December 2, 2016

Living longer with friends

By: Moira Burke

People who have stronger social networks offline live longer [1]. When it comes to longevity, social relationships are as predictive as smoking, obesity, and physical inactivity [2].

However, past studies don’t tell us whether online social interactions are as good as offline ones when it comes to better health. On the one hand, social media could make it easier to keep in touch with supportive friends, strengthening social connections and healthy behaviors. Or the internet might reduce physical activity, replacing in-person interactions with more passive, lower quality ones. What do the data say? How does Facebook activity fit in?

To answer these important questions, Facebook researchers partnered with academics at UC San Diego, and Yale. We find that online social interactions are linked to longer life.

Hobbs, W., Burke, M., Christakis, N., and Fowler, J. (2016). Online social integration is associated with reduced mortality risk. Proceedings of the National Academy of Sciences.

We find that people who use Facebook live longer than people who don’t. In a given year, the average Facebook user is about 12% less likely to die than someone who doesn’t use the site.

However, these effects may be due to differences in the kinds of people who use the site: They may have more social connections or more access to technology. So more importantly, among people who use Facebook, those with high levels of offline social integration—appearing in photos—and moderate levels of online social interaction—writing messages and status updates—have the greatest longevity.

To understand the relationship between online social interactions and longevity, we compared a random sample of California Facebook users with Facebook users who appeared in public California vital records indicating that they had died. After aggregating and de-identifying the data, we compared counts of their activities on Facebook, such as the number of photos they appeared in and the number of status updates they wrote. The research was observational—no one’s experience on the site was any different than usual—and the project was approved by UC San Diego’s Institutional Review Board, California’s Committee for the Protection of Human Subjects, and the Vital Statistics Advisory Committee at the California Department of Public Health with a goal of understanding the kinds of online social experiences that may be protective of health.

Consistent with classic studies, we find that people with larger networks have better health. Beyond past studies we can also see that people who receive a lot of friend requests have greater longevity, while people who initiate a lot of friend requests don’t live longer (Figure 1). These results suggest that better health isn’t determined solely by an individual’s ability to seek out connections. Instead, it depends on others wanting the friendship as well. Unfortunately, this suggests that public health interventions to increase people’s capacity to seek support may not have the intended effect of improving health.

friends_deciles

Figure 1. People who accept more Facebook friendships live longer but initiating friendships is not associated with significant differences in longevity. The y-axis shows mortality risk in a Cox proportional hazard model compared to the reference category (middle decile, indicated with a square), controlling for age, gender, mobile device use, and Facebook account tenure.

Our findings suggest that for most people, online interactions are related to better health. People who use Facebook more tend to live longer and are less likely to die from illnesses that other research has linked to isolation, such as cardiovascular disease or substance abuse (Figure 2). However, in the relatively rare case when people use Facebook at extremely high levels without also participating in offline interactions (e.g., they post dozens of status updates but don’t appear in any photos indicating offline activity), we find the opposite result. Online social interactions seem to be healthiest when they are at moderate levels and complement offline interaction — and for the majority of people, it looks like they are.

friendships_type

Figure 2. Having more Facebook friends, especially being the recipient of those friend requests, is linked to a lower mortality risk for cardiovascular disease (CVD), drug overdose, and suicide, causes of death known to be associated with social isolation. Most forms of cancer are not associated with social behavior offline, and the same appears to be true online.

In most cases, sharing posts and messages on Facebook is linked to a longer life. However we were surprised to find a small positive relationship between Facebook posts/messages and cancer mortality, perhaps reflecting an effort by Facebook users to seek emotional support or share updates once they are diagnosed with a chronic illness.

While we show that Facebook use and health are correlated, we cannot prove the relationship is causal. It is possible, for example, that healthy people are more likely to attract others to befriend them. On the other hand, association studies like ours can be an important first step to help other scholars design research where it is possible to identify the underlying causes of the relationships we have discovered.  The classic offline studies have spurred a lot of useful research on the role of offline social life in health, and we hope that our study does the same for the online world.

Notes:

  1. Berkman, L. F., & Syme, S. L. (1979). Social networks, host resistance, and mortality: a nine-year follow-up study of Alameda County residents. American journal of Epidemiology109(2), 186-204. The results have been replicated hundreds of times.
  1. Holt-Lunstad J, Smith T.B., Layton J.B. (2010). Social relationships and mortality risk: a meta-analytic review. PLoS Medicine 7(7):e1000316.