Treatment fidelity: its importance and reported frequency in aphasia treatment studies

Am J Speech Lang Pathol. 2013 May;22(2):S279-84. doi: 10.1044/1058-0360(2012/12-0092).

Abstract

Purpose: Treatment fidelity is a measure of the reliability of the administration of an intervention in a treatment study. It is an important aspect of the validity of a research study, and it has implications for the ultimate implementation of evidence-supported interventions in typical clinical settings.

Method: Aphasia treatment studies published in the last 10 years in 3 journals were reviewed using coding techniques that were adapted from Gresham, Gansle, Noell, Cohen, and Rosenblum (1993). The following items were noted: identifying information, study design, description of both the dependent and independent variables, and whether a measure of treatment fidelity was explicitly included.

Results: Of the aphasia treatment studies published in the last 10 years, 14% explicitly reported treatment fidelity. Most studies reporting treatment fidelity used checking of videotaped sessions by independent raters. Of the reviewed studies, 45% provided sufficient treatment description to support replication.

Conclusion: Treatment fidelity is widely acknowledged as being critical to research validity and is a foundation for the implementation of evidence-based practices, but only a small percentage of aphasia treatment studies published in the last 10 years explicitly reported treatment fidelity. Recommendations for research practices include increased attention to matters of treatment fidelity in the peer review process and explicit incorporation of 3 levels of treatment fidelity in treatment research.

Keywords: aphasia; evidence‐based practice; reproducibility of results; speech‐language pathology.

Publication types

  • Meta-Analysis
  • Review

MeSH terms

  • Aphasia / therapy*
  • Evidence-Based Practice / standards*
  • Humans
  • Reproducibility of Results
  • Research Design
  • Speech-Language Pathology / methods*
  • Speech-Language Pathology / standards*