Putting a Dead Horse in a Weighted Vest: Another Review of Sensory Integration Training

Written by Thomas Zane, PhD, BCBA-D
Director of Applied Behavior Analysis Online Programs
Institute for Behavioral Studies at The Van Loan School of Graduate and Professional Studies, Endicott College

Sensory Integration Therapy (SIT) and similar sensory-based interventions have long been common but controversial treatments for autism (Green, Pituch, Itchon, Choi, O’Reilly, & Sigafoos, 2006; Interactive Autism Network, 2008). Scientific reviewers contend that these treatments remain unproven (Arendt, MacLean, & Baumeister, 1988; Daems, 1994; Hoehn & Baumeister, 1994; Miller, 2003; Parham, et al., 2007; Smith, Mruzek, & Mozingo, 2005), yet interest in using these interventions continues to grow (Schaaf & Davies, 2010), with occupational and physical therapists serving as primary advocates. Given their continuing popularity, the purpose of this article is to comment on the most recent review of the evidence to appear in the American Journal of Occupational Therapy(AJOT), a major repository of research on SIT and similar sensory-based treatment approaches.

In this review, Case-Smith and Arbesman (2008) examined different types of occupational therapy interventions, including “Sensory Integration and Sensory-Based Interventions,” involving strategies such as Auditory Integration Training (AIT), brushing, and massaging. The authors categorized the quality of each study using the following rubric: Level I research involved randomized-controlled trials, systematic reviews, and meta-analyses; Level II research involved clinical trials that were not randomized (such as static groups); Level III research involved simple A-B or before-after designs, which, along with Level II research, can never prove cause and effect (e.g., Fraenkel & Wallen, 2009).

Case-Smith and Arbesman found only 8 studies that qualified for Level I research, of which only one pertained to SIT (others focused on auditory integration techniques and massage). This study was a systematic review of prior research on the effects of sensorimotor interventions, including (but not limited to) SIT. They found studies reporting positive effects of SIT on “modulation” (i.e., controlling arousal, decreasing sensitivity to stimuli) and participating in social interactions. However, they described these studies as having scientifically weak designs (categorized as Levels II and IV), which did not prove a causal relationship between an intervention and behavior changes. Accordingly, they concluded that evidence of the effectiveness for SIT for improving modulation and social interaction is “inconclusive at this time” (p. 418).

Case-Smith and Arbesman expressed the view that there was more compelling evidence for the beneficial effects of SIT and similar approaches on inappropriate behaviors, hyperactivity, self-stimulatory behaviors, attention, and focus. They concluded that SIT (including, at least in this instance, interventions such as therapeutic touch) was related to improvements in these areas. As corroboration, they cited only two Level I studies. Field et al. (1997) used experimental and control groups to test the effect of touch therapy on off-task behavior, touch aversion, and social withdrawal, as measured by classroom observations, teacher ratings, and a test of social communication. Participants in the experimental group received touch therapy that involved a 25-step protocol (e.g., stroking the leg, brushing the cheek, etc.). Participants in the control group sat on the lap of a volunteer student and played a game. The researchers reported that, at the end of the study, the experimental group outperformed the control group on all three outcome measures and concluded that touch therapy was causally related to improvement in the experimental participants.

However, Case-Smith and Arbesman cautioned that the study was very short-term (four weeks), that the rate of interobserver agreement in scoring the behavior observations was poor (lower than the commonly accepted 80% level) and that the accuracy of implementation of the 25-step touch procedure was not assessed. This is a crucial omission, since the procedure was quite complicated. Last, the authors admitted that the results were confounded by the possibility of increased attention to the participants by the researchers who conducted the procedures. For these reasons, it is questionable whether improvement was necessarily due to the touching therapy.

Escalona, Field, Singer-Strunck, Cullen, and Hartshorn (2001) conducted the other study that, according to Case-Smith and Ardesman, offered Level I evidence of SIT effectiveness. Escalona et al. (2001) compared an experimental group that received nightly massages from their parents before bed to a control group whose parents read to their children nightly before bed. The dependent measures were parent ratings of activity level and sleep, as well as behavior observations at school of positive response to touch, on-task behavior, stereotypical behavior and social relatedness. Although the experimental participants improved on these measures more than the control participants, the study is fraught with problems. For example, the dependent variables had no established reliability; thus, there could have been measurement error. Parents completed the sleep diaries, which is a potential problem, given the subjective nature of the information and that the parents knew what condition their children were in, potentially skewing their recordings as well.

