LONGEVITY: Social Factors

Kathryn Aegis (aegis@igc.apc.org)
Mon, 17 Aug 1998 17:11:59 -0700 (PDT)

By James Lardner
Washington Post
Sunday, August 16, 1998; Page C01 (Outlook Section)

Americans are famous for not getting too excited about differences in income or wealth, but we have outdone ourselves lately. Since the 1970s, virtually all our income gains have gone to the highest-earning 20 percent of our households, producing inequality greater than at any time since the 1930s, and greater than in any of the world's other rich nations (and many of its poor ones) today. Bill Gates alone is wealthier than half the American people put together, and from the halls of the Federal Reserve Board on down to your corner tavern, economic pundits treat the issue as a regrettable footnote in the glowing story of world-beating job creation, soaring corporate profits and all-but-invisible inflation.

Now comes a new point to consider. Arriving at the question from a variety of angles and disciplines, researchers in the United States, Canada and Britain have found evidence that more unequal societies are more unhealthy--not just in some highfalutin moral sense but in the plain old medical sense, and not just for the poor (as anyone would suspect) but for the bulk of the population. To put it more baldly, if you live in a place where differences in income and wealth are unusually large (the United States, for example), your chances of escaping chronic illness and reaching a ripe old age are significantly worse than if you live in a place where differences are
not as large (Sweden, for example).

In the scheme of risk factors, moreover, we are talking about one as dangerous as cigarettes, fatty foods or any of the other health habits that we periodically work ourselves into tizzies over.

An inkling of this came, two decades ago, from a study of 17,000 British civil servants. Surprisingly, the annual heart-attack fatality rate among clerks and messengers was four times what it was among administrators. (So much for the idea of executive stress as a big contributor to heart disease.) Even at the bottom of the ladder, these were office workers with the means to feed and house themselves decently and with access to high-quality national health care. In any case, differences in rank proved to be just as eerily accurate a guide to heart-attack rates at the top of the ladder (a senior assistant statistician was twice as vulnerable as a chief statistician), and a similar pattern held for cancer and other diseases that might have been assumed to choose their victims more randomly.

After weeding out such obvious explanations as class-related differences in smoking and diet, Michael Marmot, an epidemiologist at the University of London, began to suspect that intangible issues such as job control and sense of security--the sheer experience of rank, in other words--might be involved. Looking down from a higher statistical altitude, Richard Wilkinson, an economic historian at Sussex University, had a similar insight. Wilkinson's focus was the health differences among, rather than within, nations; here, standard of living--such a sure-fire yardstick of individual health in the Whitehall study--became, oddly, almost a nonfactor.

As long as per-capita gross domestic product exceeded about $5,000, Wilkinson found, Nation A (the United States, for example) could have twice the per-capita income of Nation B (Greece, for example) and still lag in life expectancy. By contrast, Wilkinson's data suggested that equality of income was a very reliable predictor of health. This was an idea that, like the key piece in a jigsaw puzzle, simply looked right the instant it fell in place. In fact, it seemed to solve a number of puzzles at once:

As interesting as these comparisons are, though, some of the strongest evidence involves U.S. states, cities and counties, which offer the advantages of comparable data and a largely shared culture. In the July issue of the American Journal of Public Health, a team of researchers led by John Lynch and George Kaplan of the University of Michigan describe a study of 282 metropolitan areas, finding, as Wilkinson did, that mortality rates are considerably more closely linked to relative than to absolute income. What do Biloxi, Miss., Las Cruces, N.M., and Steubenville, Ohio, have in common? High inequality, high mortality. Allentown, Pa., Pittsfield, Mass., and Milwaukee, Wis.? Low inequality, low mortality.

Ichiro Kawachi and Bruce Kennedy of the Harvard School of Public Health have used domestic data to explore the thesis that much of inequality's influence on health may be due to an erosion of trust or, to use a term coined by the sociologist James Coleman, "social capital." Kawachi and Kennedy have come up with ways of measuring these variables, which turn out to be closely related to inequality and, at the same time, to a city's or state's death rate from a wide range of causes.

They also point to the Brigadoon-like story of Roseto, Pa., an Italian American town whose inhabitants smoked heavily and cooked with lard but nonetheless had an unusually low incidence of heart attack. Alighting on Roseto in the 1950s, the physician-sociologist team of Stewart Wolf and J.G. Bruhn traced its good health to a conspicuously old-fashioned immigrant culture and closeness. By the '60s, however, they began to see younger Rosetans adopting the more individualistic ways of the wider world--for example, building terraces in back of their houses rather than porches in front. Within another decade, Roseto's heart-attack fatality rate was as high as that of neighboring towns.

As bizarre as some of these ideas may sound in the context of modern medicine, they sit well with an array of recent discoveries about how psychosocial forces affect physical health.

Social isolation or lack of friends, for example, accounted in one study of a group of coronary patients for a threefold difference in survival rate after three years; in another study, for a fourfold difference in the rate at which otherwise healthy research subjects got sick after being exposed to cold viruses.

Nor are the biological pathways that could explain such things beyond comprehension. Working with wild baboons in the Serengeti, Robert M. Sapolsky, a Stanford University neurobiologist, has found that animals lower down in the tribal hierarchy tend to resort habitually to the kind of hormonal mobilization that, in higher-status animals, is reserved for emergencies. Over time, escalated levels of cortisol and other stress-related hormones become the norm; the hippocampus gland (important to learning and memory) shrinks, and disproportionately many animals succumb to cancer, brain damage or stroke.

In a laboratory setting, Carol Shively, a psychologist at Wake Forest University, has taken Sapolsky's work a scary step farther, manipulating the status of macaques by bumping them from one social group to another. The results, for monkeys who go from dominant to subordinate, include an accelerated pulse and the buildup of fat in the arteries.

For all the questions that remain to be sorted out (assuming they ever are), it is hard not to see in this body of research a cellular-level glimpse of truths that have lain in the realm of intuition and parable: the idea, for example, that when a society creates steep discrepancies in income and wealth, it excites a preoccupation with material pleasures, money and status, and aggravates feelings of anxiety and inferiority that (it is all too accurate to say) eat away at people. These feelings "are so fundamental," Wilkinson observes, that "it is reasonable to wonder whether the effects on the quality of life are not more important than the effects on the length of life."

Wilkinson's emphasis on the psychosocial dimension of inequality has been a point of disagreement between him and others in the field who argue that material considerations and access to social services may, in fact, play a large role, especially when, as Clyde Hertzman of the Canadian Institute for Advanced Research observes, one includes the latent effects of income-related differences in prenatal care and parenting.

Whatever the balance of factors, however, it ought to be clear that these problems won't be solved with a little 1960s-style income redistribution. They involve the ways in which we distribute money (and power and status) in the first place, and, that being so, call for a basic change of thinking on the part of society, corporate chieftains included. And trailing just a millisecond behind that thought is this one: sure, in a million years.

But consider, for a moment, the slender basis for much of what we currently accept as economic gospel. Communism collapses, and the most timid forms of social planning are seen as a first step down the road to the gulag. Western Europe falters, and we are ready to throw a century's worth of social welfare measures in the trash. The American economy goes into high gear for a few years, and the idea of corporate executives who earn hundreds of times as much as their
employees becomes a law of nature. Indeed, the confidence with which eternal verities are thrown around by experts, and accepted by masses of people, in the capital- and market-driven part of the world these days recalls nothing so much as the Soviet Union a generation or two ago. This dogma too shall pass, and we will all be better off.

James Lardner writes about business and technology for U.S. News and World Report.

Copyright 1998 The Washington Post Company