It has long been known that social connections and health behaviors are related. We can see this at the individual level: If our friends are working out or dieting or quitting smoking, we are more likely to do so too. But at the population level, understanding precisely how interactions affect behavior has proven difficult. Many factors influence behavior, and while we can see correlations, cause and effect are hard to pin down.
The rise of online health communities and social media, however, has opened important new avenues for research in this area. There are now many intentionally created communities organized around particular health concerns, from dieting to exercise to smoking cessation to combating any number of conditions. Facebook, Twitter, and other social media also have given rise to commercial applications that offer radical approaches to improving health.
By studying what happens inside online communities, social scientists now can observe connections between interactions and outcomes. Investigators also can conduct what has long been the gold standard for scientific research—the randomized, controlled experiment. In much the same way that medical researchers do clinical trials on individuals, social scientists are beginning to do clinical trials on entire populations.
My own research investigates how changes to the social network in an online population can accelerate the spread of health behaviors and technologies. One of these studies surprisingly found that the network configurations among the members of an online health community that could most effectively spread health information actually inhibited the spread of behavior change. Another study, among the participants in a 3-month online fitness program, found that the characteristics of the program contacts—including their age, gender, and body mass index—could dramatically affect the likelihood that the participants (most notably those who were obese) would adopt an online dieting technology. Social scientists have long argued that health behaviors are highly susceptible to changes in the composition of people’s social networks. The results from these studies show that we can causally demonstrate how this works, opening the door to large scale, systematic research on a vast array of social influences on health.
We are now in a position that we couldn’t have imagined 20 years ago. New information and communication technologies are starting to generate real-time medical data, which will allow us to connect social interactions with specific health outcomes, such as obesity, heart disease, and diabetes. With these and other technological advances, we can systematically study the social determinants of health using large scale experiments, that are replicable over many trials. This will hopefully allow us to discover significant new strategies for improving health outcomes on an a population scale.
Damon Centola is an Assistant Professor in Behavioral and Policy Sciences at MIT’s Sloan School of Management.