ASA's 36th National Conference on Autism Spectrum Disorders (July 13-16, 2005)

• Recorded Presentation

Politicians, professionals, and parents alike speak of the importance of relying on “best practices”, or those programs and techniques whose effectiveness is established by scientific research. We describe why we must turn first to best practices. We illustrate how basic questions regarding the effectiveness of an intervention can undermine even the most apparently compelling of findings. Did the children have autism? How was the impact of the intervention measured? Do we know which element(s) of the intervention were critical? Is the intervention appropriate for all children with autism?

We review some of the criteria traditionally used to evaluate the quality of research cited in support of best practice, by describing guidelines established by a variety of professional organizations. These guidelines set forth clear standards with respect to the selection of research participants, the description of the methods employed, the assessment of outcome, and the need for replication. We consider how recent commissioned reviews in the United States, Canada, and overseas have attempted to refer to such criteria when defining best practices. We also describe the novel approach recently adopted in Delaware for evaluating the evidence in support of proposed new practices.

The controversy regarding Lovaas / IBI programs has often turned on the question of whether these constitute best practice. We describe this controversy, how it has sharpened the debate about best practices, and how it illustrates both the opportunities and potential pitfalls of turning to scientific research.

Yet even with clearly defined criteria, it can be difficult to find “best practices” if you are not a researcher with access to reliable information and trained to distinguish between valid and invalid research. Moreover, many potentially valid practices do not meet the traditionally rigorous standards described above. As a result, parents and educators are left without identified “best practices” for critical areas of teaching. We therefore review possible criteria for establishing practices that are “good enough” when best practices have yet to be clearly established. These include careful evaluation of the quality of the evidence available, the identification and use of authoritative reviews of the field, and some guidelines for weighing the costs and benefits of “good enough practices”. We present concrete guidelines for assessing the extent to which a given technique is consistent with “good enough practice”.

Under No Child Left Behind, public schools are now expected to employ best practices. What does this mean for educators faced with a plethora of choices? What does this mean for parents who are dissatisfied with their child's program and are seeking alternatives? We examine the standards set by NCLB and precedents set by case law, and the implications for parents and professionals.

Best practices are useless, however, if they are not supported by a comprehensive training protocol. We briefly describe some of the elements parents and professionals must consider when determining if training can be effectively implemented. We also provide some concrete guidelines for designing a comprehensive training program when a protocol is not readily available. .

Best practices are also ultimately limited in their effectiveness if they are not delivered by “best programs”. We conclude by describing some of the features of effective educational programs identified in recent reviews – e.g., those that include comprehensive training and/or service delivery models specialized for individuals with autism. We provide concrete guidelines for assessing the extent to which a given program is consistent with these criteria.