social media buttons on cell phone screen
Image via Creative Commons, JasonHowie’s Flickr photostream. (source)

How can social media give us knowledge that can help people live healthier lives and help scientists cure diseases?

We’re not asking this question rhetorically, we’re asking it really and truly. Over the past couple of years at RWJF, we’ve received a handful of proposals that want to explore ways to harvest social media data to provide insight into disease, health behaviors, and population health issues. We’ve also received proposals to develop and/or test ways to harness social media data to improve health and facilitate research.

While we think social media holds a lot of promise here, we’ve turned down most of these harvesting and harnessing proposals. The reasons for the turndowns have varied, but a theme underneath them has been a sense that the proposals have all been one-offs, focused tests of really early stage applications that are unlikely to move into general use after the test is over.


Time for a closer look

But we think it’s time for us to take a closer look at this question of how social media can give us knowledge and help us live healthier lives. So, let’s start by assuming that at some point in the future, social media data is easily accessed, harvested and harnessed for the cause of better health. Now, think about where we are today relative to that future. We want to hear from you about the barriers and limitations that exist between current and future states.

For instance, barriers could include insufficient privacy controls; public fears over discrimination and abuse; lack of human subjects protection/consent in data use; or simply, concerns about “creepiness” in using social media data in this way. Limitations could be technical. For example, there’s no good way to link social media data with EHR data; natural language processing hasn’t advanced to the point where it’s easy to identify and extract signal; or there are issues with representativeness of sampling. Finally, we have questions about the meaning and interpretation of study results when using these types of data.

Ultimately, we want to see if there are some key issues that are in the way of us getting to that future.


Building a culture of health

As we work to build a Culture of Health, we need to continually explore cutting edge ideas and emerging trends that have the potential to accelerate our progress. If we end up funding in this space, we think investments that benefit the whole field—that help address general barriers and limitations—will be more productive in the long run.

So please share with us your thoughts about harvesting and harnessing, barriers and limitations. Please comment below or write to us at We’ll keep track of the comments and share back a synthesis.


  1. This article discusses many challenges that exist when using social media in any kind of population study, and all the warnings apply to improving healthcare: Ruths, Derek & Pfeffer, Juergen (2014). Social Media for Large Studies of Behavior. Science Vol 346, Issue 6213, 1063-1064,

  2. I think that social media shows a lot of promise simply in the volume of people that use it. Stable healthcare relies on communication, and it is inarguable that many people keep up with social media on a regular basis. We know from Pew Research studies that almost 75% of internet users use Facebook. We also know that even 75% of low-income Americans use the internet, a number that increases substantially with income (97% of households above $75k/year!). If we’re looking for mediums to engage patients/healthcare utilizers, it’s right in front of us.

    That is not to say that social media will be a vital component of the healthcare process. Given the issues you mentioned with privacy and technical linkage, its hard to ever imagine Cerner feeding an EHR into a Facebook profile. Where it will be useful is in developing demographic profiles to better understand where specific health interventions can make the most difference. There is no single solution that covers all populations; states, communities, neighborhoods will all have discrete, unique healthcare challenges. But knowing more about the population is an invaluable step towards understanding what interventions will work best. I imagine that social media can play a larger role in the furthering that understanding.

    • Thanks Alex, I agree that the sheer volume of people using social media means it is a potential goldmine for population level health information. Since it is the primary means of communication for so many people, I expect it is only a matter of time before it gets harnessed for many other health care applications as well. The recent hacking and the public posting of private information from the Ashley Madison website makes it clear that no website (nor paper chart for that matter), is hack-proof. So, I believe, crafting a solution will always mean a balance between privacy consideration and the benefits of using the medium. Pat


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.