PMID 16685181

Early Intensity Behavioral Treatment: Replication of the UCLA Model in a Community Setting

Cohen H, Amerine-Dickens M, Smith T.

Background

The UCLA/Lovaas model was conducted with great success in an academic setting using Early Intensity Behavioral Treatment (EIBT), with 9 of the 19 children showing sustained improvements in intellectual achievement. Several studies have tried to replicate these findings in a community setting, but none have truly replicated the methods involved in the study and several different findings have been reported.

Introduction Goal: To fully replicate the UCLA/Lovaas study in a community setting.

Methods There were 21 children in the treatment group, which received EIBT for 35-40 hours per week, 47 weeks per year for 3 years. There were 21 children in the comparison group who received services from local public schools. Participant qualifications for this study were:

Diagnosis of autism or PDD-NOS Pretreatment IQ of 35 or greater Age between 18 and 42 months Under 48 months at treatment onset No severe limitations in health Residence within 60km of treatment clinic 400 hours or less of prior treatment Participant parent agreement

This experiment used a quasi-experimental design because children could not be randomly assigned to their groups. Intervention consisted of in-home instruction, peer play, and regular education classes.

Results IQ: Significant difference between comparison groups, F=5.21, p<.05. Merrill-Palmer: No significance Reynell Language Comprehension: Significant difference, F=3.82, p<.06 Expressive Language: No significance VABS Composite: Significance found, F=7.84, p<.10 Communication: Significance found, F=5.45, p<.05 Daily Living: Significance found, F=6.40, p<.05 Socialization: Significance found, F=4.03, p<.10

Conclusion Hypothesis was not confirmed. There was no significant group x time interaction as the EIBT treatment group scores did not significantly improve over time versus the comparison group, though the treatment group's success in school placement was significantly more than the comparison group.

Discussion Limitations: One significant limitation of this study was that the groups were not randomly assigned. Another limitation of this study is that the assessment protocol was not consistent or rigorous enough.