Interventions for Individuals on the Autism Spectrum and How Best to Evaluate their Effectiveness
Written by Bob LaRue, PhD, BCBA-D
Associate Director of Behavioral and Research Services
Douglass Developmental Disabilities Center, Rutgers University
& Amy Hansford, BA,
Autism is a complex disorder characterized by significant deficits in social reciprocity and communicative ability as well as the presence of repetitive behavior/restricted interests. Given the heterogeneous nature of autism, many interventions have emerged with claims of effectiveness. Applied Behavior Analysis (ABA), sensory integration, specialized diets (e.g., gluten and/or casein free diets), pharmacological interventions (e.g., Risperdal), and mercury detoxification procedures are among the most commonly-used interventions with learners on the autism spectrum. However, with the exception of ABA, many of these treatments persist in the absence of scientific data supporting their use. This absence of data is particularly problematic for parents and practitioners who are trying to provide the best possible intervention for their learners with autism. The prospect of selecting an appropriate treatment for autism can be overwhelming for anyone with the vast amount of conflicting information available.
Most experts agree that the best first line of treatment is an intensive, coordinated program of special education and behavior management. Developmentally appropriate intervention programs generally include a language-based curriculum, systematic intervention to improve communication and social skills and a structured plan to address maladaptive behavior. Behavioral intervention strategies, derived from ABA, have the most empirical support for their use. In addition, ABA has been endorsed by the U.S. Surgeon General (1999), National Institutes of Health (NIH) and the National Standards Report published by the National Autism Center (2009).
Non-behavioral treatments can generally be divided into two main categories: biological and non-biological interventions. Biological interventions include treatments such as Hyperbaric Oxygen Therapy (HbOT), vitamin therapy, specialized diets (e.g., gluten-casein free diets, Feingold diet) and psychotropic medication. With the exception of psychotropic medication research, sound empirical evidence supporting the effectiveness of these interventions for learners with autism is sparse.
Of note, several studies have shown that certain psychotropic medications (e.g., Risperdal) can decrease some aberrant behavior (e.g., impulsivity, aggression). Non-biological interventions include treatments such as Sensory Integration Therapy (SIT), Facilitated Communication and Craniosacral Therapy.
While behavioral intervention is a good, empirically-supported starting point, it would not necessarily preclude the use of alternative interventions, if the use of such alternative interventions was carried out in a careful manner using operational definitions of targets, continuous data collection and a strong design to assess efficacy. Rejecting other types of intervention without understanding and evaluating them is problematic for several reasons. First, the absence of supporting evidence is not the same as evidence against an intervention. One of the main problems with the new and emerging treatments for autism is that limited research has been done to either validate or invalidate these interventions. Complicating this further is the fact that scientific journals do not typically publish negative research findings. In other words, research indicating that particular treatments are not effective may not be accepted for publication because of a journal‘s bias to only print positive findings. Another complicating factor is the individual variability of response to different treatments for autism. As mentioned previously, autism is a complex disorder with a variety of different etiologies that may affect response to treatment. Strategies may work for some learners and not others. It is important for practitioners to approach the intervention process with an objective mind. Rather than dismissing alternative approaches to intervention, practitioners should take the opportunity to study these interventions and systematically eliminate components that are not effective. Unfortunately, alternative interventions are often carried out in an unsystematic manner which precludes a valid assessment of outcome.
Alternative therapies are going to be part of the autism treatment landscape for years to come. It is the responsibility of both parents and practitioners to become informed consumers of autism services. It is important to critically review the validity of proposed treatments, read the literature (peer-reviewed articles rather than testimonials or anecdotal reports) and look for scientific evidence of benefit, potential health risk, and financial or time cost of all treatments. While the temptation to "leave no stone unturned" may seem appealing, wasting precious time with ineffective interventions may prevent learners from maximizing their potential.
In addition to educating themselves about alternative interventions, practitioners and parents should critically evaluate the effectiveness of the intervention process. It is often the case that people do not use systematic designs to assess the effectiveness of alternative treatments. Implementing treatments in a reversal (ABAB) design or multi-element designs can provide valuable information regarding the efficacy of treatments. Furthermore, a variety of behavioral assessment and measurement procedures, such as observational data coding, preference and reinforcer assessments, and functional analyses, can be viable ways to evaluate the effects of these interventions. Using behavioral measures for the purpose of evaluation can allow for testing some of these alternative interventions to determine if there is a clinically significant response. Such procedures could allow parents and practitioners to only use interventions that are effective and discontinue components that are ineffective. These measures can be used to determine if treatment effects generalize to different settings and also to monitor the occurrence of side effects.
In summary, there exists a wide variety of behavioral and non-behavioral interventions for learners on the autism spectrum. Some of these interventions have garnered empirical support (i.e., ABA and some psychotropic medications) while many have not been studied thoroughly enough to validate the effectiveness for use with learners with autism. Unfortunately, many treatments that are ineffective persist because they often promise miraculous results and are easy to use. The best solution is for parents and practitioners to educate themselves regarding these alternative therapies and critically evaluate the effectiveness of these interventions. In doing so, parents and practitioners can ensure the best possible intervention for their learners, prevent time from being wasted and can ultimately help others through the process by sharing their findings with others.