How Facebook May Improve Your Health

What role can—and should—Facebook play in our health? Can social networks impact determinants of health, and if so how?

It’s a fascinating question. To get some more insight I turned to Dr. Freddy Abnousi. Freddy is an old friend, a practicing interventional cardiologist, and the head of healthcare research at Facebook. Freddy shared his thoughts recently on my podcast, A Second Opinion.

About three years ago, Freddy came to the same realization I once had: no matter how many patients I could heal as a cardiac surgeon, there were scores more I couldn’t help, patients I wouldn’t get to know until their health was too compromised by a lifetime of health decisions. Patients for whom I arrived too late.

“Regardless of how many heart attacks you fix, it ultimately will never move the needle for the public’s health,” Freddy told me. “If your focus is trying to impact that, then you have to be able to step out of the hospital and think about where health really actually happens, which is outside of the hospital.”

Many years ago, I turned to Washington, committed to impacting health through policy. More recently, Freddy sought to better understand what he thinks is the biggest driver of health outcomes in America: poorly understood social and behavioral variables.

Health in the United States is impacted by exposure, access to quality care, genetics, and social and behavioral variables. Freddy believes that social and behavioral variables—the last category—add up to 55% of overall health outcomes in the U.S.

We’ve been missing more than half the picture.

Until recently prevention meant giving general health advice: exercise; don’t smoke; eat a diet that is calorie-restricted and rich in vegetables and fish. We haven’t been able to address prevention with a complete understanding of communities. Social and behavioral variables comprise income, education, and race, which can be broken down further into issues of transportation, housing insecurity, food insecurity, and much more.

This isn’t just a theory. Freddy mentioned a study published in Circulation Research in 2017 that followed a population of nurses in Massachusetts for about 20 years and identified the most impactful risk factors for a heart attack. Generally healthy nurses signed up, and researchers identified risks from clinical issues (hypertension, diabetes, obesity, etc.), lifestyle (smoking, drinking, diet), and social integration.

Social integration in this study was defined by the Berkman-Syme Index, which includes the number of friends you have, the number of groups you’re a part of, the time spent in these groups, whether you’re married, whether you go to church, and more.

The study found that if you control for all the clinical risks and all the lifestyle risks—including smoking, which was surprising to us both—the least socially-integrated nurses had a significantly higher chance of having a deadly heart attack than the highest integrated quartile.

“This is not simply a food and transportation issue,” Freddy pointed out. “The way they looked at it was, ‘How locked into your community are you?’ My suspicion is that if we do a bunch of research toward this, at the end of the day, we will be trying to recreate what human beings were like a couple of hundred years ago where we were really in tune to each other.”

How does Facebook—certainly not facilitating relationships one hundred years ago—hope to help?

The vast majority of Americans interact on Facebook and other social media platforms, Freddy explained, and we must be able to mine those data for health findings.

Our datasets thus far have been lacking the granularity and complete understanding of a human being and their families and friends and communities, he argues. “And there’s only one place in the world that really solved for that very well today.”

Already research has introduced the concept of “social biomarkers”, behavioral indicators that predict health and disease. For example, research out of MIT found that young women with worsening mental health problems, particularly at risk of suicide, changed the way they interact within their social network from essentially having a few relationships that are fairly deep to having many relationships that are fairly shallow.

Freddy expects careful study of social media data will reveal more markers that can help us intervene much earlier in crises, but also better illuminate other social determinates of health.

“Perhaps there’s a set of principal components that go into making up income, education, and race, which maybe are actionable. If we understand those, we start to walk towards understanding this ‘nurture source code’, if you will, where we can finally have a scalable solution that’s not just applicable to one group versus another,” Freddy said.

There are certainly issues of privacy to address, but I agree with Freddy’s assessment. These are the types of studies we can use to understand and impact health in important ways that stretch far beyond diet and exercise.

“We continue year on year, in my opinion, the greatest nation in the world, being at the bottom of the OECD list of premature mortality for essentially every major category every single year.” Freddy said. “What have we got to lose?”

You can hear our whole conversation (and a lot of other great ones) on the August 5 episode of the A Second Opinion podcast.