The first sentence of the General Medical Council guidance on ‘Decision making and Consent’ (which came into effect in November 2020) says: “Shared Decision Making and Consent are fundamental to good medical practice”1.
Here is the 2021 NICE guideline definition2:
“Shared decision making is a collaborative process that involves a person and their healthcare professional working together to reach a joint decision about care. It could be care the person needs straightaway or care in the future, for example, through advance care planning. It involves choosing tests and treatments based both on evidence and on the person's individual preferences, beliefs and values. It means making sure the person understands the risks, benefits and possible consequences of different options through discussion and information sharing. This joint process empowers people to make decisions about the care that is right for them at that time (with the options of choosing to have no treatment or not changing what they are currently doing always included).”
So, in other words … in every consultation, there are two experts in the room: the clinician and the patient. Shared decision making is a process which embraces and builds on this. And allows the scales of a consultation to be more evenly balanced between the participants.
It involves sharing of evidence-based information about the options, benefits and risks and working together to agree on the most appropriate tests, treatments, management, or support packages. It’s relevant in any healthcare situation in which there may be different choices. Choosing to do nothing is one of them.
And most clinicians and most patients agree that providing more information, along with a move away from a ‘doctor knows best’ culture does help people reach more informed decisions.
It’s true – most of us practice in this way already. But when patients are asked for their views, we might not be doing as well as we think we are! It may be helpful to know more about how best to structure the conversation, how to explain risk and which phrases may lead to misunderstanding. And crucially, ways to ensure consultations do not become prolonged as a result.
Although we try to involve people as much as possible in their decisions and explain as much as we can, it’s well documented that clinicians often interrupt patients within 15-20 seconds of hearing their story3. and analysis of consultations shows that only 5% of the time is spent by the clinician answering questions 4. Often not enough time is taken listening to and allowing the patient to talk about their ideas, concerns and expectations.
And one size does not fit all. Our recommendation might be the right thing in one sense but completely impractical or unwanted from the patient’s perspective. Usually our treatment should prolong life, decrease pain and improve quality of life. But for some patients it may do the opposite. It’s a strange fact that doctors often choose less treatment / intervention for themselves than they would suggest for their patients.5 Just because we can do something does not mean that we should, or that it is what the patient really wants.
Imagine, with a growing family, you needed a bigger house and you spoke to an estate agent. Doubtless they could come up with several options to address that issue, but quite rightly you would only consider those that fitted with work, the desire to be near family or other leisure pursuits. It’s much more than just having 3 bedrooms. Only you can know about these preferences, but they are very important, and, just like estate agents we need to try to be aware of people’s preferences to be of most help to them.
This approach addresses uncertainty and internal conflict which people may have, whilst at the same time builds trust, and lessens subsequent regret about wrong choices. By being mindful of preferences, we also reduce the built-in inequity of our systems, which tend to prefer a one size fits all approach. Not everyone can travel to different parts of our area for out-patient services or has access to digital means of communication.
Firstly, this works much better when people are prepared. If there’s any information that can be shared ahead of the conversation, this is most helpful and helps when people are trying to take in information and make decisions at the same time. Inviting someone to accompany the person can be helpful.
Sometimes it may be appropriate to send test results in advance and it’s also useful to encourage people to think about questions in advance, e.g. the BRAN mnemonic:
• What are the Benefits of this test or procedure
• Are there any Risks or side-effects
• Are there any Alternatives
• What would happen if I did Nothing?
People could also be invited to record the conversation on their own mobile device, to help with recall of the conversation.
There are two useful (and similar) models for structuring a shared decision-making conversation.
We are advocating the DECIDE approach. The other commonly used approach is the ‘3 talk model: Team Talk, Option talk, Decision talk’. They are closely linked.
Firstly, we need to consider how information is ‘framed’. ‘Gain framing’ is where the focus is on the advantages of intervention e.g. 95 out of 100 people having this procedure will have a good outcome. ‘Loss framing’ focuses on the disadvantages e.g. 5 in every 100 will have a complication.
It’s also easier to be consistent about the ‘denominator’ – ideally 100 – and to say e.g. 5 in every 100 rather than 1 in 20; and rather than 1 in 4, say 25 in every 100.