Overall, the most recent review published in AJOT (conducted by occupational therapists who are proponents of SIT), find inconclusive results on SIT and other sensory-based interventions. The continued belief by some caregivers and parents that SIT is an effective treatment (e.g., Miller, Coll, & Schoen, 2007) brings to mind the old adage that “one is entitled to one’s own opinion, but not to one’s own facts.” The fact is that the evidence about effectiveness of SIT and similar interventions remains inconsistent at best. Furthermore, the interventions that seem most common in treating persons with autism, such as weighted vests, brushing, swinging and joint compression, have few, if any, well controlled research indicating their effectiveness. At this time there is simply no proof that SIT and similar interventions can promote improvement in behavioral or social functioning of persons with autism.


Arendt, R., MacLean, W. E., Jr., & Baumeister, A. (1988). Critique of sensory integration therapy and its application in mental retardation. American Journal of Mental Deficiency, 92, 401-411.

Baranek, G. T. (2002). Efficacy of sensory and motor interventions for children with autism. Journal of Autism and Developmental Disorders, 32, 397-422.

Case-Smith, J., & Arbesman, M. (2008). Evidence-based review of interventions for autism used in or of relevance to Occupational therapy. American Journal of Occupational Therapy, 62(4), 416-428.

Daems, J. (Ed.). (1994). Reviews of research in sensory integration. Torrance, CA: Sensory Integration International.

Escalona, A., Field, T., Singer-Strunck, R., Cullen, C., & Harts-horn, K. (2001). Brief report: Improvements in the behavior of children with autism following massage therapy. Journal of Autism and Developmental Disorders, 31(5), 513-516.

Field, T., Lasko, D., Mundy, P., & Henteleff, T. (1997). Brief report: Autistic children’s attentiveness and responsivity improve after touch therapy. Journal of Autism and Developmental Disorders, 27(3), 333-338.

Fraenkel, J. R., & Wallen, N. E. (2009). How to design and evaluate research in education. (7th ed.). New York: McGraw-Hill.

Green, V. A., Pituch, K. A., Itchon, J., Choi, A., O’Reilly M., & Sigafoos, J. (2006). Internet survey of treatments used by parents of children with autism. Research in Developmental Disabilities, 27,70-84.

Hoehn, T. P., & Baumeister, A. A. (1994). A critique of the application of sensory integration therapy to children with learning disabilities. Journal of Learning Disabilities, 27, 338-350. Interactive Autism Network (2008). Retrieved November 29, 2010 at http://www.iancomunity.org/cs/ian_treatment_reports/occupational_therapy.

Miller, L. J., Coll, J. R., & Schoen, S. A. (2007). A randomized controlled pilot study of the effectiveness of occupational therapy for children with sensory modulation disorder. The American Journal of Occupational Therapy, 61(2), 228-238.

Miller, L. J. (2003). Empirical evidence related to therapies for sensory processing impairments. Communique, 31, 34-37.

Parham, L. D., Cohn, E. S., Spitzer, S., Koomar, J. A., Miller, L. J., Burke, J. P., Brett-Green, B., Mailoux, Z., May-Benson, T. A., Roley, S. S., Schaaf, R. C., Schoen, S. A., & Summers, C. A. (2007). Fidelity in sensory integration intervention research. The American Journal of Occupational Therapy, 61(2), 216-226.

Schaaf, R. C., & Davie, P. L. (2010). Evolution of the sensory integration frame of reference. The American Journal of Occupational Therapy, 64(3), 363-367.

Smith, T., Mruzek, D. W., & Mozingo, D. (2005). Sensory integrative therapy. In J. Jacobson, R. Foxx, and J. Mulick (Eds.), Controversial Therapies for Developmental Disabilities: fad, fashion, and science in professional practice. Pergamon Press.

Please use the following format to cite this article:

Zane, T. (2011). Putting a dead horse in a weighted vest: Another review of sensory integration training. Science in Autism Treatment, 8(1), 18-19.