Thursday, 18 February 2010

Social Media and Medicine: Let's get #hyperlocal!

So after 16 months of blogging, twittering and social bookmarking how do I find social media to be useful to my work as a GP? Ummm. Well, I don't really....yet. Let me explain.

Early 2009: Phil Bauman's post on 140 Healthcare Uses for Twitter caused a lot of excitement. Then there was talk about surgeons twittering from operating theatres. It all left me a little bit cold. I couldn't see how I would use social media to support my work as a GP. But then I came across a review by Richard Smith from the BMJ in 1996, "What Clinical Information do Doctors Need" which helped me understand my resistance and the potential. To summarise there are three kinds of informations that doctors need
  1. Information about the patient
  2. Information about disease and management
  3. Information about local services to help manage the patient's condition.
I could (and might) do a whole other blog post about how social media might be useful in direct interactions with patients. In medicine we are trained almost exclusively for synchronous communication with patients. When I am in the same room as a patient we can usually communicate so much better. Speaking on the telephone is the next best thing and has the added bonus that it is much more convenient for the patient.  But as I say this needs a lot more room to explore so I'm not going to focus on this now, but safe to say, I am unclear as to see social media could help me gain information about patients at the moment.

Next, there is information about diseases and management. I'm a generalist, and I work in the UK and in a practice with several other doctors. For many conditions such as diabetes or cardiovascular disease, my management is determined by national guidelines, which are then tailored to individual patients through shared decision making. I will often look for information about these more common and less common conditions, and I will use sites such as TripDatabase which will find me relevant research papers and guidelines quickly. But I don't bookmark the information I find there. Why would I? The next time I need some information about that condition, which may not be for a few months, I will search again because the evidence may have changed. Now, this is different to how I use social bookmarking for my work as an educator or as a student (I'm registered for an EdD). Social bookmarking is useful for infomation that I have come across through serendipity (perhaps through Twitter), or because I am specifically looking for information that I don't have time to completely study now but want to be able to find again. But my infomation needs in clinical medicine tend to be more just-in-time. I don't think that I am so alone in this and it probably explains why I have found it hard to find delicious doctors
(Should I be reading generalist journals to keep up to date? The RSS feeds of the BMJ, NEJM and the Lancet swamped my google reader and made me feel inadequate! But here is somewhere where social media is useful. If you have any interest in what is happening in the big medical journals sign up to the RSS feed from Richard Lehman's Journal Watch blog. He writes with wit, and cuts to the chase.)

And lastly there is how social media could be useful with regards to information about healthcare services locally. This is where I currently feel the largest gap, and not uncoincidentally, where I feel social media could contribute most. Back in 1996, Richard Smith pointed out that this information is often diffuse and rapidly changing. Doesn't that sound like something that social media could grapple with? Well, next month I am going up to Glasgow to meet people interested in gathering information that is diffuse and rapidly changing and that could help people living with long-term conditions. 

I'll tell you more about the ALISS project and what they are hoping to achieve when I get back. Then we can start thinking about how we can all get useful #hyperlocal information. What do you think?

PS. By last summer @drves was referring to me as a "Web 2.0 sceptic". But I was so enthused about this last year that I actually started a new blog to try and get some discussion going.


  1. Web 2.0 skeptic to Web 2.0 maniac.

  2. Great post! I do think you're right - doctors and other healthcare professionals are trained synchronously (hmmm - maybe all humans are). And not only that, but one-patient-at-a-time is about the maximum anyone could cope with. So web 2.0, with all that aggregation and 'continuous partial attention', doesn't seem a good fit at all.

    I wonder whether the notion of the 'backchannel' - as embodied in Twitter for example - might prove to be a more useful output of web 2.0? In some ways that's what Patient Opinion can provide (disclosure: I'm on the PO team) - you can hear (some of) the things the patient won't say to your face. Sometimes that's a real learning opportunity, and one we're interested in exploring.
    I'll be at the ALISS day - let's talk!

