CHAPTER 20
Health and Happiness
OF ALL THE DOMAINS in which I have traced the consequences of social capital, in none is the importance of social connectedness so well established as in the case of health and well-being. Scientific studies of the effects of social cohesion on physical and mental health can be traced to the seminal work of the nineteenth-century sociologist Émile Durkheim, Suicide. Self-destruction is not merely a personal tragedy, he found, but a sociologically predictable consequence of the degree to which one is integrated into society—rarer among married people, rarer in more tightly knit religious communities, rarer in times of national unity, and more frequent when rapid social change disrupts the social fabric. Social connectedness matters to our lives in the most profound way.
In recent decades public health researchers have extended this initial insight to virtually all aspects of health, physical as well as psychological. Dozens of painstaking studies from Alameda (California) to Tecumseh (Michigan) have established beyond reasonable doubt that social connectedness is one of the most powerful determinants of our well-being. The more integrated we are with our community, the less likely we are to experience colds, heart attacks, strokes, cancer, depression, and premature death of all sorts. Such protective effects have been confirmed for close family ties, for friendship networks, for participation in social events, and even for simple affiliation with religious and other civic associations. In other words, both machers and schmoozers enjoy these remarkable health benefits.
After reviewing dozens of scientific studies, sociologist James House and his colleagues have concluded that the positive contributions to health made by social integration and social support rival in strength the detrimental contributions of well-established biomedical risk factors like cigarette smoking, obesity, elevated blood pressure, and physical inactivity. Statistically speaking, the evidence for the health consequences of social connectedness is as strong today as was the evidence for the health consequences of smoking at the time of the first surgeon general’s report on smoking. If the trends in social disconnection are as pervasive as I argued in section II, then “bowling alone” represents one of the nation’s most serious public health challenges.1
Although researchers aren’t entirely sure why social cohesion matters for health, they have a number of plausible theories. First, social networks furnish tangible assistance, such as money, convalescent care, and transportation, which reduces psychic and physical stress and provides a safety net. If you go to church regularly, and then you slip in the bathtub and miss a Sunday, someone is more likely to notice. Social networks also may reinforce healthy norms—socially isolated people are more likely to smoke, drink, overeat, and engage in other health-damaging behaviors. And socially cohesive communities are best able to organize politically to ensure first-rate medical services.2
Finally, and most intriguingly, social capital might actually serve as a physiological triggering mechanism, stimulating people’s immune systems to fight disease and buffer stress. Research now under way suggests that social isolation has measurable biochemical effects on the body. Animals who have been isolated develop more extensive atherosclerosis (hardening of the arteries) than less isolated animals, and among both animals and humans loneliness appears to decrease the immune response and increase blood pressure. Lisa Berkman, one of the leading researchers in the field, has speculated that social isolation is “a chronically stressful condition to which the organism respond[s] by aging faster.”3
Some studies have documented the strong correlation between connectedness and health at the community level. Others have zeroed in on individuals, both in natural settings and in experimental conditions. These studies are for the most part careful to account for confounding factors—the panoply of other physiological, economic, institutional, behavioral, and demographic forces that might also affect an individual’s health. In many cases these studies are longitudinal: they check on people over many years to get a better understanding of what lifestyle changes might have caused people’s health to improve or decline. Thus researchers have been able to show that social isolation precedes illness to rule out the possibility that the isolation was caused by illness. Over the last twenty years more than a dozen large studies of this sort in the United States, Scandinavia, and Japan have shown that people who are socially disconnected are between two and five times more likely to die from all causes, compared with matched individuals who have close ties with family, friends, and the community.4
A recent study by researchers at the Harvard School of Public Health provides an excellent overview of the link between social capital and physical health across the United States.5 Using survey data from nearly 170,000 individuals in all fifty states, these researchers found, as expected, that people who are African American, lack health insurance, are overweight, smoke, have a low income, or lack a college education are at greater risk for illness than are more socioeconomically advantaged individuals. But these researchers also found an astonishingly strong relationship between poor health and low social capital. States whose residents were most likely to report fair or poor health were the same states in which residents were most likely to distrust others.6 Moving from a state with a wealth of social capital to a state with very little social capital (low trust, low voluntary group membership) increased one’s chances of poor to middling health by roughly 40–70 percent. When the researchers accounted for individual residents’ risk factors, the relationship between social capital and individual health remained. Indeed, the researchers concluded that if one wanted to improve one’s health, moving to a high-social-capital state would do almost as much good as quitting smoking. These authors’ conclusion is complemented by our own analysis. We found a strong positive relationship between a comprehensive index of public health and the Social Capital Index, along with a strong negative correlation between the Social Capital Index and all-cause mortality rates.7 (See table 6 for the measure of public health and health care and figure 86 for the correlations of public health and mortality with social capital.)
