Pioneering partnerships to deliver evidence-based patient decision aids

The ultimate goal of evidence-based medicine is to get accurate, evidence-based information about the benefits and harms of treatment options to policymakers, practitioners and patients. Unfortunately, the evidence is rarely getting to the patient. We aim to help change that.

Patient decision aids have not been based on systematic and up-to-date syntheses of critically appraised evidence
Shared decision-making is a process whereby the doctor and patient work together to make a treatment decision, having discussed the options and their benefits and harms, and considered the patient's values, preferences and circumstances. Patient decision aids (PDAs) are evidence-based tools designed to help patients make well-informed and deliberate choices among healthcare options. They are designed to complement, rather than replace, clinicians' counselling about the most appropriate course of action. While the movement toward PDAs should be applauded, the concern now is that the vast majority of PDAs are not based on a rigorous or systematic summarisation of benefits and harms of the various treatment options. This has been documented most comprehensively by Hoffmann and colleagues for the International Patient Decision Aids Standards (IPDAS) Collaboration, which found that 87% of all publicly available PDAs did not report any steps in the evidence summarisation process. Hoffmann’s article made proposals to further emphasise the evidence summarisation process aspect of PDA preparation, and these are currently being reviewed by IPDAS. The previous focus of IPDAS has been on design and presentation aspects of PDAs; for example, whether patients can better understand natural frequencies versus percentages, but increasingly it is being recognised that the accuracy of the underlying evidence is at least as important. 

Can existing systematic reviews and evidence-based clinical practice guidelines provide the evidence to support evidence-based PDAs?
For some conditions, the answer is ‘yes’. However, for many conditions or patient questions, the answer is not so simple. For many patient questions, there may be no systematic reviews or the available reviews may not be current, robust or adequately address the relevant patient question. Reviews of the evidence and conclusions from guidelines may also be influenced by conflicts of interest. Often an independent review of the evidence is needed. In the recent initiative to establish evidence-based PDAs in a project for a hospital in Germany, it was found that 68% of the PDAs required a de novo systematic review (and another 15% required a review update). Synthesizing evidence on the effects of treatment options for questions most important to patients could be a major undertaking. How can it be supported and efficiently advanced?

Cochrane, EQUATOR Network and evidence summarisation approaches for PDAs
One idea is to draw on the knowledge of systematic reviews and evidence synthesis conduct and reporting from organisations that have advanced this work for decades, such as Cochrane and EQUATOR Network. But where to start? How can we translate the existing standards for the conduct of evidence syntheses for PDAs? We identified four approaches that we presented as four posters at the 11th International Shared Decision Making (ISDM) conference held June 2022 at Kolding, Denmark.

First, we asked how the process can be accelerated given that traditional systematic reviews are time consuming and expensive? For guidance on this, we looked to methods used for rapid reviews, a form of knowledge synthesis that speeds the process of conducting a traditional systematic review by streamlining or omitting a variety of methods to produce evidence in a resource-efficient manner. We focused on the Cochrane Rapid Reviews approach and, although this approach offer some shortcuts compared to traditional Cochrane reviews and thus may save time and resources, it is still resource intensive. Just like traditional Cochrane reviews, Cochrane rapid reviews require that a systematic and reproducible approach be used to search for, select and critically appraise the best evidence available for the research question.  

Second, we asked when to use an existing systematic review and when to conduct a new one? One consideration is its currency. Have new trials been published that would make the systematic review’s findings obsolete? For guidance on this, we looked to a method developed by the US Agency for Healthcare Research and Quality (AHRQ) that includes criteria and definitions of qualitative and quantitative ‘signals’ from new evidence for the need to update systematic reviews. For example, one criterion for a signal of a ‘potentially invaliding change in evidence’ would be if a high-quality new trial suggests ‘opposing findings’ to those in the original review (with such definitions/criteria operationally defined in detail in the guidance). Using this AHRQ method as a tool for summarising evidence for PDAs may take some of the guesswork out of deciding whether it is necessary to incorporate new trials into an existing systematic review.

Third, when conducting a systematic review and summary for a PDA, how can the process be reported? The EQUATOR Network is an international initiative to support the use of study reporting guidelines to improve the value and transparency of health studies. The EQUATOR reporting guideline for systematic reviews of interventions has recently been updated (PRISMA 2020). Can it be applied to the development of PDAs? The key challenge that limits the applicability of PRISMA 2020 for reporting evidence summarisation processes for PDAs is that it was developed for systematic reviews of interventions with randomised trials as the unit of analysis, whereas systematic reviews may be used in the evidence summarisation for PDAs. However, two EQUATOR reporting guidelines that include systematic reviews as a data source are under development (overviews of reviews and rapid reviews). Aligning PDA evidence summarisation methods with this guidance, once available, may promote harmonisation of methods across developers. Such harmonisation of approaches may support transferability of the summarised evidence for different purposes (PDAs, guidelines, etc.) and potentially reduce duplication of effort.

