Image via Wikipedia
The principal investigator was Dr. David Mandrell from the University of Pennsylvania for the Center for Medicare and Medicaid. This scan looks at the literature for behavioral and psychosocial treatments but not biomedical ones and evaluates the studies on treatments for autism. It also looks at the quality of the studies. This study is interesting because it sheds some light on where public funding may be spent in the future. The government is concerned about how to best spend declining resources so it wants to know what treatments have been definitively proven to be efficacious and what treatments are promising and what treatments are in the experimental stage.
At the end of the day, it seems like experts who try to identify "gold standard" treatments for autism really don't have much to say definitively. When you skim to the Conclusions sections of the report, it seems that the best that the experts have to say is that there is "growing and encouraging support for behavioral interventions for young children with autism. Data are scare or nonexistant altogether, however, on the efficacy of treatments for adolescents and adults with autism."
It seems that they are having some basic problems in the design of studies:
- Nature of outcome measures used in efficacy studies: A lot of the studies don't measure outcomes of therapy in terms that address the mission of Medicaid to maximize health. Medicaid addresses health conditions. Many of the studies address cognitive and behavior conditions. My guess is that there is a lack of a proven model that shows how medical conditions translate directly into cognitive and behavioral improvements. When decision makers want to decide how to direct billions of dollars, they need to stick to their mission (in the case of Medicare/Medicaid is medical) and they really want gold standard studies that justify what they are doing as they are audited by the GAO and Congressional committees.
- Lack of data on effectiveness: There is a difference between efficacy studies done in university research settings under ideal conditions and community based practice settings (ie your doctor's office) where there are lots of different types of patients.
- Ambiguity regarding how the intervention translates into Medicaid-reimbursed services: Medicaid address health conditions, not learning conditions. If a service doesn't conform to a state's Medicaid standards, a patient doesn't get reimbursement.
- and Lack of studies examining interventions in combinations designed to address more holistically the challenges facing individuals with autism and their families.
When you start looking at "Transitioning Youth" (aka adolescents) and Adults, the number of studies decreases dramatically. For "Transitioning Youth", the Antecedent Package was rated as most efficacious. Promising treatments for adolescents did not show any that any were maintained after treatment.
For adults, Structured Employment was rated as efficacious. No promising treatments were identified.
Net result: we need more studies.