Abstract
Purpose and background:
Implementation fidelity (IF) is the extent to which an intervention is implemented as intended by its developers, and increases confidence that changes in study outcomes are due to the effect of the intervention itself and not due to variability in implementation. The aim of this study was to evaluate the IF within a behaviour-change self-management intervention for people with chronic low back pain and/or osteoarthritis, consisting of six weekly sessions (SOLAS ISRCTN49875385).
Methods:
In a sample of data, the intervention was delivered by physiotherapists (n=9) in seven sites. IF was assessed using self-report (by physiotherapists) of all sessions (n=60), direct observations (by the research team) of 40% of the sessions (n=24) and audio-recorded observations (by the research team) of all sessions (n=60) using checklists. Data were analysed in SPSSv20 to assess % agreement between methods and fidelity scores.
Results:
Agreement between direct observations and audio-recordings was 86.6%, with inter-rater reliability of 84.6% of the audio-recordings. Agreement between the direct observations and self-report was 74.6%. Four sites scored above 80% (‘high’ fidelity) with three scoring between 70 and 80% (‘moderate fidelity’). Four physiotherapists scored between 83–100% with five scoring from 72.5% to 78.8%. There was no significant difference in the fidelity scores across sessions with scores ranging from 74% to 88.8%.
Conclusions:
Self-report data may be of limited use for assessing IF, though it may have benefits in terms of enhancing IF. Overall, fidelity levels ranged from moderate to high which will provide valuable information for the interpretation of the trial's clinical effectiveness.
