PRB

The more social ties people have at an early age, the better their health in young adulthood and old age, according to a 2016 study in the Proceedings of the National Academy of Sciences.

These findings are based on data from four nationally representative, federally funded longitudinal studies that follow the same people over many years.¹

Such studies measure changes in an individual’s health over time, allowing researchers to identify cause and effect, according to Kathleen Mullan Harris, a sociology professor at University of North Carolina-Chapel Hill, faculty fellow at the Carolina Population Center, and a member of the research team.

Their study was one of the first to definitively link social connections with specific health measures such as abdominal obesity, inflammation, and high blood pressure—all of which increase the risk of disease, including heart disease, stroke, and cancer.

PRB spoke with Mullan Harris about their research. A partial transcript follows; an audio recording is available above.

PRB: These results are surprising. Can you help us understand how our friendships and social ties affect our health?

Kathleen Mullan Harris: Sure. We were interested in how social connections affect health. It’s been known in the literature for some time that social connections were associated with longevity, in other words, lower death rate and a longer life expectancy, but mainly this research has been done for just older aging populations.

We have the ability to try to understand the underlying biological mechanism that explains this relationship. 

For example, when we experience some threat or a challenge, our brain sets off a distress signal to other bodily systems that get us ready for that challenge. Certain hormones are released to respond to that threat. It gets us ready, our muscles receive additional energy. We release epinephrine as well as cortisol and that’s a normal healthy response to stress.

PRB: So what’s an unhealthy response to stress?

Kathleen Mullan Harris: The problem occurs when people experience stress on a regular basis; you might think of living in a dangerous neighborhood where you’re always vigilant for hearing gunshots or something like that. In those situations, our body responds by keeping alert to this threat, and it never returns to its equilibrium where cortisol levels will go back down to normal. When we’re experiencing  the chronic daily stress in this way, our body has a physiological response that is bad for our health. It means that we have an increase in inflammation. Our heartbeat is always elevated. We have high blood pressure. We have cortisol circulating in our bloodstream all the time, which adds fat to the middle parts of our body.

That’s a long way of saying that what we suspect is going along with the social connections in terms of affecting health is that it acts as a buffer to these types of chronic daily stressors. It provides a sort of social and emotional support used for the kinds of things that we face every day. 

We looked at the relationship between the social connections that people with these biological outcomes have in their lives and find that indeed, the more social connections that individuals have, the lower these health risks or the lower inflammation, obesity, and hypertension.

PRB: I think from your findings, you say that age makes a difference, that our relationships impact our health in somewhat different ways at different ages. What did you observe there?

Kathleen Mullan Harris: That’s really the other new part of our research. As I mentioned this relationship has mainly been examined in aging populations and what we did in our research was put together sources of data from four different studies that allowed us to examine the relationship between social connection and health risk across the life course, so beginning in adolescence into middle adulthood and then in aging or older adults.

What we found is that social connections are very important in reducing the health risk in young adulthood and later life, so among aging individuals. [But] in middle adulthood, we found that it wasn’t so much the amount of social connections that people have, but it was really the quality of those connections—what those connections actually give you in terms of social support or strain.

PRB: So for middle-age adults, quality matters more than quantity?

Kathleen Mullan Harris: Yes, that was really an interesting finding, that the social connections, the actual number of different domains in which individuals were embedded wasn’t as important for middle-age adults. But what was important was the quality of their connections. What we think is going on here is that middle-age adults are already highly embedded in many different social networks. They have people that they work with, their neighbors, they have children. They likely have living parents, and they’re involved in networks with their children’s friends’ parents, and so they had many different connections, but what those connections actually give them in the sense of social support or strain, is what we found made a difference with their health.

The extent that those connections provided social support, provided someone that they could talk to about their problems, or who was always there for them improved their health, lowered these health risks that we were measuring. But if those connections actually created strain in their lives by making demands on them or criticizing them, letting them down, then that increased the poor health outcomes that we were examining. We find it very interesting that early in life or later in life [friendship] is more voluntary—where you go out and get involved with social life, and you meet friends and you join organizations. In the middle of life it’s less voluntary. You’re naturally in many of these different networks, and so it’s really what those networks give you that seems to matter for your health in middle adulthood.

PRB: Toxic colleagues at work or difficult family relationships take a toll during middle age?

Kathleen Mullan Harris: Exactly.

PRB: What does social isolation look like? Are introverts and shy people less healthy for example?

Kathleen Mullan Harris: Yes that’s a good question. So we measured social integration, or what I’ve been referring to as social connections. What’s important to understand is that we examined integration across multiple domains of social life so in the adolescent period, for example, we examined the extent to which adolescents were connected to their families, to their friends, within religious institutions, and in the school in terms of activities. We measure social connections across many different domains and then we do the same in middle adulthood and older adulthood, just changing the context to what’s developmentally appropriate for that stage of adulthood.

Social isolation then is being isolated across multiple domains. It’s not just having friends, but it also means that you’re not really actively engaged with your family or within the community or school, so that’s what isolation looks like. I think the question about being shy or introverted, I mean, of course that makes sense, but because we capture this in multiple domains, we think that social isolation is capturing more than just somebody who’s shy or introverted. Even very introverted individuals are actively engaged with their family, perhaps belong to some community organization, or go to church. We’re less concerned that we are really picking up that personality aspect.

