in the studies summarized here include lack of equivalency between groups (e.g., in participant severity, amount of treatment), un-blinded outcome assessment where the study personnel conducting the assessments were aware of the participants’ group assignments, and lack of randomization to a comparison group.
There were also concerns about study procedures that limit the generalizability of findings beyond the study. For example, there are many studies where the sample size was relatively small for the characteristics of the population being studied. A small sample size increases the chance that repeating the study with a different sample would not yield the same results. Thus, it is possible that a study with a small sample size might find a difference between groups when one does not exist or, conversely, not find a difference when one does exist. There is an even greater risk of this happening when the population being studied has a great deal of variability as is the case with the major populations of interest here (e.g., cerebral palsy, stroke, and traumatic brain injury). In more variable, heterogeneous populations such as these, sample sizes often need to be in the hundreds to have confidence in the generalizability of the results.
There were some studies where the choice of statistical tests did not seem appropriate for the study design. This included the use of statistical tests designed for large samples in studies with small sample sizes. In other studies, outcomes measures used to identify change in response to NDT and/or other procedures were inappropriate for the subject population or to capture the types of change likely to occur as a result of NDT. There were concerns in some studies about the fidelity of the NDT intervention, i.e., how closely the treatment that was given adhered to NDT principles and intervention strategies. This was especially a concern for those studies that relied on a manual to train interventionists in the NDT approach.
In reviewing these summaries, we recommend that the reader evaluate the studies carefully for these types of issues. In order to assist with that process, we have added editor notes to a few of the studies to point out examples of concerns; however, reviewers’ comments are not intended to constitute a complete analysis of the articles.
The NDTA Research Committee
Nancy S. Darr, PT, DSc, NCS
Kathleen Ganley, PT, PhD
Janet M. Powell, PhD, OTR/L
Fran Redstone, PhD, CCC-SLP, C/NDT
Debbie Evans-Rogers, PT, PhD, PCS, C/NDT
Marcia Stamer, PT, MH, C/NDT