Newcastle University

Project:

Decision Analysis in Routine Treatment II (DARTS II): A randomised controlled trial (efficacy study) of a decision aid to support shared decision making for patients with atrial fibrillation.

From November 2001 to December 2004
Project Leader(s): Prof Richard Thomson, Prof Martin Eccles, Dr Nick Steen
Staff: Mrs Jane Greenaway, Ms Lynne Stobbart
Contact: Prof Richard Thomson
Sponsors: Wellcome Trust HSR Project Grant (£74,482 + £24,312)

Background - Patients with non-valvular atrial fibrillation (NVAF) are at increased stroke risk; this is reduced by aspirin or warfarin - treatments with different harm/benefit profiles. The decision is sensitive to patient preferences.

Aim - To determine the efficacy of a computerised decision support tool when deciding on whether to treat patients with atrial fibrillation with warfarin.

Design - Efficacy RCT

Methods - Patients over 60 with NVAF, recruited from general practice, randomised to one of two clinics: the first (N=53), using the decision support tool within a shared decision making consultation, the second (control, N=56) giving a doctor-led recommendation using paper-based guidelines. The primary outcome measure was the decision conflict scale. Secondary outcome measures included anxiety, knowledge, decision making preference and the treatment decision.

Results - Decision conflict was lower immediately after the clinic in the intervention than control arm (1.89, SD 0.43 compared to 2.07, SD 0.43, mean difference -0.18 95% CI -0.34 to – 0.01). In patients not already on warfarin, the proportion of subjects who made decided to start warfarin was significantly lower in the intervention arm (4/16, 25% compared to control 15/16, 93.8%, RR 0.27, 95% CI 0.11 – 0.63).

Conclusions - A computerised decision aid applied in a shared decision making context produces lower post-intervention decision conflict than doctor-led guidelines advice, consistent with studies of similar aids. The differences in decisions made was marked in those who were not already on warfarin. This suggests that patients’ decisions, where the treatment has marked potential for harm, may be more conservative with balanced presentation of harm and benefits.

Conference Presentations

• R Thomson. Invited Seminar Leader, 7th European Forum on Quality Improvement in Health Care, Edinburgh, "Shared decision making and professional education: introducing choice to patients" March 2002
• R Thomson. Invited Short Course Leader, 8th European Forum on Quality Improvement in Health Care, Bergen, "Shared decision making and professional education: introducing choice to patients" May 2003
• R Thomson. Invited workshop 4th Biennial Sunderland Stroke Meeting, “Shared decision making: anticoagulation in AF”, Sunderland, June 2003
• Tim Rapley ‘Ordering contingency: some observations on how not to 'confound' a randomized controlled trial’ Paper presented at the International Institute of Ethnomethodology and Conversation Analysis - Producing Local Orders Conference, Manchester Metropolitan University, July 2003
• R Thomson. Invited speaker, Annual Assembly of the Swiss Society of Internal Medicine, “Decision analysis”, Basel, Switzerland 2003
• R Thomson, M Murtagh, T Rapley. Workshop on Shared Decision Making 2nd International Shared Decision Making Conference, Swansea, September 2003
• MJ Murtagh, M Eccles, R Graham, B Heaven, E Kaner, CR May, T Rapley, N Steen, L Stobbart, RG Thomson. 'Cessation of one arm of a randomised control trial (RCT) of a complex intervention: a case study of the value of contemporary observational data' Paper presented at Society for Social Medicine: 2004 Annual Scientific Meeting, Birmingham, United Kingdom, September 2004