educational objectives must describe specific behaviors to be acquired or changed.

Research on outcomes (or whether goals of independence and responsibility have been attained) can be characterized by whether the goal of an intervention is broadly defined (e.g., “best outcome”) or more narrowly defined (e.g., increasing vocabulary, increasing peer-directed social behavior); whether the study design involves reporting results in terms of individual or group changes; and whether goals are short term (i.e., to be achieved in a few weeks or months) or long term (i.e., often several years). A large body of single-subject research has demonstrated that many children make substantial progress in response to specific intervention techniques in relatively short time periods (e.g., several months). These gains occur in many specific areas, including social skills, language acquisition, nonverbal communication, and reductions of challenging behaviors. Often the most rapid gains involve increasing the frequency of a behavior already in the child’s repertoire, but not used as broadly as possible (e.g., increasing use of words) (Watson et al., 1989). In single-subject reports, changes in some form are almost always documented within weeks, if not days, after the intervention has begun. Studies over longer periods of time have documented that joint attention, early language skills, and imitation are core deficits that are the hallmarks of the disorder, and are predictive of longer-term outcome in language, adaptive behaviors, and academic skills. However, a causal relationship between improvements in these behaviors as a result of treatment and outcomes in other areas has not yet been demonstrated.

Many treatment studies report postintervention placement as an outcome measure. Successful participation in regular education classrooms is an important goal for some children with autism. However, its usefulness as an outcome measure is limited because placement may be related to many variables other than the characteristics of a child (such as prevailing trends in inclusion, availability of other services, and parents’ preferences).

The most commonly reported outcome measure in group treatment studies of children with autism have been IQ scores. Studies have reported substantial changes in IQ scores in a surprisingly large number of children in intervention studies and in longitudinal studies in which children received nonspecific interventions. These are discussed in more detail in Chapter 7. However, even in the treatment studies that have shown the largest gains, children’s outcomes have been variable, with some children making great progress and others showing very small gains. Overall, while much evidence exists that education and treatment can help children attain short-term goals in targeted areas, gaps remain in addressing larger questions of the relationship between particular techniques and both general and specific changes.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement