Permit me to propose a new eponymous law of social physics. It is a variation on Godwin’s law, which famously states that “As an online discussion grows longer, the probability of a comparison to the Nazis or Hitler approaches one.” Thus, O’Connor’s Specific Variation of Godwin’s Law states: “As an online discussion of social media monitoring grows longer, the probability of a reference to Orwell or Big Brother approaches one.”
I’ve been moved to propose this new law by the “Case of the Twitterining Kidney Patient”- not, alas, a new Sherlock story, but rather a fascinating case study of social media monitoring written by Colin Wren (with prologue by Alex Talbott), over at NHSSM, a splendid website dedicated to discussing the opportunities and challenges presented to the UK National Health Service by social media. You should really take a look at the full case study (which is a model of clarity), but in short, it is the story of how the St George’s Healthcare NHS Trust in London began monitoring one of their kidney patients’ Twitter feeds, resulting ultimately in the patient’s consultant physician actually contacting the patient by Twitter mentioning to the patient during a ‘real-life’ consultaton that he or she had read a tweet in which the patient seemed to have made a medical error about some aspect of their condition. the physician then corrected the error. The patient, though a bit startled by the contact, actually approved and, indeed, encouraged more such doctor-patient interaction.
It was the consultant in question, however, who provided the trigger instance of O’Connor’s Specific Variation on Godwin’s Law, reporting that they “felt like ‘Big Brother watching over the patients life’.
There ensued a fascinating discussion at the NHSSM board about the case, touching notably on the lack of useful guidelines or policy for using social media within the NHS. What really interested me was how the case study, and the attendant Big Brother concerns, might be used to illustrate some pressing ethical issues regarding social media monitoring within the healthcare sector. It largely boils down to this: is it ever acceptable for a hospital, or indeed any healthcare organization, to be monitoring patients’ social media content? The idea that your physician is constantly looking over your social media shoulder raises some unpleasant spectres. As a colleague in bioethics, Nathan Emmerich, put it in a tweet regarding the Case of the Twittering Kidney Patient: “For good or for bad (it’s) new surveillance medicine!”, referencing David Armstrong’s famous article in which Medicine’s regulatory gaze is expanded to all aspects of our lives, not just illness. If a specialist can tweet you about the condition for which they are treating you, what’s to stop another healthcare professional sending a disapproving tweet every time you tweet ‘time for another glass of wine!’ or ‘looking forward to chips for dinner’? It sounds dystopian, of course, but that’s Big Brother for you.
Behind the spectre of Big Brother, however, there lie three critical ethical issues which I think it might be useful to address, not just in the Case of the Twittering Kidney Patient, but in all instances of social media monitoring in the healthcare sector. Those ethical issues are: 1) privacy, 2) justice, and 3) trust.
Privacy: This is the big one, the issue that concerns most people when it comes to social media monitoring / surveillance in the healthcare sector. The anxiety is that monitoring / surveillance may violate a patient’s right to privacy, that is to say it may violate thier right to keep certain information about themselves to themselves. The problem is that social media does nothing if not complicate the idea of privacy. Let’s say this had happened without benefit of Silicon Valley and that, rather than monitoring a patient’s tweets, St George’s hospital had tapped the patient’s phone line. There would be little question that this would constitute a violation of privacy – especially if the patient’s physician then phoned them up to correct them about the content of the conversation they’d just been having. But Twitter, and social media in general, is not a phoneline. Most Twitter accounts – including that of the patient in question – are public. They can be read by anyone, in much the same way as an online newspaper can be.
There are two general responses to the un-private nature of social media. The first, which I call the responsibility argument, stresses that users (in this case patients) have a duty to be aware that their social media content is publicly accessible and that they should take steps to protect that content if they wish to maintain their privacy. The second, which I call the propriety argument, acknowledges the public nature of social media content, but questions the license this supposedly gives to others to listen in to, or comment upon, that content. Basically, just because you can listen, doesn’t mean you should. As GP Anne Marie Cunningham put it when discussing the case (paraphrasing Shirky), “Although I might be speaking to you in a public place, most people would think it socially unacceptable that I actually eavesdropped (on someone else talking in public).”
These schools of thought both set privacy up as a sort of oppositional relationship between patient and physician/surveyor. The responsibility argument focuses upon the patient, expecting him or her to defend their privacy against surveillance, whilst the propriety argument focuses upon the physician (or whoever is doing the social media monitoring), expecting them to respect the privacy of the patient, even though the content they have created is clearly in the public domain.Physicians need to know what their patients are doing – information is key to diagnosis and treatment. Patients do not want, always, to share everything with healthcare professionals whom they may view as judgemental. One could do an entire post on competing notions of privacy online (and I will!), but it’s worth noting for now that a more productive way of looking at online privacy may be to conceive of it not as a commodity which can be lost (through a patient’s irresponsibility on Facebook or similar) or violated (through online medical surveillance) but rather as a process in which patient and physician work together to ensure that the physician has enough information to work with, and that the patient is aware of the quasi-public nature settings such as Twitter.
