Assessing fidelity of delivery of the Community Occupational Therapy in Dementia (COTiD-UK) intervention

Abstract

Background: Measuring whether interventions are delivered as planned (fidelity of delivery) is key to evaluating effectiveness. Without assessing fidelity it is not possible to make conclusions about the planned intervention as intervention effects reflect the outcomes of what was delivered, not what was planned. This study aims to measure the fidelity of delivery of Community Occupational Therapy in Dementia (COTiD-UK), within a multi-site (n=15) randomised controlled trial across the UK. COTiD-UK is delivered by an occupational therapist over ten hours to the person with dementia together with their family carer, using seven key skills. Methods: This study has a longitudinal observational design. All intervention sessions are audio-recorded. 10% of sessions (n=140) will be sampled from across sites and occupational therapists (n=27). Selected sessions will be transcribed and analysed for fidelity using checklists and coding guidelines developed for this study. These will be piloted and amended through an iterative process, until good inter-rater agreement (Kappa > 0.8) is achieved. Researchers will then independently code transcripts to measure fidelity. A percentage of transcripts will be double-coded to check for coder drift. Data will be analysed using descriptive statistics. Expected results: Findings will demonstrate which components were delivered as planned. Findings will be compared across sessions, providers and sites. Current stage of work: This project is in the pilot phase. Discussion: These findings will be used to interpret the results of the randomised controlled trial of the COTiD-UK intervention.