Another way to express information in a helpful way is to talk about the ‘number needed to treat’. That is the number of people who need to have the intervention in order to have one positive outcome. In other words, if the number needed to treat is 23, this means that 23 people will need to have the treatment in order to have one good outcome. Conversely, the ‘number needed to harm’ is the number of people undergoing the treatment before a single harm occurs. If the number needed to harm is 200, this means that 200 people will have the intervention before one negative outcome occurs.
Another very useful decision aid if the information is available is a ‘pictogram’ such as this one from NICE guidelines showing the effect of stopping smoking on preventing a cardiovascular event.
National Institute for Health and Clinical Excellence. Guideline development Group. Improving the experience of care for people using NHS services: summary of NICE guidance March 2012, BMJ (online) 344(mar16 1):d6422
Sometimes what we say can be misinterpreted, so click here to view language to avoid and possible alternatives.
One of the main concerns of clinicians is the potential for consultations to become longer, meaning that less people can be seen. And it’s true, consultations might take longer. But not much and not always. And there are ways to deal with this.
A Cochrane review6 looked across the medical scientific literature at effect of the use of decision aids on consultation length. They found 10 out of 105 studies on the subject, which looked at consultation length. Eight of the studies showed no effect on length of consultation and two studies showed an increase. Combining the data, the difference in consultation length overall was 2.6 minutes longer in the decision aid consultations.
Giving people information in advance of the clinic could also help. And potentially allows more time for considering the issues and asking the right questions.
But maybe we need to make more time. A more considered consultation may result in more people opting for a more conservative approach with less requiring subsequent investigation or procedures.
Fair point. Some people are wary of this, as they are used to the normal style where their clinician makes a recommendation and advice is generally followed. There is evidence to suggest that this view is more common among older patients but an increasing trend for younger people (with ready access to multiple sources of information) to prefer the more informed approach of Shared Decision Making7. Remember also that there can be a significant ‘power imbalance’ in these conversations and some people find it difficult to speak up even when they want to be more involved in decision making.
Health literacy, the ability of people to access, understand, weigh up and use health information, is important. Many vulnerable people have lower levels of health literacy, but actually stand to gain the most from the greater opportunity to consider more information about proposed treatment6.
Shared Decision Making is usually not possible in true emergency situations when a person is seriously ill or injured and requires immediate intervention to preserve life.
Where a person temporarily or permanently lacks capacity, they should be supported to be involved in decision making. This may not be possible with the person themselves, but we should make every effort to find out a person’s likely wishes from people close to them or their Health and Welfare Power of Attorney.
Health professionals are not required to provide care that is not indicated or appropriate or is not of overall benefit to the person. If someone is asking for an intervention that is not appropriate, try to explore their reasons in order to try to reach an agreement. We may sometimes need to ask for further advice or a second opinion.
The GMC cites Seven key principles.1 Read them here.
There is a very good on-line module available for staff on the TURAS learning site (Sign in and search "Realistic Conversations").
In NHSGGC (in conjunction with NES and EC4H) we are running on-line ‘Realistic Conversation’ communication training courses covering Shared Decision Making. Links to forthcoming courses can be found here.
1. Decision making and Consent: guidance on professional standards and ethics for doctors. The General Medical Council. November 2020. https://www.gmc-uk.org/-/media/documents/updated-decision-making-and-consent-guidance_pdf-84160128.pdf
2. Shared Decision Making. NICE guideline (NG197) June 2021 https://www.nice.org.uk/guidance/ng197/chapter/recommendations#shared-decision-making
3. The effect of physician behaviour on the collection of data. H B Beckman, R M Frankel. Ann Intern Med.1984 Nov;101(5):692-6.
4. Patient participation in the cancer consultation: Evaluation of a question prompt sheet. P. N. Butow, S. M. Dunn, M. H. N. Tattersall & Q. J. Jones. Annals of Oncology 5: 199-204, 1994.
5. Physicians Recommend Different Treatments for Patients Than They Would Choose for Themselves. P A Ubel, A M Angott, B J Zikmund-Fischer. Arch Intern Med. 2011: 171(7): Arch Intern Med. 2011 Apr 11; 171(7): 10.1001/archinternmed.2011.91.
6. Decision aids for people facing health treatment or screening decisions. Dawn Stacey et al. Cochrane Database Syst Rev. 2017 Apr; 2017(4): CD001431.Published online 2017 Apr 12. doi: 10.1002/14651858.CD001431.pub5
Alastair Ireland, Realistic Medicine Lead, NHSGGC
Claire Macaulay, Realistic Medicine Lead, NES