Treatment fidelity: relation to treatment acceptability and change over time.

Abstract

The purpose of this study was: (a) to examine the relation between acceptability and fidelity of an intervention package in natural classroom settings, and (b) to examine how fidelity of implementation varies in relation to high vs. low treatment acceptability over the course of an academic school year. Participants were 44 teachers, from 15 schools, working with students at-risk for or diagnosed with emotional and behavioral disorders (EBD). After two weeks of training, teachers implemented the multi-component intervention package (Classroom Organization and Management Program, Good Behavior Game, and Self Monitoring) in their classrooms. Once a week, trained observers conducted direct observation of treatment fidelity and offered feedback to teachers. The study was carried out for an academic school year with fidelity data collected for an average of 15.5 weeks (excluding training, school breaks, and missing data). At the end of the school year, teachers filled out acceptability ratings for the combined and individual components of the intervention package. The relation between treatment acceptability and fidelity of implementation was studied by examining the fidelity of implementation on weeks 2, 7, and 12, and the total treatment acceptability score. The findings of Pearson correlations revealed a significant but weak relation between teachers' ratings of acceptability and their fidelity of implementation (ranging from .28 to .53). The results of Repeated Measure Analysis of Variance (RM-ANOVA) revealed that teachers' fidelity of implementation did not significantly change over time in relation to their high vs. low ratings of treatment acceptability. Finally, the findings revealed that within the context of an efficacy trial, with extensive training and weekly consultation, the fidelity of implementation was sustained over time. Results are discussed in terms of implications for school consultants, limitations and future research.