Making sense of sensation in autism

Posted By News On November 25, 2013 - 4:00pm

(PHILADELPHIA) – In one of the first randomized control trials studying an intervention for sensory problems in children with autism, researchers found that occupational therapy using the principles of sensory integration (OT-SI) provided better outcomes on parent-identified goals than standard care, according to results published November 10th in the Journal of Autism and Developmental Disorders.

OT-SI therapy for autism is based on the premise that difficulty interpreting sensory information affects the ability to participate fully in everyday activities including eating, dressing, learning and play activities. Typical sensations are often perceived as distracting or distressing for children with Autism. "Some children will scream and hold their hands over their ears when the toilet flushes, or can't tolerate certain textures of clothing," says Thomas Jefferson University occupational therapist and neuroscientist Roseann Schaaf, Ph.D., from, the lead investigator on the study. "Others don't use sensation from their body to guide movements and thus have difficulty participating in active play and other movement activities. When we get improved processing of sensation, the children are often better able to participate in everyday tasks," she says. Schaaf and colleagues collaborated with Children's Specialized Hospital in New Jersey where the children were treated.

The current standard of care is behavioral training, which reinforces properly completed tasks, but can take up to 25 to 40 hours per week for up to 2 years to see a change in behavior. Instead, Schaaf and colleagues used sensory integration strategies, a form of treatment that is frequently requested by parents. This intervention identifies the type of sensory difficulties and then designs playful activities to help make sense of the sensation. Significantly less time consuming, the intervention was performed three hours per week for 10 weeks.

For example, one goal identified by parents was for their child to take a shower without becoming distressed and exhibiting overly disruptive behaviors. Whereas this behavior would be treated by a behavioral therapist by providing rewards for incremental increases in time spent in the shower, an occupational therapist would assess whether there were any sensory factors affecting this activity. The occupational therapist would assess the child's ability to tolerate the water hitting their skin, or managing the auditory, visual, tactile, and olfactory sensations during the shower, as well as whether the child was managing their body sensations—called proprioception—and use that information to design specific activities that address these difficulties. Then, the OT-SI therapist might work with the child in a large ball pit to decrease tactile sensitivity and improve body awareness. Importantly, the therapy is playful and the child is actively engaged.

"One approach is shaping the behavior. The other is addressing the sensory needs, and helping children manage them better," says Schaaf.

There has been some debate among researchers as to whether sensory integration therapy is indeed effective, in part because of lack of rigorously controlled trials. Dr. Schaaf's team addressed some of those concerns by blinding the investigators who performed the assessments from knowing which therapy the children had completed. The researchers randomized 32 children, aged 4-8, to either occupational or standard care group, and found that the children in the occupational group scored significantly higher on their goal attainment (an average of over 13 points higher). In addition, using a standardized test of functional skills, the children receiving the occupational therapy intervention also required less assistance from their parents in self-care activities and socialization with others. Problem behaviors associated with poor sensory processing also improved for the treatment group and approached significance.

"The rationale behind this approach," says Schaaf, "is that by changing how sensations are processed and integrated by the brain we help children with autism make better sense of the information they receive and therefore use it to better to participate in everyday tasks." Schaaf's group is in the process of publishing the treatment manual that describes the intervention used in this study so others can follow it. In addition, future studies will include additional outcome measures and follow children for a longer period of time to determine if changes in behavior are maintained over time without additional intervention.

These findings sound too good to be true. In my experience as a parent of a child with asd, too good to be true findings usually do not stand up to proper scrutiny and waste a good deal of family time and energy.

Dear Editor:

I recently stumbled across your article regarding the study published in the Journal of Autism and Developmental Disorders by Schaaf et al. Whether one of your writers penned this, or it was picked up “off the wire” from Thomas Jefferson University’s public relations department, I applaud your dissemination of scientific research in autism treatment. We need more of this!

As a science consumer, I am always interested in studies which purport new findings that overturn the status quo, and positive outcomes from sensory-integration therapy definitely fit that bill. As a Board member of the Association for Science in Autism Treatment, I constantly comb through article after article in an attempt to separate the wheat from chaff, providing guidance to parents of children with autism who may not possess adequate expertise and who certainly do not find themselves with excess time.

The news copy on your website describes this new study as one of the first randomized control trials and indicated that it yielded better outcomes than “standard care.” This is good news, if it continues to be demonstrated, because it would lead to another science-based tool for families impacted by autism. However, I caution the reader of your article to be careful, and implore that your science writers be equally cautious about statements that imply that this new discovery can replace behavioral training. In fact, the study does not compare a manualized Sensory Integration intervention with behavioral treatment, but rather compares it and spotty behavioral treatment against nothing.

Let me clarify why I use “spotty” to describe the behavioral treatment. Firstly, it is unclear in the manuscript exactly how many hours the students received. It appears that the experimental group received an average of 9 hours total [across 10 weeks?] but could also be interpreted as 9 hours/week. The control group received a reported 23 hours [again, weekly or total?], but did indicate one participant reported 240 hours of support in the school. With this amount excluded, the control group either received a total of 7.8 hours [total] or 7.8 hours weekly. The recommended amount of behavioral treatment to address symptoms of autism tends to be between 20-40 hours a week. Neither group received that level of behavioral treatment. This, in and of itself, is not a problem. Generally, treatments are tested against their absence, rather than compared to a well-established treatment. Only after it has been demonstrated and replicated that an approach does provide improvement would a research team then begin comparative evaluations.

Unfortunately, the way your web article was written, it sounds as if the study compared OT-SI package to standard care [which is later defined in your article as 25-40 hours/week of behavioral services]. This could lead a casual reader to believe that the study demonstrated that OT-SI provides improvements beyond what is gained by 25-40 hours of behavioral treatment. Not only is this inaccurate, it may result in parents opting for a sensory integration approach instead of behavioral treatment, a choice with no science supporting it. It is especially worrisome given the inaccurate statement that behavioral treatment requires two years before any gains are observable. This is patently false as there are a myriad of studies that show socially significant improvement (on the order of months) from a behavior analytic approach.

As a scientist, I am always excited to see new studies examine treatment for autism with scientific rigor. It is very important that the results of each are conveyed in a manner that tempers the hope for a breakthrough with the reality of each study’s limitations. In a world in which everything is valued for convenience and immediacy, I understand that people want results now; but rushed science is generally botched science and we are likely going to find solid science-based treatment through a very systematic, methodical approach. While it may feel like advances move at glacial speeds, the benefit this approach will bring is incredible. Families affected by a diagnosis of autism deserve no less.

Joshua K. Pritchard, PhD, BCBA
Board Member, Association for Science in Autism Treatment

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Allowed HTML tags: <h> <a><em><img><strong><iframe><table><object><cite><p><br><i><b><center><ul><li><div><html5:figure><html5:figcaption><td><tr>
  • Lines and paragraphs break automatically.

More information about formatting options

Sorry, we know you're not a spambot, but they're out there