  3. It would be great if I could Tweet my symptoms to my GP (by DM of course) and get online diagnosis/reassurance or advice. Today though I think only "hands on" would do.

  4. @dundeechest I'm still a sceptic- hence trying to focus on what I can see the needs are, rather than trying to find problems to fit the technology.

    @JamesM Well, I will definitely try and catch you in Glasgow. I suppose this post is about the information needed for an interaction with a single patient rather than feedback on a service as a whole. I've seen some links to feedback on NHSChoices which would certainly be useful to GP pratices, and it looks as if they are going to act on it. I'm not so sure that backchannel would give information on how a patient is managing with illness or treatment... yet. But in the future perhaps.

    @Jane Thanks for posting your tweet here. Personally as a doctor I would not like patients tweeting me their symptoms. When I am in the practice I triage requests for telephone consultation as I am going through a surgery- and before and after. Sometimes I can look at the information the receptionist has taken down and know instantly that it would be better to see the patient face to face so they are added on to the end of the clinic. Sometimes I think I need to speak to that patient now, so I'll phone them but that sets things back a bit. Sometimes I think I'll speak to them when I get a chance. It's really hard for me to think about what information could be got over in 140 characters that would be useful. Could you make up some symptoms so that we could try it out? Post them here!

  5. ok here goes. In last 7 days I have pain when I press near the middle joint on middle finger L hand. Can feel a small lump under the skin.

    (139 characters!!)

  6. Hello Jane. What is your main concern about your finger? Swelling or pain?

  7. both. I worry the swelling may be arthiritis as my great aunt & great grandma had it. The pain is a concern as I need finger for dancing!

  8. Jane, I think it's best if I see your finger, then we can work out what to do.

  9. The problem with only accessing latest research, of course, is that it isn't necessarily accurate - and the problem with national guidelines is that they're based on older research by definition, potentially giving you dissonance between the guidelines and the research.

    What might be more useful - and doable via social media - is a kind of uber-meta review; you post the annual results of patients in your practice with Disease A, as does every other GP with an interest in the experiment. You can see how your results compare with anyone else; if you're an outlier - either better or worse - there's some learning opportunities, either from you or to you.

  10. @Nicky
    I'd love to know more about your background and interest in this. I can't tell so much from your blogger profile.
    Your suggestion is very interesting and I've heard a few other people talk about this kind of crowd-sourcing. @laikas , a medical librarian, did a great series of posts about evidence-based medicine (EBM) and web 2.0. Here is the first:
    EBM does hold meta-analysis at the highest level and this depends on aggregating research. A single paper should usually not change practice. With regards to diabetes, questions are started to be raised about how low HbA1c should be in those with Type2 Diabetes. How low can you go? When the ACCORD trial was published showing that too low might not be good, guidelines were not changed. But now that other papers are saying something similar I'd expect a major review of guidelines. We will also discuss this research in my practice and decided if we will change our protocols in advance of guideline changes. That's why I say that a single paper rarely changes practice.
    You mention doctors discussing how well they perform in the management of different diseases. Well, that information is also available online.
    The link shows how well every practice did in Tower Hamlets at meeting one of the GP contract info targets on HbA1c.
    I completely agree with you that learning how and why some practices do better than others is a real opportunity for social media. We could all be sharing how we manage our diabetes clinics- what works, what doesn't work- and how we use local services, and patients could contribute on how they have found exercise referral programmes for instance. That's one aspect of the way I think #hyperlocal is the future.
    Many thanks for your comment!

  11. Comments on this post on doc2doc

  12. In terms of access to local health services, there are a number of projects that use interactive mapping to get information to the general public about health and social services in their area.

    Check out:

    Mapping HIV/AIDS and TB services in Kenya:


I am reintroducing word verification to cut back on spam posts. I'm sorry if you find it frustrating,