Table 6: Which State Has the Best Health and Health Care?
Morgan-Quitno Healthiest State Rankings (1993–1998):
1. Births of low birth weight as a percent of all births ( – )
2. Births to teenage mothers as a percent of live births ( – )
3. Percent of mothers receiving late or no prenatal care ( – )
4. Death rate ( – )
5. Infant mortality rate ( – )
6. Estimated age adjusted death rate by cancer ( – )
7. Death rate by suicide ( – )
8. Percent of population not covered by health insurance ( – )
9. Change in percent of population uninsured ( – )
10. Health care expenditures as percent of gross state product ( – )
11. Per capita personal health expenditures ( – )
12. Estimated rate of new cancer cases ( – )
13. AIDS rate ( – )
14. Sexually transmitted disease rate ( – )
15. Percent of population lacking access to primary care ( – )
16. Percent of adults who are binge drinkers ( – )
17. Percent of adults who smoke ( – )
18. Percent of adults overweight ( – )
19. Days in past month when physical health was “not good” ( – )
20. Community hospitals per 1,000 square miles ( + )
21. Beds in community hospitals per 100,000 population ( + )
22. Percent of children aged 19–35 months fully immunized ( + )
23. Safety belt usage rate ( + )
The state-level findings are suggestive, but far more definitive evidence of the benefits of community cohesion is provided by a wealth of studies that examine individual health as a function of individual social-capital resources. Nowhere is the connection better illustrated than in Roseto, Pennsylvania.8 This small Italian American community has been the subject of near
ly forty years of in-depth study, beginning in the 1950s when medical researchers noticed a happy but puzzling phenomenon. Compared with residents of neighboring towns, Rosetans just didn’t die of heart attacks. Their (age-adjusted) heart attack rate was less than half that of their neighbors; over a seven-year period not a single Roseto resident under forty-seven had died of a heart attack. The researchers looked for the usual explanations: diet, exercise, weight, smoking, genetic predisposition, and so forth. But none of these explanations held the answer—indeed, Rosetans were actually more likely to have some of these risk factors than were people in neighboring towns. The researchers then began to explore Roseto’s social dynamics. The town had been founded in the nineteenth century by people from the same southern Italian village. Through local leadership these immigrants had created a mutual aid society, churches, sports clubs, a labor union, a newspaper, Scout troops, and a park and athletic field. The residents had also developed a tight-knit community where conspicuous displays of wealth were scorned and family values and good behaviors reinforced. Rosetans learned to draw on one another for financial, emotional, and other forms of support. By day they congregated on front porches to watch the comings and goings, and by night they gravitated to local social clubs. In the 1960s the researchers began to suspect that social capital (though they didn’t use the term) was the key to Rosetans’ healthy hearts. And the researchers worried that as socially mobile young people began to reject the tight-knit Italian folkways, the heart attack rate would begin to rise. Sure enough, by the 1980s Roseto’s new generation of adults had a heart attack rate above that of their neighbors in a nearby and demographically similar town.
The Roseto story is a particularly vivid and compelling one, but numerous other studies have supported the medical researchers’ intuition that social cohesion matters, not just in preventing premature death, but also in preventing disease and speeding recovery. For example, a long-term study in California found that people with the fewest social ties have the highest risk of dying from heart disease, circulatory problems, and cancer (in women), even after accounting for individual health status, socioeconomic factors, and use of preventive health care.9 Other studies have linked lower death rates with membership in voluntary groups and engagement in cultural activities;10 church attendance;11 phone calls and visits with friends and relatives;12 and general sociability such as holding parties at home, attending union meetings, visiting friends, participating in organized sports, or being members of highly cohesive military units.13 The connection with social capital persisted even when the studies examined other factors that might influence mortality, such as social class, race, gender, smoking and drinking, obesity, lack of exercise, and (significantly) health problems. In other words, it is not simply that healthy, health-conscious, privileged people (who might happen also to be more socially engaged) tend to live longer. The broad range of illnesses shown to be affected by social support and the fact that the link is even tighter with death than with sickness tend to suggest that the effect operates at a quite fundamental level of general bodily resistance. What these studies tell us is that social engagement actually has an independent influence on how long we live.
Figure 86: Health Is Better in High-Social-Capital States
Social networks help you stay healthy. The finding by a team of researchers at Carnegie Mellon University that people with more diverse social ties get fewer colds is by no means unique.14 For example, stroke victims who had strong support networks functioned better after the stroke, and recovered more physical capacities, than did stroke victims with thin social networks.15 Older people who are involved with clubs, volunteer work, and local politics consider themselves to be in better general health than do uninvolved people, even after accounting for socioeconomic status, demographics, level of medical care use, and years of retirement.16
The bottom line from this multitude of studies: As a rough rule of thumb, if you belong to no groups but decide to join one, you cut your risk of dying over the next year in half. If you smoke and belong to no groups, it’s a toss-up statistically whether you should stop smoking or start joining. These findings are in some ways heartening: it’s easier to join a group than to lose weight, exercise regularly, or quit smoking.
