Glyn Elwyn: Expecting too much of “patient decision aids”

Too many people hoped that by introducing tools called “patient decision aids” we would quickly bring about shared decision making in clinical encounters. As if the tools, by themselves, could generate collaboration based on respect for the informed preferences of those involved.

In my view, those people were mistaken. “Patient decision aids” is not a great name for a starter. These tools are not about “decisions.” Nor do most patients feel comfortable “making decisions”not solely on their own at least. Decisions in healthcare are difficult and most people will look for support and guidance. These tools essentially provide information about options, ideally in a balanced accessible way, based on trustworthy up to date evidence. Their goal is to get people ready to discuss options, reach preliminary views, and foster joint discussion and deliberation with health professionals (Glyn Elwyn et al. 2017)

Are “decision aids” solutions in search of a problem?

Patient decision aids have been around since the late 1980s. Over 100 randomized trials have shown that the tools lead to better informed patients, who in turn make different decisions as a result (Stacey et al. 2017). A recent review of tools used solely in clinical encounters reported the same outcomes, with little or no effect on visit duration (Scalia et al. 2019). However, there are very few examples of adoption in routine clinical settings. All of which suggests that these tools are not viewed as essential elements of standard practice by the majority of clinicians or by their organizations.

To expect healthcare professionals to provide patients with patient decision aids may be an example of wishful thinking. Health professionals face competing priorities and, moreover, are increasingly constrained by electronic record straight jackets that make their work more difficult (Pieterse, Stiggelbout, and Montori 2019). Pausing to access and use tools, whose quality are uncertain, explaining the existence of reasonable options, comparing trade-offs, and diagnosing individual preferences represents new extra work for most clinicians. Currently, neither workflow design nor the electronic health record support these tasks (G. Elwyn, Cochran, and Pignone 2017). If we add clinicians’ common concerns that these tools take up more time, cause patients to doubt their expertise, and in some situations, may reduce their income when better informed patients elect to decline or defer procedures, it’s no surprise that implementation efforts have failed (Glyn Elwyn et al. 2013).

Send just-in-time information tools

Yet, despite the unfortunate label, when people are given well-designed “patient decision aids” they are almost universally enthusiastic. People facing difficult healthcare choices embrace tools that help them compare reasonable alternatives, and, in the US, consistently ask for more information about the cost of different options, as insurance coverage increases direct financial burdens. If these tools are given to people at relevant points along the patient journey, and in ways that are easy to access, I predict enthusiastic uptake. For a woman invited to have a screening mammogram, she needs to know whether or not it is in her long term interest to have such a test. For a man diagnosed with early prostate cancer, he and his family are keenly interested to know which treatment option is likely to be the most effective and have the least harm. There will be even more uptake if the tools viewed have been endorsed by their clinician, and are considered integral to a normal process of care.

Make easy-to-use tools using UX 

Academics have typically designed these tools, with limited resources or attention to the principles of user-centered design (UX). Very few are brief, visual, readable, or use animation or video: the kind of information formats and media that people find useful. A handful of commercial companies, mostly in the US, also produce these tools, but they also often fall into the trap of making them too long or too complex. Whether or not integration into electronic healthcare records will lead to widespread clinician adoption remains unknown (Coylewright et al. 2019).

Will clinicians be receptive to better informed patients?

There is strong consistent evidence that patients who use these tools gain knowledge (Stacey et al. 2017). Of course, not everyone is willing to assert their new knowledge when meeting clinicians, and not all clinicians will be receptive to better informed patients (Frosch et al. 2012). Yet, as more people search for information it is neither sensible or logical to dismiss the need to coproduce patient centred care (Batalden 2018). Some clinicians also acknowledge that well-informed patients actually make their work easier, saves time, and helps them raise their game.