Justice: Whether or not the Case of the Twittering Kidney Patient constituted a violation of privacy, it had one unquestionably beneficial outcome: the patient emerged with a more accurate knowledge of their condition. Twitter, basically, may have helped them deal with their illness in a new, more appropriate fashion. This is great, of course, but what of the patient who does not have a Twitter account which their physician can helpfully monitor in case of medical error? Indeed, what of the patient who lacks access to the internet at all? Internet usage in the UK runs at about 80%, which leaves still millions without access. This, then, is the justice issue presented to us by the Case of the Twittering Kidney Patient: the digital divide. Whilst the opportunities presented by social media use within the healthcare sector are fabulous, particularly when it comes to empowering patients to make better decisions about their care, we run the risk in our excitement of unjustly excluding from those opportunities those who do not have reliable internet access. People lack access for numerous reasons, but it is mostly a question of economics – those with less money have less access. Talking about access to Twitter as a question of social justice might seem wilfully perverse, but if the Twittering Kidney Patient tells us anything, it’s that access to social media can improve your health outcomes.
In some ways, the justice question is an inversion of the privacy question. It’s not so much that there is an ethical issue with being surveyed, but that there may be an ethical issue in not being surveyed, that is to say in being passed over for the benefits coming to everyone else. Surveillance technologies, when deployed by the state or other public bodies have historically been viewed as malign, but as Edward Higgs points out in The Information State in England, “much information gathering by the state is not for the purpose of keeping people under surveillance, but is to enable citizens to safeguard their own rights through the recognition that state registration provides.” (p.8). The same argument might be made from a point of justice regarding medical surveillance using Twitter – that it is to enable patients to safeguard their own health through the better information that the medical profession provides.
Trust: I’ll cite GP Anne Marie Cunningham from the NHSSM discussion of the Twittering Kidney Patient again here: “I think we need to tease out why a hospital would want to monitor social media for mention of their services.” There are three potential answers, which represent rising levels of mistrust between patient and healthcare provider: 1) that the hospital merely seeks to improve patient care, which I think we would all like to be true, 2) that the hospital is worried that patients are bad-mouthing the hospital online, which I think we can all acknowledge is concern of any brand, or 3) that the hospital does not trust their patients to tell them the truth when talking to physicians and thus considers it necessary to put them under social media surveillance by way of extracting the information necessary to care and treatment.
Surveillance, despite the potential justice-centered benefits, almost always carries with it the sense that the object of surveillance is not entirely trustworthy, that they must be observed if the truth about them is to become apparent. If social media monitoring within the healthcare system is not to undermine the trust that is at the heart of the physician-patient relationship, then that sense of mistrust must be avoided. The key to this, I think, may be for healthcare organisations to be a bit more ‘social’ in their social media – that is, not just to ‘monitor’ but to put out some content that can be ‘monitored’ by interested or concerned patients. In other words, make your healthcare organization known to the social media sphere in which it seeks information, spend some of your organisation’s social capital (the excellent healthcare information it has!) and that mistrust can perhaps be avoided.
So, returning to my original question: is it ethically acceptable for a healthcare organization to monitor patients’ social media? The answer, as I have tried to show, is a qualified yes, where the qualifications are a) a concern for patient privacy as a process rather than a violable commodity, b) an awareness of the social justice implications of surveillance, and c) that the monitoring is done in the open and honest spirit of the social web. No answer in medical ethics is ever final (believe me!) and doubtless you will have your own ideas, which I invite you to share in the comments below.
Very many thanks to Colin Wren and the St Georges NHS Trust for sharing this excellent case study with the world.
Amended 20/10/10 at 8.23EST.
Hi Dan, excellent blog post. Really well thought out and you provide some great arguments.
I’d like to correct one thing, the consultant in the case study never contacted the patient via Twitter, he read the tweets and when the patient next saw the consultant, the consultant told the patient about the tweets being read and corrected the patients diet.
As for the motive, the monitoring was set up for the first 2 options in your post, I set them up to see how we could help patients via Twitter (by sending links to patient information and the like) and also to try and protect the St George’s brand on the internet.
