« High Stakes Testing: There is No Free Lunch | Main | Dead Horse Flogged »

September 02, 2005

Friends for Life

There's no question that our social environments help determine how healthy we are. In fact, I have now in front of me a small mountain of studies to that effect. Rather than bore you with the particulars of their findings, I'll simply summarize this way: study after study shows that close social relationships--a spouse, loved one, or a close friend--help people live longer.

Interestingly, social bonds don't appear to prevent the onset of a disease. There's no association, for instance, between social isolation and sudden cardiac death. But for survivors of a life-threatening event or disease, social connectedness significantly increases average longevity and functioning. One study even suggests that the effect of social isolation "is comparable to the effect of cigarette smoking on total mortality reported in some studies." 

In other words, close relations with friends and family are really, really good for your health.  Or--stated in the converse--loneliness kills.

Perhaps more intriguing, an emerging body of research is pointing out that suburban sprawl is an impediment to social networks. Does sprawl erode social networks which are critical for health? Is sprawl bad for our health because it diminishes our personal relationships? Well, that's where things get confusing.

I've already aired my skepticism that sprawl erodes social capital. (Or more precisely, I'm skeptical that the existing research proves it.) But for the moment, I'll abandon my skepticism and go along with the multitude of voices arguing that sprawl is bad for social capital. And indeed, there is some compelling research showing that traditional city neighborhoods are better at fostering incidental contact between neighbors and promoting loose associations that may be important for well functioning civil society and even democracy. Plus, it's clear that declining social capital over the last several decades has been coincident with ever-more dispersed suburbs and highways.

One might be tempted to conclude that if sprawl is bad for social capital (which is good for health), then sprawl must also be bad for health. One would even have plenty of company among researchers who have claimed just that. But one would still have to convince me. Here's why.

My reading of the available literature suggests that there's a pervasive equivocation at work. The term "social capital" (or, variously "social networks" or "social ties") is used in at least two different ways. In one sense the term(s) refers to informal neighborhood associations, participation in civic life, or belonging to a church or community group. I'll call this "loose social capital." In the other sense, the term(s) means the presence of close supportive individuals, such as a spouse. I'll call this "tight social capital."

The research makes it clear that tight social capital is good for health, but I haven't heard of anyone arguing that sprawl reduces tight social capital. Moreover it's not at all clear that loose social capital--the kind that is allegedly eroded by sprawl--has anything at all to do with health outcomes. But because the literature's terminology for both tight and loose social connections are the same--social capital--it's easy to assume that they are the same thing and have the same effects.

Now, admittedly, there is some evidence to suggest that people with more friends are healthier (and this is especially true for men for some reason). But that, of course, doesn't mean that having friends makes you healthier. And it's also true that at least one study shows that too many social connections can actually be inimical to health--interpersonal conflicts cause stress-related diseases (and, again, this is especially true for men for some reason).

I do think there's probably a link here--that is, a link from sprawl to loose social capital to health--but I'm not quite convinced yet. The best research I've seen is from Ichiro Kawachi, a Harvard researcher. On a state-by-state basis he examined the results of two simple questions that have been found to be closely correlated with loose social capital---Do you think most people can't be trusted? and Do you think most people would take advantage of you if given the chance?

States where people thought they couldn't trust others also had worse rates of self-reported health. And states where people felt that others would take advantage of them had higher mortality rates. So Kawachi's study makes me think that there must be some connection between loose social capital and health. And if sprawl does indeed weakens loose social capital, then it may be to blame for worse health outcomes. I'm hoping there's more research emerging that will document these connections, if they exist. In the meantime, I'm also hoping not to find more research that seems to equivocate between loose and tight social capital.

Postscript: In a curious though unrelated side note, one study found that among spouses, husbands had a higher risk of depression if their wives suffered a cognitive disability. On the other hand, if their husbands suffered a cognitive disability, wives were at no greater risk for depression. I'm not really sure what to make of that.

Posted by Eric de Place | Permalink