We will need structural fixes 

Tools to aid decision making need to be provided to the right patients at the right time. Systems need to change so that the patients are identified ahead of their clinical visits and sent well-designed materials to help them prepare for visits. Policy mandates are also required: it is too easy to call any leaflet a “patient decision aid.” Patient information tools need to be certified so that they as free as possible of bias that might arise from commercial interests, and based on the very best evidence synthesis. Washington is the only state in the U.S. that is providing certification for these tools. Norway and Taiwan have developed systems to certify tools, and the UK has considered a certification process. Finally, thought should be given as to how to motivate clinicians to accept the use of these tools in workflows and to refine their ability to welcome patients who have used them to compare options treatments (Ryan et al. 2019).

Only when these structural changes occur will we see some movement towards shared decision making. In the meantime, we are a bit stuck when it comes to the use of these tools. Nevertheless, I do not think people will put up with the status quo for much longer. 

That’s my optimism kicking in.

Glyn Elwyn, professor, The Dartmouth Institute for Health Policy and Clinical Practice

Competing interests: Glyn Elwyn is editor-in-chief of the Option Grid tools produced by EBSCO Health, adviser to PatientWisdom and is the chief clinical research scientist for Abridge AI Inc, a developer of an app that empowers patients to securely record, transcribe, and share health recordings.


Batalden, Paul. 2018. “Getting More Health from Healthcare: Quality Improvement Must Acknowledge Patient Coproduction-an Essay by Paul Batalden.” BMJ  362 (September): k3617.

Coylewright, Megan, Jon G. Keevil, Keren Xu, Shayne E. Dodge, Dominick Frosch, and Michael E. Field. 2019. “Pragmatic Study of Clinician Use of a Personalized Patient Decision Aid Integrated into the Electronic Health Record: An 8-Year Experience.” Telemedicine Journal and E-Health: The Official Journal of the American Telemedicine Association, August.

Elwyn, G., N. Cochran, and M. Pignone. 2017. “Shared Decision Making-The Importance of Diagnosing Preferences.” JAMA Internal Medicine 177 (9).

Elwyn, Glyn, Marie Anne Durand, Julia Song, Johanna Aarts, Paul J. Barr, Zackary Berger, Nan Cochran, et al. 2017. “A Three-Talk Model for Shared Decision Making: Multistage Consultation Process.” BMJ  359 (November): j4891.

Elwyn, Glyn, Isabelle Scholl, Caroline Tietbohl, Mala Mann, Adrian G. K. Edwards, Catharine Clay, France Légaré, et al. 2013. “‘Many Miles to Go …’: A Systematic Review of the Implementation of Patient Decision Support Interventions into Routine Clinical Practice.” BMC Medical Informatics and Decision Making 13 (Suppl 2): S14.

Frosch, Dominick L. D. L., Suepattra G. S. G. May, K. A. S. Katharine A. S. Rendle, Caroline Tietbohl, and Glyn Elwyn. 2012. “Authoritarian Physicians and Patients’ Fear of Being Labeled ‘Difficult’ among Key Obstacles to Shared Decision Making.” Health Affairs 31 (5): 1030–38.

Pieterse, Arwen H., Anne M. Stiggelbout, and Victor M. Montori. 2019. “Shared Decision Making and the Importance of Time.” JAMA: The Journal of the American Medical Association 322 (1): 25–26.

Ryan, Padhraig, Saturnino Luz, Pierre Albert, Carl Vogel, Charles Normand, and Glyn Elwyn. 2019. “Using Artificial Intelligence to Assess Clinicians’ Communication Skills.” BMJ  364 (January): l161.

Scalia, Peter, Marie-Anne Durand, Julia L. Berkowitz, Nithya P. Ramesh, Marjan J. Faber, Jan A. M. Kremer, and Glyn Elwyn. 2019. “The Impact and Utility of Encounter Patient Decision Aids: Systematic Review, Meta-Analysis and Narrative Synthesis.” Patient Education and Counseling 102 (5): 817–41.

Stacey, Dawn, France Légaré, Krystina Lewis, Michael J. Barry, Carol L. Bennett, Karen B. Eden, Margaret Holmes-Rovner, et al. 2017. “Decision Aids for People Facing Health Treatment or Screening Decisions.” Cochrane Database of Systematic Reviews  4 (April): CD001431.