I'm sometimes asked how I make sure I'm using the "best techniques" in ABA to teach children. Since autism is a spectrum disorder, what's the best way to individualize a behavioral treatment program for a child with autism? Here's my response:

1. Start with the long-term outcome data from peer-reviewed research. Lovaas, 1987; Sallows, 2005; and Cohen, 2006 resulted in 29-48% of the children reaching best outcome (i.e., testing within the normal range on IQ and completing first grade in a regular education classroom without the help of a 1:1 aide). This research included 2-3 years of intensive behavioral treatment and each study provides a basic summary of the treatment procedures used. This research is the most extensive and rigorously controlled research published to date. Thus, it stands to reason that the treatment procedures described in this research should serve as the foundation of any evidence-based behavioral treatment program.
2. Consider other short-term, single-subject design research. The long-term research not only identified a group of children who reached best outcome, but also demonstrated that a majority of children make significant progress through behavioral treatment. However, the results, while encouraging, also show that not all children benefited to the same extent from the treatment. Many professionals in the field of applied behavior analysis have attempted other strategies and procedures to try and increase a child's ability to learn a particular skill (either more quickly or when other procedures have failed). This solid research should be considered when a particular child demonstrates difficulty in learning new skills. More research is needed to help clarify which procedures may work best, based on the type of skill being taught and a specific child's strengths and weaknesses.
3. Continually assess and consider the documented progress of the particular child. Data collection is a critical component of applied behavior analysis. Data needs to be taken frequently enough and provide enough information so that particular difficulties (e.g., discrimination errors, non-responsiveness, difficulties with retention) can be quickly addressed.
4. Consider the practical experience of other behavior analysts. Practical experience of other behavior analysts who work with children with autism can also provide insight into particular prompts, creative programming, or other strategies that worked with one child and may also prove beneficial for another child. The Lovaas Institute hosts an email discussion group that was created to keep its behavior consultants throughout the country in contact with each other, as well as to gain insight from other behavior analysts at replication sites throughout the world. ABA International's annual convention or local ABA chapters are other opportunities to meet and discuss practical interventions with colleagues.
5. Consider your individual experience in providing ABA treatment with children with autism. The work of a behavior consultant requires more than just book knowledge. The Lovaas Institute typically requires at least 2-3 years of supervised, full-time experience in implementing ABA therapy plus advanced training before a staff member learns to design programs for a child with autism on their own. The Behavior Analyst Certification Board requires at least 1500 hours of supervision to obtain certification. And, the Autism Special Interest Group of ABAI has created "Revised Guidelines for Consumers of Applied Behavior Analysis Services to Individuals with Autism and Related Disorders" that demonstrate the breadth of both knowledge and experience considered necessary to competently implement behavioral treatment.

In my opinion, the above 5 factors, listed from greatest to least, demonstrate the important elements to consider when individualizing behavioral treatment for a particular child. Do you agree? Have I left anything out?