I’m confused about the potential utility of social media in health, and I suspect I’m not alone. I’ve been struggling for a term to help me articulate my feelings towards the use of the terms “social media” in the context of “health” and “healthcare” for a while. Struggling to the point where I’d avoid conversations for fear of the inevitable twitch in my left eye as I recognise, yet again, that I simply can’t compartmentalise the concepts as well as I’d like.  This is despite malignant curiosity leading me to use most major new communication and technology trends around since Boyz II Men groupies were still teenagers.

Following almost every discussion around the use of social media in health (and healthcare) I’m left with a recurring feeling that people are taking part in the one same conversation, using identical words, but talking about different concepts.  I’m sitting here at the kitchen table after a weekend of anaesthetising unwell patients, trying to crystallise in my own mind what these different concepts are, and how they relate to these unwell patients, their well relatives, and the staff caring for them.

I’ve recently understood that “social media”, “health”, and “healthcare” are best described as a “suitcase words”.  Artificial intelligence researcher Marvin Minksy described “suitcase words” as words containing many different concepts. These evolve to improve the efficiency of daily conversations, but can be singularly unhelpful when trying to match these jumbled concepts to real actions and outcomes. I feel it’s the difficulty in matching of words and concepts during these conversations to real improvements in outcomes that hopeful, but confused, clinicians are struggling with.

Minsky talks about unpacking these suitcase words into the smaller, more actionable concepts.

Lets start with the term “social media”.  These can be roughly unpacked into public networks based on common interests (twitter.com), private networks based on social and organisational relationships (Facebook.com, yammer.com), or mixed public/private networks (google+).  Social media in its essence is networked multi-directional content. Email and chat channels, widely available since the early 90′s fit this description, but various factors have meant that the network effect fuelling the uptake of newer communication tools never developed to the same extent.  Content creation and distribution has since become more efficient with advancements in technology and evolution of culture, and now almost anyone can create content via their affordable devices and data plans with a unique human behaviour that has been “shaped” to share.

What about the word “Health”?  Are we talking about wellness and its tremendously broad determinants, or are we discussing the management of illness (healthcare).

What about “Healthcare”? Lets unpack this to slightly more focused suitcase terms of quality of care, productivity, and branding (staff and patient recruitment).

At a recent Sax Institute forum entitled “Bringing the social media revolution to healthcare“, I sat in a mixed audience of administrators, clinical staff, journalists, private hospital body representatives, marketers, and various other stakeholders in healthcare.  We listened to Mayo Clinic‘s experience of exposing their already successful brand to the unpredictably stormy seas of social media.  I came away thinking that social media led to positive improvements in Mayo Clinic’s brand, as well as improvements in patient outcomes through distribution of information to patients for which Mayo Clinic had the expertise to manage.

I spoke to attendees whose primary interest was in organisation branding and its potential for staff and patient recruitment and who thought primarily about twitter, linkedin, and facebook.  I spoke to health promotion practitioners whose primary interest was in assessing sentiment and promoting behaviour shaping to improve population health through tools such as twitter, facebook, and youtube. The benefits flowing from the evolution of communication has been obvious to private healthcare services and health promotion researchers because their primary roles are to assess sentiment and shape behaviour as marketers, and that’s what the two-way mass communication platforms of twitter and facebook are particularly good for.

I spoke to administrators in healthcare organisations thinking about how to grapple with privacy, legal, and productivity risks. Why their staff would want access to youtube, twitter, SMS, facebook, what they are likely to overshare, and what social media policy documents need to be put in place.

Unwell people are beginning to think about tools to help them connect to people with shared experiences (patientslikeme, curetogether ), illness information produced by experts (mayoclinic, Wikipedia, quora), and the people and services that help them get well (ratemydoctor), as well as tools that improve communication with their clinicians (hellohealth, healthvault, teleconferencing).

Healthy people have picked up on the potential of tools that connect them with people and information that keep them healthy.  They think of twitter and blog streams dealing with nutrition, exercise, and wellbeing. They think of socially connected health metric applications that they hope will positively shape their behaviour (the eatery, runkeeper, dailymile, trackyourhappiness).

Clinicians’ interests at this point in time seem to be in the understanding of the implications of social media tools to them and their patients. They are, for good reasons, unable to provide specific clinical advice through public networks, and the vast majority of clinical staff have no access private organisational communication networks that may improve productivity within their organisation. The default position for people directly responsible for the health outcomes of others is one of well-deserved skepticism. They are unable to crystallise in their own mind which of the jumbled concepts in “social media” would help them do their job better.

Over the next few years communication tools will evolve and clinicians will be given access to communication networks with more appropriate privacy controls for the information being exchanged.  Discussions will also start to focus on narrower and more relevant concepts, and as this happens, the use cases for clinicians and the problems these connected technologies are able to solve should become clearer.

Note: I’d like to attribute the concept of “suitcase words” to @arcwhite and the clarification of behaviour “shaping” vs “change” to @yhpo and @bjfogg

The Milkmaid, Johannes Vermeer

I’m a fan of both longitudinal population studies and of the work of public health researcher Michael Marmot from the University College London. His previous work on Whitehall studies I and II revealed a correlation between a person’s health indicators, and their position within the British public sector. Much of his work has revolved around the notion of how personal autonomy affects a person’s ability to choose healthy behaviour.