PRB: What about for young people? For many of them, being social involves smart phones and social media. Do those connections count too or is that something different?

Kathleen Mullan Harris: Good question. Well, we did not include that, and probably because we had questions about Internet access, about social media. The survey in which the young people were interviewed started in 1995 for adolescents. Then we’ve been following them over time. Social media wasn’t such an incredibly a huge part of our social life back then as was today. That might be important probably for more the friendship connections than some of the other connections.

PRB: And for the older people, I wondered if health doesn’t also affect our ability to socialize? For some elderly people, getting out of the house is difficult. Do we know that it’s not health affecting social connections rather than the other way around?

Kathleen Mullan Harris: Right, absolutely. Something we always worry about in our social research. Fortunately, what we were able to do here is control for health, which means we take advantage of the fact that we know what the individual’s health is before we measure their social integration. That allows us to actually say, okay given what your health is now, we’re going to see how many social connections you have and then see what the impact of those social connections are three years from now. We’re able to actually look at the change in health.

PRB: So you’re following people over time and it’s the individual change that you’re looking at?

Kathleen Mullan Harris: Right. It’s the individual change. We can’t do that for all of the data sets but we can do it for some of the data sets and that gives us confidence. In fact, I think for one of the findings, we try to provide a way to interpret the importance of social connections and we found that the effect of social connections was as important as being engaged in physical activity for reducing health risk in adolescence for example.

PRB: Wow. Are there takeaway messages that you would give parents or school administrators based on this work, particularly related to adolescents?

Kathleen Mullan Harris: Yes. Two things that are really important to me. One of the reasons that I am interested in bringing together the sort of biological data with the social data to look at relationships like this is to bring the importance of social factors in health to the medical community. I think that our results show that when doctors, for example, are conducting an annual checkup either with young people or old people, they should be asking them what their social life is like, how often do they go out during the week, who are their best friends. That’s one implication that we hope we’ll carry over into the medical field.

Then for the young people, we were surprised actually to find social connections to be so important early in the life course. It seems that schools could think about adolescents who are particularly isolated and reach out to them, maybe create certain structures or activities that might pull in some of the adolescents who aren’t as engaged in the social life of the school.

For parents for sure, it seems like this is an important educational finding that parents should be sort of monitoring their adolescent’s social life and making sure that they’re getting out and being around other people.

PRB: That their social life has an impact on their health so it’s just as important as eating well or getting enough sleep or … ?

Kathleen Mullan Harris: Or exercise, exactly.

PRB: You mentioned that you used four different separate surveys for this study. What was unique about each of them?

Kathleen Mullan Harris: These studies are all nationally representative so that means that our findings relate to the U.S. population, and they were all longitudinal as you mentioned. That means that they follow the same people over time which is very important. 

What was unique about them is they cover different stages of the life course so we use the Add Health study [National Longitudinal Study of Adolescent to Adult Health] which is a study of adolescents who are being followed into early adulthood, young adulthood. Then, we use another study called MIDUS [Midlife in the United States] which is a study of middle adulthood and then we have two national studies of aging adults NSHAP [National Social Life, Health, and Aging Project] and HRS [Health and Retirement Study] which study adults above age 50. Basically, what we did is we harmonized all of our measures of the social integration as well as our measures of social support and social strain. We harmonize all of the biological mechanisms. We ran the same analysis in all four data sets. That allowed us to put together our findings across the life course or the age of adults and observe how they were the same and different.

PRB: You could take a young person and look at them over time and see what happened with obesity or inflammation based on the kind of social life they had.

Kathleen Mullan Harris: Right. We could say for this individual who comes from this area of the country and is this race or ethnicity and gender, here’s what social connections means for their health in adolescence, in middle adulthood, and then in older adulthood.

PRB: Do you hope that these findings will provide evidence for policy change or program interventions?

Kathleen Mullan Harris: I hope so. One of the nice things about identifying social factors is that you can usually change them. I mean some are more expensive than others but they’re usually changeable. [Findings related to social factors] usually  result in a number of different implications for policy.

PRB: So these findings might be the basis for heading off the obesity epidemic or future diabetes …

Kathleen Mullan Harris: Yes. I mean I think where our policies would help in terms of the obesity epidemic is that we can curtail or put a dent in that relationship between obesity and disease. I think there’s been some movement. I think from what I gather, we’ve got the plateauing of obesity so in other words, obesity isn’t rising anymore. It’s kind of flat, so I imagine that there will be continued policies to work on lowering obesity. For our cohort, the main thing that we’re trying to follow is that the time from health risk, which is obesity, to disease. We would like to extend that time by identifying what factors matter most and then in turn, the policy implications—[policy changes or new health recommendations] that could act on those factors. Even if you are overweight or obese but you exercise, that’s healthy. That’s going to lower your blood pressure. I don’t know how much of an impact it will have on lowering inflammation but it is still healthier than being inactive.

Reference

The four, federally funded longitudinal studies are the National Longitudinal Study of Adolescent to Adult Health (Add Health); Midlife in the United States (MIDUS); National Social Life, Health, and Aging Project (NSHAP); and the Health and Retirement Study (HRS).