Social Networking and Connected Health
Tuesday, March 23, 2010
| Joseph Kvedar, MD
About the Author: Joseph Kvedar, MD is the Director of the Center for Connected Health.
Connected health has three synergistic components: collecting accurate information about you, sharing it with you in context and using that information as a coaching tool. We keep seeing how this triad improves patient engagement and health outcomes. We’ve had some debate about the role of social networking in connected health. We know that social networks in health can be quite powerful. The best examples I know of are Patientslikeme, where patients with a variety of life altering illnesses are banding together online to share information and do amazing things and the Nike+ program, where thousands of runners have banded together to share data about their athletic work outs, both competing with and helping one another. One question is what do these two phenomena have in common? In our experience at the Center for Connected Health, patients with life-altering illness display different behaviors compared to those obsessed with fitness. Could it be that the social network has an accountability component that could be analogous to the coaching component that we’ve noted to be such an important aspect of improving health outcomes? In the connected health synergy, could the social network act as a health coach? Is the sharing of objectively derived health information likely to prompt individuals on one’s network to encourage one to achieve a healthier behavior state? Since all health care professionals, including health coaches, are in short supply, this could be a boon to achieving better health outcomes for the population. The idea would be to convince an individual to share some health or illness-related information (it would have to objectively derived and automatically uploaded) and see if that person’s social network would begin to spontaneously cheerlead and motivate the person to achieve improved health status.
Last November, I read about the Withings wifi enabled scale. This is a very easy to use scale that connects to a home wifi network and begins to immediately upload weight readings over the Internet to the Withings website. One can review one’s weight trends, BMI, add info such as diet etc. When I saw that Withings offers seamless Twitter integration, I saw this as my chance to test my hypothesis. How would my Twitter followers react to me posting my weight automatically daily? Since I know that day to day weight fluctutaions are largely due to water/salt intake, I decided to enter my step count (obtained by the automatically uploaded data from my Fitlinxx pedometer). I wanted to see if either data point was better at bringing out the organic health coach in my network. Finally, I hypothesized that my networks on Twitter and Facebook would be quite different in their response. Twitter is mostly about people broadcasting their thoughts. The hypothesis is that people follow because they find one’s thoughts interesting. Their contract is mostly to listen. Facebook is more like a conversation with friends and family. One expects people to react to one’s status updates. So I hypothesized that posting my weight and steps to Facebook would yield more intrinsic coaching than posting them to Twitter.
To help with the comparisons, some more detail: At the time I did this, I had 123 friends on Facebook and 368 followers on Twitter. I posted my vitals to Twitter for 35 days and to Facebook for 15 days. In reporting the reactions to these posts, it seems like a good way to normalize would be to report as posts or comments per network participant per day (p/np/d). The results then are 0.002 p/np/d for Twitter and 0.004 p/np/d for Facebook (I don’t think these two numbers are different in a meaningful way). Qualitatively speaking, the comments were pretty similar. They could be categorized as questions (about the technology or my intent), educational comments (diet mavens reminded me that checking daily weights in not recommended practice), or inspirational comments. They sorted roughly 1/3 of each. The inspirational comments came largely from a small number of people (less than 5). The end result: my weight remained stable. It has fluctuated + or – 3 lbs for some time now.
Here are some possible lessons:
- I was not intrinsically motivated to lose any weight. Sure it’d be nice to drop 5 lbs, but I had been given no such advice by a health care provider. Connected health can help people who are not motivated to become more motivated, but it probably requires a more dedicated coach.
- I purposefully did not broadcast to either network that I was seeking to recruit the network as my health coach. I wanted to see if this phenomenon would spontaneously develop. It did not. Both the Nike+ network and Patientslikeme are based on the concept that people who share a common goal can support one another. Maybe I would have had better luck if I first sought to find a network of folks interested in losing weight through mutual reporting and coaching.
- I can’t prove it but I think I was slightly more active during this time. Not enough to sustain any weight loss, but the discipline of daily reporting of my step counts to match the automatically uploaded weights kept my activity top of mind. I found ways to get more steps in and I’ve kept those habbits. One of the priniciples of connected health is that its more powerful to share your objectively derived health data with a coach than to simply be aware of it yourself. Even though the network effect of encouraging me to be more active or lose weight was not overwhelming, the fact that I was openly sharing helped me be more thoughtful about both activity and calorie intake (especially over the Christmas and New Year’s holidays).
I thought I was done until a few days ago when one of my most important network members (my son) sent me a text message – “I miss the daily weights”. It is likely that this was a case of good-natured ribbing, but I did take notice. Clearly there is more to learn here.
I’ll continue to look for new, interesting sensors that automatically transmit their data (this is an important developmental step for connected health because it makes sharing one’s data so easy) and exploring ways to create natural coaching situations or take advantage of applications that offer automated coaching. The synergy between objectively collected health information and coaching as a tool for improving health behaviors is undeniable. Now it’s a matter of fine tuning.