Fourth, how can we get an accurate assessment of the harms of an intervention? Accurate harms assessment is particularly important for PDAs because patients weigh benefits vs. harms in deciding on a treatment option. If the benefits are well-reported and quantified with reliable estimates, for example, from robust randomised trials and systematic reviews, but the harms are selectively reported or completely unreported, a summary of reported benefits and harms may give patients an overly optimistic view of the effects of the treatment. How do PDAs address the widespread under-reporting of harms? Do ‘no harms’ reported really mean there are no harms of the intervention or were they just not observed or investigated? For this, we illustrated the ORBIT II method, which has been developed and refined over the past decade (and has a dedicated website with training) to assess risk of selective reporting of harms. Providing patients with robust information about harms – with uncertainty due to risk of incomplete/selective outcome reporting acknowledged – respects informed patient choice and patient autonomy, which are governing principles of medical ethics.

Impact and lessons learned
In addition to our posters, we presented a pre-conference workshop on evidence-based PDAs at ISDM 2022. Our workshop consisted of two parts. Part I focused on the topics of the posters as well as a worked example demonstrating the challenges of an evidence summarisation process for a specific PDA. Part II focused on insights, challenges and results from developing PDAs, with emphasis given to practical insights and cases where robust evidence is lacking.
 

The impact thus far of this exploratory work has been raising visibility on the evidence summarisation approaches for PDAs at the ISDM 2022. We received many positive comments from conference participants that evidence-based PDAs were necessary and timely, and that the optimal approach was to align the development of PDAs with international standards for evidence summarisation. One practical consideration is the need for funding to support this work and these partnerships. Unfortunately, methodological research is often not recognised as important by funding agencies, and building capacity to do methodological research to support evidence-based PDAs does not align with relevant requests for applications. 

Conclusions
For shared decision-making to reach its full potential, the information presented to the patient in PDAs must provide an evidence-based and accurate account of the benefits and harms of treatment options. This should be a key consideration for PDA standards and credentialing as well. Summarising the evidence needed to support the development of a critical mass of evidence-based PDAs is a large-scale undertaking. A first step is to learn from relevant evidence-based methods already developed by leading international organisations, such as Cochrane and EQUATOR. These organisations have led evidence-based clinical decision-making for decades, and the shared decision-making community can benefit by building partnerships with these EBHC organisations and adopting/adapting their methodology. The investment will pay huge dividends in trustworthy, reliable, evidence-based PDAs, which will lead to the best health outcomes for patients through the shared decision-making process.

Conflicts of interest: None

Acknowledgments
We would like to acknowledge our partners and collaborators on the posters and pre-conference workshop. 

We would like to thank our poster collaborators, Yoon Loke, Sunita Vohra, Liliane Zorzela, Chantelle Garritty, Candyce Hamel, and Margaret Sampson, who are senior EBM methodologists. Dr. Loke and Dr. Vohra lead the Cochrane Adverse Effects Methods Group, and Dr. Zorzela led the development of PRISMA Harms in collaboration with Drs. Vohra and Loke. Dr. Garritty and Dr. Hamel are leaders of the Cochrane Rapid Reviews Methods Group. Dr. Sampson managed the Children's Hospital of Eastern Ontario medical and family resource libraries.

We would also like to thank our workshop collaborators, Fülöp Scheibler, Marion Danner, Anne Rummer, and Marie Debrouwere, who led the Share to Care program at University Medical Center Schleswig-Holstein at Kiel, Germany. The Share to Care program was a large-scale implementation of evidence-based PDAs for shared decision-making. Our workshop was introduced by Tammy Hoffman, who is the Professor of Clinical Epidemiology and leads the Centre for Evidence-Informed Health Decisions in the Institute of Evidence-Based Healthcare, Bond University.

Authors
Eric Manheimer, Susan Moss

Disclaimer
The views expressed in this World EBHC Day Blog, as well as any errors or omissions, are the sole responsibility of the author and do not represent the views of the World EBHC Day Steering Committee, Official Partners or Sponsors; nor does it imply endorsement by the aforementioned parties.