A systematic review on communicating with patients about evidence

Abstract

Objective: To conduct a systematic search for (1) the effectiveness of evidence-based communication tools to increase patient understanding of evidence, (2) effective formats for representing probabilistic information and (3) effective strategies for eliciting patient preferences about evidence. A case scenario is used to illustrate some of the difficulties of putting these results into practice.

Data sources: Systematic search of The Cochrane Library, Medline, Psychinfo, Embase and Cancerlit.

Review methods: Systematic reviews of randomized controlled trials (RCTs) and high quality RCTs were included. Studies were excluded if they did not address the question, were focused on behavioural outcomes without attempting to increase understanding, were concerned with counselling as a therapeutic intervention, or were specific to communication regarding clinical trial participation.

Results: We found 10 systematic reviews of RCTs and 30 additional RCTs addressing our questions. Communication tools in most formats (verbal, written, video, provider-delivered, computer-based) will increase patients' understanding but are more likely to do so if structured, tailored and/or interactive. Probabilistic information is best represented as event rates (natural frequencies) in relevant groups of people, rather than words, probabilities or summarized as effect measures such as relative risk reduction. Illustrations such as cartoons, or graphs (vertical bar charts) appear to aid understanding. Values clarification exercises may be better than standard utility techniques for eliciting preferences in individual decision making. Looking for effective evidence-based communication tools for prostatic specific antigen testing highlighted the challenges for clinicians and consumers in accessing tools that are evidence-based in design as well as content.

Conclusion: There is an increasing body of evidence supporting the design of effective evidence-based communication tools but variable access to such tools in practice.