Skip to main content

Verified by Psychology Today

A New Look at Links Between Isolation and Mortality Risk

A new study investigated their relationship across race and sex categories.

Andrea De la Parra/Shutterstock
Source: Andrea De la Parra/Shutterstock

By Lucy Hicks

Many studies have found evidence for a link between social isolation and risk of death. But a recent investigation, published in the American Journal of Epidemiology, provides support from a sample of the population that is much larger than usual—more than 580,000 people—and also examines whether that connection differs based on race and sex.

Previous studies measuring social isolation and health have largely involved homogenous, predominantly white samples, according to Kassandra Alcaraz, lead author of the study and Strategic Director of Health Disparities Research at the American Cancer Society. “We know that there is consistent underrepresentation of certain groups in research and this area is an example of that,” she says. Due to the size of the sample in the new study, researchers were able to analyze subgroups of black and white men and women.

“There are, unfortunately, racial disparities in a lot of health outcomes,” says Matt Pantell, a doctor and assistant professor at the University of California San Francisco who studies social isolation and its health impacts. “In order to understand those disparities—not only for the benefit of understanding but also for the benefit of determining interventions—it's important to determine if there are different associations between a risk factor and a health outcome.”

In this study, all data were taken from The Cancer Prevention Study II, a pool of more than one million patients initially enrolled in 1982 and 1983. Patient information was gathered from enrollment until 2012 or the years of their death.

The researchers gauged social isolation at the time of enrollment using surveys that determined whether participants were married; attended a religious institution at least once a month; participated in clubs or group activities at least once a month; and had at least seven close friends or relatives. They quantified social isolation by scoring each category with either 0 (more connected) or 1 (more isolated), then compiling the scores from each category to determine an overall social isolation score—the most isolated scoring “4” and the least isolated scoring “0”).

The results showed that social isolation scores were correlated with general mortality risk over the follow-up period, and that was the case across sex and racial groups. The most highly isolated black respondents had an all-cause mortality risk more than double the rate of their well-connected peers, and in white men and women, the most isolated individuals had a 60 and 84 percent higher risk of death, respectively—though these differences between the black and white cohorts were not statistically significant. There was some suggestive evidence that being unmarried might be more closely connected with mortality risk among men than among women and that a lack of religious service attendance may be more connected to risk for black women than black men.

Social isolation was also associated with an increase in mortality risk due to cardiovascular disease in particular—a link that was stronger for white women than for white men. Notably, an association between social isolation and cancer mortality was found only in white respondents. Future research could differentiate by cancer type to see what exactly is driving this difference, Alcaraz says.

A correlational study such as this one cannot prove that factors such as being unmarried or having fewer close friends necessarily cause higher mortality. But it could be the case, for example, that having social support makes a person more likely to attend doctor’s appointments and be held accountable for regularly taking medications. Also, socially isolated individuals tend to have higher levels of anxiety, stress, and depression, Alcaraz says, which can impact physiologic health.

Previous studies have found that highly socially isolated individuals have elevated levels of C-reactive protein, a biomarker for inflammation, Pantell notes. “They're trying to work out the pathway, but in some way, [social isolation] is getting under the skin and leading to biological effects,” he says. “There are a lot of upstream pathways that you could think about—in terms of having resources—and also more internal pathways in terms of the physiological mechanisms that people are investigating now.”

Alcaraz hopes that these findings will inspire future research investigating the mechanisms behind these associations and increase attention to the links between health and social resources. “We take a person’s medical history, but we rarely consider a more comprehensive social history,” she says. “I think the main issue is to just have an awareness that something like social isolation can have deleterious implications down the line regarding one’s health.”

Lucy Hicks is an Editorial Intern at Psychology Today.


Alcaraz, K. I., Eddens, K. S., Blase, J. L., Diver, W. R., Patel, A. V., Teras, L. R., . . . Gapstur, S. M. (2018). Social Isolation and Mortality in US Black and White Men and Women. American Journal of Epidemiology. doi:10.1093/aje/kwy231