Abstract
Lovaas and his associates at UCLA reported substantially better treatment outcomes for children with autism than had been reported previously. Many authors have closely scrutinized the UCLA research, and some have published major criticisms. Recently two critical reviews have appeared. Howlin (1997) evaluated the most common treatment approaches for children with autism, including the UCLA model. Additionally, the UK Department for Education and Employment commissioned a review of treatment research on autism written by Jordan, Jones, and Murray (1998). Although both reports contain useful information, they also contain inaccurate statements about the published results of the UCLA project. An aim of this paper is to illustrate the errors of fact and interpretation contained in the two papers and to clarify the existing data. A second aim is to identify points of agreement, which is also important when attempting to facilitate critical and accurate assessments of available scientific evidence. It is argued that Lovaas and colleagues made a convincing case for the efficacy of one model of early intensive behavioral treatment by presenting outcome data showing clinically meaningful, longβterm gains in a population historically characterized by a severe and chronic disability, and an unfortunate history of treatment failures. Copyright Β© 2001 John Wiley & Sons, Ltd.