Skip to Main Content

Getting to the hospital quickly is essential for treating heart attacks, strokes, and other medical emergencies. You might guess that spouses or family members would be best at reacting quickly. You’d be wrong.

My team at Brigham and Women’s Hospital studies social networks. Few humans are solitary creatures; most of us are embedded in one or more social networks that include family members, friends, and acquaintances. We believe that these networks are an under-recognized factor in illness and health, and could be harnessed to improve our lives.


For our most recent test of this idea, we recruited 175 men and women who had recently experienced a mild or moderate stroke. Strokes usually happen in front of other people, and such witnesses are the life-lines who detect, plan, and act when a person is in trouble. After interviewing each participant, we mapped their social networks to identify the patterns of close and distant individuals in their lives.

As we reported in the journal Nature Communications, the connection between social network and hospital arrival was seemingly counterintuitive. Those who arrived at the hospital quickly tended to have larger networks of loosely connected people (an average of 8). Those who arrived later tended to have smaller networks of closely knit people (an average 5).

Regardless of age, race, or stroke type, the social network affected response times. The difference in arrival times had to do with communication patterns in the networks. Individuals in small, tight networks discussed or argued about the plan and then agreed to watch and wait. Those in large open networks negotiated less and disrupted plans to wait and see.


One of my patients, I’ll call her Mary, was enjoying the holidays with her family. After dinner one night, she and several family members noticed that her words were jumbled. They mulled it over and agreed that she should take a nap. When she woke up and was still jumbling her words, they rushed her to the hospital. But by then she was past the window of opportunity for her to receive emergency stroke treatment.

Another patient, I’ll call him Mark, was a clear example of how social networks can help. He had trouble speaking in front of Harry, an acquaintance. Harry decided the symptoms were serious and, without negotiation, called 911. By doing that, Harry likely had a profound long-term effect on Mark’s health and ability to function.

Delay, more than any other factor, limits treatment for strokes and heart attacks. Only 25% of patients arrive within three hours of the onset of heart attack or stroke symptoms. Because of this, only 5% to 10% of patients receive advanced medications or procedures. Delay in seeking treatment is a major reason for disability and death worldwide. Mass education campaigns to increase awareness have done little to change these numbers.

The role of the social environment on arrival time in heart attacks was first studied in the 1980s. Researchers calculated the time that witnesses spent before calling for help. It took nonrelatives 20 minutes to call, family members 30 minutes, and spouses 35 minutes. The closer the personal connection, the longer individuals analyzed symptoms, negotiated possibilities, and avoided taking control.

Many people assume that their most familiar contacts will make accurate medical observations. In fact, the importance of the less familiar is a well-known pattern in sociology. In the “strength of weak ties” theory, a weakly connected familiar is a potential source of new ideas. In an emergency, the weak tie can be the outside voice of reason who isn’t afraid to disrupt the group’s ideas. The close-knit core, in contrast, is vulnerable to recycling and over-analyzing information. They are also more likely to build a plan that is easy and agreeable — and not necessarily right. Though well-intentioned and loving, the closed group is a flawed echo chamber.

The take-home message of our work is that important life decisions, such as whether to seek medical care, are often social. We like to discuss, weigh options, and make a group decision with others. Unfortunately, deliberations with intimates may work against us in emergency situations. In such scenarios, an outsider might be the best advocate. Learning when and how to include an outsider could lead to exciting new interventions. What if Amazon’s Alexa, for example, could be the outside voice and activator of emergency services?

Witnesses are essential in medical emergencies. In such situations, love or care is only one of the key ingredients. The network needs to think together. A diverse network that is open to new ideas is valuable when we are most vulnerable.

Amar Dhand, M.D., is an assistant professor of neurology at Brigham and Women’s Hospital and Harvard Medical School and director of the Dhand Lab. The stroke research described here was supported by a grant from the National Institutes of Health.