I’m glad that the case study has inspired you, and I certainly hope that people don’t start comparing me to Hitler
Hi Colin,
Thanks for commenting, and for the complements. It really is a great case study. Sorry about the mistake in detail; I’ve amended the post accordingly.
I guessed from your original post that options 1 and 2 were your primary motivations for setting up the monitoring – I think the problem occurs when patients, for whatever reason, come to believe that such projects occur for option 3) lack of trust in them by physicians. I wonder, have you any sense of how other patients at St George’s feel about your social media activities?
And if anyone compares you to Hitler, just tell them that Bernie Eccelstone thinks he was a politician who could get things done. Actually, don’t…
Hello Dan,
I’m very glad that you have across the #nhssm as this is an excellent contribution to the discussion. Your post clarifies several issues that I have clumsily raised in my comments on the other blog and in my own head. Here are some responses:
- With regards to personal responsibility for privacy, I hope that in the coming months we will see increasing discussion around health-related digital identity issues. And I, personally will be considering what is my role as a GP, in helping patients to be aware of their health-related digital footprint. It doesn’t apply in this case but I am very concerned about any organisation (NHS, voluntary or other) establishing spaces where patients have less privacy than they are used to in discussions with clinical staff, without making that reduction in privacy clear to them. I have blogged about some cases: http://wishfulthinkinginmedicaleducation.blogspot.com/2010/08/paying-for-privacy-patient.html and http://wishfulthinkinginmedicaleducation.blogspot.com/2010/08/my-thoughts-on-health-professionals-and.html
- The digital divide is also a concern to me so I have interest in your arguments around justice. I also think that service providers are obligated to consider how best to use scarce resources. Moving resources from traditional services to social media because it is fashionable, only, may do many a disservice. With regards to public health messages the use of social media is much more likely to have an economic advantage over traditional ways of working. But the jury doesn’t even have a case to consider with regards to individual, personalised healthcare- yet.
I have some other thoughts but that will do for now.
Many thanks again,
AM
Anne Marie, thanks so much for taking the time to comment – much appreciated. I’m glad to hear that you’ll be integrating social media awareness into your work as a GP. I venture to say that many of your colleagues, with good reason perhaps, might consider this to be too burdensome in their already full work lives. How do you think we might go about encouraging GPs and other practitioners to see this as a part of their everyday duties? Is it just a question of training and remuneration, or would it require a larger cultural shift? I’m looking forward to reading your own blogposts on the issue.
As to the digital divide, you are right to note the obligation to consider the distribution of scarce resources (especially in light of the CSR today! do we have a Navy now? I missed that bit). I agree with you; it’s vital that any migration to social media doesn’t happen merely as a cost-saving exercise (although that should be a consideration), but rather has at its heart a concern for better physician-patient communication.
Hello again Dan,
I didn’t say I would be integrating social media into my work- I think there needs to be much consideration of many factors first. No one is going to change their practice unless there is something to be gained for patients.
I don’t think that moving to social media use would necessarily be cost-saving at all. In fact I have concerns about the opposite. Face to face communication is probably the most efficient use of my time- my patients aren’t distributed around the world- so getting together locally still feels a pretty sensible approach. If you and I were chatting about this face to face- even by phone or Skype- we would avoid some of the misunderstandings that we are having now. But the benefit of having this text based conversation is that it is public. I wouldn’t bother emailing you back and forward about this. I’d just pick up the phone. The advantage of having this conversation on your blog is the fact that others can join in, or at least muse on what we are saying.
The next step is considering when those same affordances bring benefit to a health professional-patient relationship. With public health messages the utility is much more obvious. I am happy to be challenged about this but as yet I am only aware of one doctor, @endogoddess, who is using private channels in Facebook to communicate with her patients with Type 1 Diabetes who wouldn’t use email.
I think your question about remuneration reflects your understanding of this from a US perspective. It just isn’t an issue in the UK. But effective use of time is.
To be continued!
Anne Marie – I didn’t mean to imply that you’d be using social media in your work, I just meant that I was glad that you were thinking about how you might make you patients aware of their social media footprint. Sorry for the confusion.
I think you are absolutely right about the difference between public health messages and individual patient-physician communication when it comes to social media. There’s an obvious economy of scale at play with public health messages, especially if the message goes viral online. A little public health comms work can go a very long way with social media. Like you say though, with individuals it may not be time effective.
And I’m British! I just live and work in the States. I understand the difference between the two healthcare systems (both have their benefits, but all in all one does miss the NHS). When I mentioned remuneration I was referring to the way that social media is sometimes seen as something *extra* on top of all the other work people have to do. It’s a big culture shift to go from thinking about blogging and tweeting as an added extra at work to seeing them as integral tools.
