Article Text
PDF
Statistics from Altmetric.com
Introduction
The work of transforming scientific publications into tools that can support people in comparing treatments, tests and other interventions has been driven by efforts to support shared decision-making. The International Patient Decision Aids Standards (IPDAS) Collaboration has published many articles that guide this process. IPDAS, and others in this field, have considered the challenges of representing complex concepts such as effect sizes, ORs and relative risk rates in terms and formats that are easier to understand by people with varying levels of health literacy and numeracy.
The underlying task is how to simplify research results without misleading people, which is essential when communicating healthcare information. Selecting outcome probabilities and comparing data collected from different populations with various study designs would be misleading. Further, only providing relative instead of absolute risk estimates would lead to misinterpretation in most situations, framing that affects both professionals and patients.1 For example, imagine a person with a 4% 10-year risk of breast cancer recurrence: telling that person their risk of breast cancer coming back could be cut in half, or reduced by 50%, using treatment Y (relative risk) is misleading. The more informative approach would be telling them their risk could be reduced from 4% to 2% (absolute risk), a risk that is perhaps even clearer if portrayed in natural frequency (n in 100).
This article tackles the challenge of untangling complicated, heterogeneous evidence to deliver comparative information to non-scientists while at the same time avoiding misrepresenting the underlying data. Our focus here is to describe some of the editorial challenges of presenting complex scientific information, particularly when published as continuous outcomes in ways that are accurate yet accessible to a broad public, irrespective of their health literacy and numeracy levels. Those editorial challenges have materialised when developing tools to promote shared decision-making. The …
Footnotes
X @glynelwyn, @ThomasAgoritsas
Correction notice Since this paper first published, the author affiliations for MM have been updated.
Contributors GE initiated the article. MM, TA and MAD contributed to the final version, which has been approved for publication by all authors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests MM is employed by EBSCO Clinical Decisions, which makes various decision support products for both clinicians and patients. MAD and GE are advisers for EBSCO and developed Option Grid patient decision aids, now licensed to EBSCO Clinical Decisions. They receive consulting income from EBSCO Health, and royalties. TA is Chair of the board of the MAGIC Evidence Ecosystem Foundation, a not-for-profit organisation that provides authoring and publication software (MAGICapp) for evidence summaries, guidelines and decision aids.
Provenance and peer review Not commissioned; externally peer reviewed.