But the findings are sobering, too. As we saw in section II, there has been a general decline in social participation over the past twenty-five years. Figure 87 shows that this same period witnessed a significant decline in self-reported health, despite tremendous gains in medical diagnosis and treatment. Of course, by many objective measures, including life expectancy, Americans are healthier than ever before, but these self-reports indicate that we are feeling worse.17 These self-reports are in turn closely linked to social connectedness, in the sense that it is precisely less connected Americans who are feeling worse. These facts alone do not prove that we are suffering physically from our growing disconnectedness, but taken in conjunction with the more systematic evidence of the health effects of social capital, this evidence is another link in the argument that the erosion of social capital has measurable ill effects.
We observed in chapter 14 the remarkable coincidence that during the same years that social connectedness has been declining, depression and even suicide have been increasing. We also noted that this coincidence has deep generational roots, in the sense that the generations most disconnected socially also suffer most from what some public health experts call “Agent Blue.” In any given year 10 percent of Americans now suffer from major depression, and depression imposes the fourth largest total burden of any disease on Americans overall. Much research has shown that social connections inhibit depression. Low levels of social support directly predict depression, even controlling for other risk factors, and high levels of social support lessen the severity of symptoms and speed recovery. Social support buffers us from the stresses of daily life. Face-to-face ties seem to be more therapeutic than ties that are geographically distant. In short, even within the single domain of depression, we pay a very high price for our slackening social connectedness.18
Figure 87: Americans Don’t Feel As Healthy As We Used To
Countless studies document the link between society and psyche: people who have close friends and confidants, friendly neighbors, and supportive coworkers are less likely to experience sadness, loneliness, low self-esteem, and problems with eating and sleeping. Married people are consistently happier than people who are unattached, all else being equal. These findings will hardly surprise most Americans, for in study after study people themselves report that good relationships with family members, friends, or romantic part-ners—far more than money or fame—are prerequisites for their happiness.19 The single most common finding from a half century’s research on the correlates of life satisfaction, not only in the United States but around the world, is that happiness is best predicted by the breadth and depth of one’s social connections.20
We can see how social capital ranks as a producer of warm, fuzzy feelings by examining a number of questions from the DDB Needham Life Style survey archives:
“I wish I could leave my present life and do something entirely different.”
“I am very satisfied with the way things are going in my life these days.”
“If I had my life to live over, I would sure do things differently.”
“I am much happier now than I ever was before.”
Responses to these items are strongly intercorrelated, so I combined them into a single index of happiness with life. Happiness in this sense is correlated with material well-being. Generally speaking, as one rises up the income hierarchy, life contentment increases. So money can buy happiness after all. But not as much as marriage. Controlling for education, age, gender, marital status, income, and civic engagement, the marginal “effect” of marriage on life contentment is equivalent to moving roughly seventy percentiles up the income hierarchy—say, from the fifteenth percentile to the eighty-fifth percentile.21 In round numbers, getting married is th
e “happiness equivalent” of quadrupling your annual income.22
What about education and contentment? Education has important indirect links to happiness through increased earning power, but controlling for income (as well as age, gender, and the rest), what is the marginal correlation of education itself with life satisfaction? In round numbers the answer is that four additional years of education—attending college, for example—is the “happiness equivalent” of roughly doubling your annual income.
Having assessed in rough-and-ready terms the correlations of financial capital (income), human capital (education), and one form of social capital (marriage) with life contentment, we can now ask equivalent questions about the correlations between happiness and various forms of social interaction. Let us ask about regular club members (those who attend monthly), regular volunteers (those who do so monthly), people who entertain regularly at home (say, monthly), and regular (say, biweekly) churchgoers. The differences are astonishingly large. Regular club attendance, volunteering, entertaining, or church attendance is the happiness equivalent of getting a college degree or more than doubling your income. Civic connections rival marriage and affluence as predictors of life happiness.23
If monthly club meetings are good, are daily club meetings thirty times better? The answer is no. Figure 88 shows what economists might call the “declining marginal productivity” of social interaction with respect to happiness. The biggest happiness returns to volunteering, clubgoing, and entertaining at home appear to come between “never” and “once a month.” There is very little gain in happiness after about one club meeting (or party or volunteer effort) every three weeks. After fortnightly encounters, the marginal correlation of additional social interaction with happiness is actually negative—another finding that is consistent with common experience! Churchgoing, on the other hand, is somewhat different, in that at least up through weekly attendance, the more the merrier.
BOWLING ALONE Page 40