You might be interested in the folks over at http://hellohealth.com – a practice in Brooklyn NY, which relies almost entirely on the web to interface with patients.
More, later!
We currently have not asked patients how they feel about St George’s social media activities, I’d love to one day get to a position where we could ask our patients if they know these social media services exist and how and what they would use them for. Unfortunately this probably won’t happen for a while due to people in high positions (and their peers) attitudes towards social media (Andrew Marr ‘s ‘inadequate, pimpled and single’ stance on Blogger and David Cameron’s ‘too many tweets might make a twat’ spring to mind). And the cost involved to launch such a survey.
Of course even if we were to ask patients their opinions and get their backing 100% to start talking to patients actively on Twitter I feel that this question of how ethical this interaction is will still pop up and people will still think that the we are just trying to make sure they are not trying to lie to us.
It’s a hard subject to push towards people as the potential to improve patient care is there but the confidentiality aspect is too much of a risk for it to actively happen.
You’re so very right about ‘upstairs’ attitudes towards social media. I fear that the ‘still-printing-out-my-emails-and-dictating-replies-to-my-secretary’ brigade are becoming a serious stumbling block to the effective deployment of social media within many healthcare organizations. It’s partly Luddism, of course, and partly fear of legal reprisals regarding confidentiality, as you say.
How, then, can we get overcome this sort of resistance? Is it a matter of collecting more and more case studies, such as the one you shared, and assembling an irrefutable body of knowledge demonstrating the efficacy of social media within healthcare?
I think there is a need to work closely with patients and try out these ways of delivering out of hospital care. And as you say write up more case studies, once someone in my position (in comms) makes a clinical connection and the 2 units work closer to achieve this then I think we can start to push towards meeting this resistance.
A policy for this kind of engagement will need to be drawn up though, I know NHS Peterborough have a social media policy and I’m hoping they will share their thoughts on this.
Hi Dan,
I think it’s fantastic that you’ve taken the time and effort to write such a thorough piece. The ethics of monitoring patients are a tough one to crack and I believe it comes down to a very personal stance. I would love to be able to get medical advice over Twitter if I had a longer term condition, or even just be able to ask which entrance I should use to access service x or y. Others of course may not like the idea. I don’t think we’ll ever get to a stage where it is acceptable for everyone. In the case of the Kidney patient I may have checked that they were OK for me to pass the details onto the consultant, but otherwise I think it is fantastic that a consultant was open to that information source.
Regarding the justice argument. I’m afraid I don’t think there really is one. There will always be inequalities in any system and we need to minimise those, but when it gets to an argument along the lines of: ‘we shouldn’t do this because not everyone can access it’ it’s ploitical correctness going too far. By communicating with patients in different ways we will help add a lot of value to the system, helping to get information to where it is needed, when it is needed.
Great to see #nhssm flourish with Colin and Tim Lloyd at the helm!
Hi Alex, thanks for responding – and for introducing the case study.
I’m not sure that ethics can always come down to a personal stance, but I know what you mean about individual variation in comfort levels when it comes to social media monitoring. I, like you, would be happy to get basic medical info via Twitter,but I know plenty of people who would freak the freak out at the very notion.
I appreciate what you say about the lack of a justice issue, but I still think the concern pertains. Just because ‘there will always be inequalities in a system’ does not mean one has to accept new situations which may perpetuate or exacerabte those injustices. I was not suggesting that social media in healthcare must benefit *everyone* in order for its adoption to be considered just; rather I was arguing that we ought to be mindful of the ways in which such an adoption can leave those who are already excluded in some ways even further behind. We should definitely, as you say, be excited about adding value to the system through new communications channels, but I think we should also look at ways of making those channels as accessible as possible.
I’m really looking forward the next #NHSSM discussion – it’s a great network you all have established.
How about if I propose a twist to the example? I will pretend this particular patient began dialysis. Because the hospital system had a social media team and used searchable terms to monitor public comment on Twitter, it was noted that a patient had some confusion about her diet. The hospital social media team did decide to share the tweeter information and suggested the physician monitor the tweets. Since the physician was fearful of breaking a trust with the patient or addressing a situation that seemed “Big Brother”-ish, the physician decided not to discuss the patient’s tweets about diet. Let’s pretend she didn’t bring up her thoughts about the diet she was supposed to be following and the physician didn’t review diet recommendations at follow-up visits. Let’s also pretend for her case and her situation, the extreme happened. She died and her death was due to too much liquid in her diet combined with too much of a salt intake. We will pretend that in the phone book or on the web or whatever, it is readily and easily known that this particular hospital system did have a social media department. In this particular situation, a known problem and misinterpretation existed which contributed to death. Who’s liable? The patient tweeted a public tweet that did use the name of the hospital.. the hospital social media department did have the tweet pop up in the searchable column… the physician chose to ignore the public information out of fear of breaking trust and thinking the monitoring was too “Big Brother”-ish. I know if that were one of my loved ones, I would be very saddened to learn that due to a physician or a system having fear of the unknown that my loved one’s misinterpretation of information wasn’t addressed.
