Thank you Mohammed for raising the issue. 

For me, it isn’t just that both clinicians have similar problems to patients in this area. They may share difficulties in perceptions about risks and benefits but it is likely that they are different - clinicians one from another and from one patient to another. I regularly ask groups of clinicians to estimate stroke risks in 1000 people with atrial fibrillation over the next year and estimates vary in primary care clinicians between 1 and 100 (when the best estimate for a patient with the characteristics of the case we have is 84). In secondary care clinicians I’ve had estimates of 200-300 for the same case. 

But even if people used the decision aids which describe the average risk for that population, we have to accept that it is inherently difficult to apply them to individuals with any hope of technical accuracy. For elements of the risk and benefit calculation rely on outputs from RCTs which almost always have a highly selected population without much in the way of co-morbidities. The reality is that co-morbidities is the norm, and there is a necessary extrapolation for an individual patient. 

Then there is the knowledge asymmetry / power imbalance between patient and clinician - even though both often work with an incomplete knowledge, perception and understanding of the risks and benefits.   

And then there are the differences in values and preferences between individuals, both clinicians and patients. In stroke prevention in atrial fibrillation one individual may greatly fear a stroke or a major bleed, another - rightly or wrongly - hardly has any fear of one or other of the major benefits or risks. People are very different one from another. 

Gradually all of this has become much more explicit over the last five years - at least for me and I recognise some others were ahead of my own recognition. It does seem to me that the recognition of the complexity is spreading much wider, and recognising the problem is the first step to trying to improve matters. But what to do and when is not clear. 

Technical knowledge dominates the curriculum for undergraduates and for postgrads in training and actually for lifelong learning. And yet clinical decision making with and for patients has to be about having really skilled conversations, largely about values and preferences. In the face of the complexity a technical solution will often not indicate what to do with any certainty, not least because we often cannot predict what will happen for this individual. 

The original challenge was to incorporate evidence into the decision making for individuals. It looks like we have instead collected together the information, but not answered at all the original challenge of how to integrate it. I caught up with Kieran Sweeney’s descried of his own mesothelioma this week http://www.bmj.com/content/339/bmj.b2862.long and I suspect may will be shocked at the consultation skills. Even when the technical care is exemplary, the conversation skills are as important - perhaps more important, though of course there is never an excuse for poor technical aspects of care. 

We need a link between the consultation skills,  the EBM and the risk communication arenas. Reading http://www.bmj.com/content/339/bmj.b2862.long and recognising that there was a need for an “hello my name is” campaign means I'm not sure the last 20 years of having communication skills on the undergraduate curriculum have achieved all we and the patients would want…..and its in that direction that I think we as EBM fans and experts will need to turn to begin to answer Mohammed’s very important question. 

Neal Maskrey

[log in to unmask]

Visiting Professor of Evidence-informed decision making, Keele University

Co-Lead ADVOCATE Field Studies, University of Amsterdam; a Horizon 2020 European Oral Health Project  www.advocateoralhealth.com

Mobile: 07976276919

On 19 Jan 2017, at 11:42, Mohammad Zakaria Pezeshki <[log in to unmask]> wrote:

Dear List

    Many thanks to Juan for sharing this eye-opening systematic review. It is interesting that the authors' 2015 systematic review, assessing the same research question among patients, found similar problem ( https://www.ncbi.nlm.nih.gov/pubmed/25531451) . The authors conclude that "Clinicians should discuss accurate and balanced information about intervention benefits and harms with patients, providing the opportunity to develop realistic expectations and make informed decisions.". The authors' current systematic review among clinicians tells us that clinicians themselves underestimates the harm and over estimate the benefits. My question: What should be done for solving this :

BOTH patients and clinicians have SIMILAR problem: Overestimating of benefits & Underestimating of the harms. How clinicians can solve the problem of patients while they have same problem?  What are you suggestions?

I am looking forward to receive your ideas. I hope this list will have a good discussion about these two systematic reviews.

Best regards, Mohammad

Mohammad Zakaria Pezeshki, M.D.
Associate Professor
Department of Community Medicine,
Tabriz Medical School, Golgasht Avenue, Tabriz, Iran,
Tel:  ++ 98 413  336 46 73
Fax: ++ 98 413 336 46 68

On Wed, Jan 18, 2017 at 1:49 PM, Juan Gérvas <[log in to unmask]> wrote:

Los médicos clínicos suelen sobre-esperar beneficios y sobre-ignorar daños de tratamientos/cribados/pruebas.
Clinicians’ Expectations of the Benefits and Harms of Treatments, Screening, and Tests. Clinicians more often underestimated rather than overestimated harms and overestimated rather than underestimated benefits.
http://jamanetwork.com/journals/jamainternalmedicine/article-abstract/2596010

-un saludo juan gérvas