Recently released analysis of the English Longitudinal Aging Study (ELAS) reveals some potentially interesting results around, amongst other things, measured biomarkers and an individual’s wealth.

The analysis of data from British citizens over the age of 50 shows what appears to be a statistically significant correlation between the serum level of dehydroepiandrosterone (DHEA) and their wealth quartiles. (see first table)

What exactly does this mean? At the moment it’s just a correlation between two variables, and its difficult to draw any conclusions. What exact role this steroid androgen precursor may play in the health outcomes is unclear. There is nothing to suggest that replacement of low normal DHEA levels with will improve outcomes, and if the correlation is robust, may simply be an indirect indicator of multiple health behaviours in people in the upper wealth quartiles. You’ll see if you examine the data (second table) on physical activity and wealth, wealthy people are significantly more likely to be physically active. DHEA levels might simply be an indirect reflection of this and other behavioural disparities.

The significance of this research is in its ability to improve the evidence base for policy development. If we understand the factors at that lead to poor health behaviour (that ends up costing both the individual and the tax payers), we’re in a better position to effect real change.

Interestingly, Edward Jenner picked up on one of the first wealth biomarkers in 1796. He was a general practitioner in England who observed that milkmaids were spared the telltale facial scars of smallpox that seemed to afflict the female aristocracy. He postulated that cowpox exposure amongst the milkmaids protected them from developing this disfiguring disease. He tested his theory by innoculating a semi-willing subject with cowpox, later exposed him to smallpox, and by documenting his resistance, developed the world’s first official smallpox vaccine.

You can get a copy of the ELAS analysis here. http://www.ifs.org.uk/elsa/publications.php?publication_id=5315

Posted via email from Pieter Peach

With social media gaining traction since SARS (2003) and Avian H5N1 (2006), it will be interesting to see what role it might now play as a media tool in the current pandemic du jour.  The benefits of social media are clear, namely speed of communication, and monitoring sentiment.  The cost in accuracy is not insignficant, and it will be simply a matter of how to best use it.  Can new micro media services like twitter add anything useful? Will good information float well enough above the bad to make it worthwhile keeping track of?

I don’t think the question is whether people should, its a question of how they should. Twitter is just an open conversation tool, and people will use it to talk about issues important to them, and if swineflu/H1N1 does take hold, it will become one of those issues.

Access and distribute reliable information. Thankfully, several streams of information are available from official government and international agencies. (WHO -website & twitter CDC – Website, Twitter, Email).  The higher the official signal to unofficial noise, the better.  You could argue that it is often slower and more deliberate than other sources, but they have significant cost/benefit analyses to make with each official release.  Partially uncertain information is occasionally communicated, but only after due consideration.

Focus on facts and confirmed cases. Real numbers are much less than that reported in the media. If somebody publishes something without a link to either official or reliable press (eg. AFP) sources and you’re still interested, try looking for a pattern of multiple first hand accounts rather than a chain of retweets. Specifically with the current H1N1 Swineflu outbreak, “suspected” cases can be very misleading.  Once an “area” (eg. often a city) has a single laboratory confirmed case, everybody who presents to the emergency department, or general practitioner, with at least two of 1) runny nose or nasal congestion, 2) sore throat, 3) cough, 4) fever or feverishness gets labelled as a “suspected case”. This is all in the CDC case definition here.  You can imagine the number of patients with otherwise innocuous colds that come through like this everyday, let alone when the population is on heightened alert.

The next bit of information of interest to most people will be a change in the global pandemic alert phase which can be found here. For Australians, the federal government has a website up at http://www.flupandemic.gov.au/ with some information on pandemic preparedness with links to each state health departments. Its not the best, but at least its something local for both clinicians and the public.

The best source of CONFIRMED US cases are to be found here http://www.cdc.gov/swineflu/

Below is an unofficial map from http://flutracker.rhizalabs.com/ of human cases of H1N1 infection.

Know of any good, reliable sources of information people might find useful?

What a Saturday.  I had the chance to sit in on some interesting sessions at the inaugral Trampoline Melbourne, an unconference organised by Pat Allen, Melina Chan, and Steve Hopkins.  It was held at Donkeywheel, Melbourne’s newest social change projects venue.  100 people from various disciplines came together to speak about their biggest ideas.  I heard some interesting talks on complexity theory, persuasion psychology, trust systems as alternative economies, biomimicry, advertising and mass-collaboration, and missed out on hearing some apparently interesting talks on Zen IT and permaculture, personal prototyping, amongst others.  The format of an unconference is simple.  People turn up and the session agenda evolves as people put their hand up to speak.

I had a chance to present on future health, and explored the way technolology and open, accessible, data will have significant impacts on health outcomes when applied to the social determinants of health, namely, education, information equity, income, empowerment, and looked briefly at the potential of rapid learning systems to improve clinical processes and aid in clinical decision support.

Slides from my talk are below (it probably doesn’t make much sense as slides were only visual cues, but links to interesting sites are on slide 99), and video of the presentation here.

There will be another Trampoline in the future, so keep your ears out.

Humour detour.  @provoost just pointed me towards a peek at a future of less labour intensive aged care.

“…..as the cleansing bubbling action kicked in, Toshiko Shibahara, 89, settled back to enjoy the wash and soak cycle of her nursing home’s new human washing machine.”

I think it’s pretty self explanatory.  I think the manufacturers may have taken inspiration from Barbarella