Hello SnipperPhysTher,
Your scenario demonstrates why hospitals are not Starbucks. There is much more on the line that just brand reputation. If communications departments decide to start monitoring tweets then they must have a plan/policy in place that all of the hospital is signed up to. The situation you describe would be a tragedy, and it shows why copying the practices of other businesses before there is any consensus on how we should manage this situation.
Thanks,
AM
Hi, SnipperPhysTher – thanks for the intriguing scenario. There’s clearly a balance of harms to be made: is it worse to violate someone’s privacy or worse to let them eat unhealthy food? depends on how unhealthy the food… if it’s very dangerous, like the salt in your scenario, then the calculation is simple: the rule of rescue applies – one has a duty to help a fellow human in danger of serious harm. If it’s just a donut, then…
If your scenario demonstrates anything, it’s that healthcare organisations need to have proper strategies in place for deciding when and how to intervene via social media. They need to think not only about what scenarios would trigger interventions, but they should consider the way in which those interventions are going to be made – just as, in the original case study, the physician brought the issue up in person, not via Twitter.
I hear what everyone seems to be saying about policies… but sometimes policies get in the way of common sense. Now, seriously, back in the day of paper policies in 3 inch 3-ring binders, does everyone really and truly believe that the employees of that department knew (and followed) all the policies within those 3 to 5 3-inch 3-ring binders? I think sometimes we try to create a false security – if we have a policy, we are protected. Kind of a false protection, don’t you think, especially if the employee doesn’t follow a policy or forgets a policy. In fact, I’d even argue that maybe having a policy and an employee not following a policy will be very problematic if a serious situation arose. On the flip side, is it realistic to have a policy for every single possible situation which might arise within a department? I mean, do companies want people freezing and not acting because there isn’t a policy?
In the original case study, obviously there weren’t any policies in place and it seemed as though common sense led to a good outcome. That company knew what the intent of monitoring social media was… found a tweet that would be considered an outlier of the company’s intent – monitored the tweeter – decided it was relevant for a physician to be aware and the issue was properly handled by the physician.
I tend to believe we believe we have more privacy than we think we have. Are our lives really private? All the data mining that occurs… the organizations with the data knows what I buy, when I buy, where I buy, who I’m loyal to… depending on the companies and their strategies, I have information “pushed” to me all the time about products or organizations because of something I did in the past. My actions and choices are highly connected to the level of privacy I think I have. I view social media sort of the same way – someone tweeting is choosing to reduce his/her level of privacy to whatever degree. The health care world may not be Starbucks, but the health care world can jump in and interact. Common sense should kind of dictate how an issue at hand is tackled. Personally, I think it is ridiculous to believe a decision tree of scenarios is the best way to handle strategies and interventions.
Dan: This is a fascinating question, in terms of new social media, but what I found striking about it was how familiar the dynamic is to one my grandfather described to me that he worked in from the early 20th century. He was doctor who worked in a small community, and occasionally had intimate knowledge about his patients’ lives and choices because that small community environment precluded some levels of privacy. For example, he was told by a local butcher if his patients were eating food that he, as a cardiologist, did not want them eating. I wonder if the new social media is helping us reform these communities. The relationship the case study describes between the doctor and his patient, via twitter, sounds far more “thick” than ones many patients commonly have to their caregivers.
Fox! What a pleasure to see/hear you! How is the deepest south?
The historical aspect of this is, indeed, interesting. Our current norms regarding doctor-patient confidentiality are pretty recent additions to the medical ethics register. It’s only a few decades back that doctors would regularly withold information from terminally ill or dying patients, telling families instead. This was gendered, too: well into the 70s, some physicians would inform husbands of their wives’ medical issues, in much the same way they might do regarding a child.
I think that we may be too ‘close’ to the social media revolution – ie: it’s happening right now and we’re in the middle of it – to tell what its historical consequences will be. It seem likely, though, that the reshaping of both traditional communities and traditional moral norms regarding privacy, wil be radically reshaped.
[...] more on this subject see The case of the Twittering Kidney Patient by Dan O’Connor Share and [...]
[...] The Case of the Twittering Kidney Patient: Healthcare and the Ethics of Social Media Monitoringfrom http://www.rmmlondon.com [...]