Activities to support the implementation of complex interventions as part of routine care: a review of the quality of reporting in cluster randomised controlled trials. Issue 10 (9th October 2015)

• Record Type: Journal Article
• Authors: McMahon, Naoimh E; Holland, Emma-Joy; Miller, Colette; Patel, Kulsum; Connell, Louise A
• Is Part Of: BMJ open. Volume 5:Issue 10(2015)
• Publication Date: 2015-10-09
• Subjects: Medicine -- Research -- Periodicals; 610.72
• DOI: 10.1136/bmjopen-2015-008251
• Languages: English
• ISSNs: 2044-6055

Abstract

Abstract : Objective: To review a sample of cluster randomised controlled trials and explore the quality of reporting of (1) enabling or support activities provided to the staff during the trial, (2) strategies used to monitor fidelity throughout the trial and (3) the extent to which the intervention being tested was delivered as planned. Design: A descriptive review. Data sources and study selection: We searched MEDLINE for trial reports published between 2008 and 2014 with combinations of the search terms 'randomised', 'cluster', 'trial', 'study', 'intervention' and 'implement*'. We included trials in which healthcare professionals (HCPs) implemented the intervention being tested as part of routine practice. We excluded trials (1) conducted in non-health services settings, (2) where the intervention explicitly aimed to change the behaviours of the HCPs and (3) where the trials were ongoing or for which only trial protocols were available. Data collection: We developed a data extraction form using the Template for Intervention Description and Replication (TIDieR checklist). Review authors independently extracted data from the included trials and assessed quality of reporting for individual items. Results: We included 70 publications (45 results publications, 25 related publications). 89% of trials reported using enabling or support activities. How these activities were provided (75.6%, n=34) and how much was provided (73.3%, n=33) were the most frequently reported items. Less than 20% (n=8) of the included trials reported that competency checking occurred prior to implementation and data collection. 64% (n=29) of trials reported collecting measures of implementation. 44% (n=20) of trials reported data from these measures. Conclusions: Although enabling and support activities are reported in trials, important gaps exist when assessed using an established checklist. Better reporting of the supports provided in effectiveness trials will allow for informed decisions to be made about financial and resource implications for wide scale implementation of effective interventions.