W

22

Unequal America
Causes and consequences of the wideand growinggap between rich and poor

hen m ajid ezzati thin ks about declining life expectancy, he says, I think of an epidemic like HIV, or I think of the collapse of a social system, like in the former Soviet Union. But such a decline is happening right now in some parts of the United States. By ELIZABETH GUDRAIS Between 1983 and 1999, mens life expectancy decreased in more than 50 U.S. counties, accordshould instead pull themselves up by their bootstraps. The study ing to a recent study by Ezzati, associate professor of internaof inequality attempts to test inequalitys eects on society, and tional health at the Harvard School of Public Health (HSPH), it is delivering ndings that command both sides attention. and colleagues. For women, the news was even worse: life exEzzatis results are one example. There is also evidence that pectancy decreased in more than 900 countiesmore than a living in a society with wide disparitiesin health, in wealth, in quarter of the total. This means 4 percent of American men and educationis worse for all the societys members, even the well 19 percent of American women can expect their lives to be o. Life-expectancy statistics hint at this. People at the top of shorter than or, at best, the same length as those of people in the U.S. income spectrum live a very long time, says Cabot protheir home counties two decades ago. fessor of public policy and epidemiology Lisa Berkman, but peoThe United States no longer boasts anywhere near the worlds ple at the top in some other countries live a lot longer. longest life expectancy. It doesnt even make the top 40. In this Much is still unknown in this dynamic eld, where Harvard is and many other ways, the richest nation on earth is not the home to pioneers who rst recognized income inequality as worhealthiest. Ezzatis nding is unsettling on its face, but scholars thy of study and younger scholars at the forefront of its study nd further cause for concern in the pattern of health disparities. today. The variety of disciplines featured in presentations of the Poor health is not distributed evenly across the population, but Universitys Multidisciplinary Program on Inequality and Social concentrated among the disadvantaged. Policyeconomics, sociology, political science, public policy, Disparities in health tend to fall along income lines everyhealth, medicine, education, law, and businesshighlights the where: the poor generally get sicker and die sooner than the rich. elds broad importance. But in the United States, the gap between the rich and the poor Because of the subjects complexity and the scarcity of consisis far wider than in most other developed democracies, and it is tent data that would allow comparison between countries and getting wider. That is true both before and after taxes: the across wide timespans, research ndings are often highly specic United States also does less than most other rich democracies to or framed in the language of interesting coincidences, rather than redistribute income from the rich to the poor. as denitive conclusions. Even when discernable patterns exist, Americans, on average, have a higher tolerance for income inthere tend to be counter-examples; for instance, the United States, equality than their European counterparts. American attitudes with high inequality, has low life expectancy compared to Denfocus on equality of opportunity, while Europeans tend to see mark and Finland, with very low inequalitybut in Spain and fairness in equal outcomes. Among Americans, dierences of Italy, with inequality somewhere in between, life expectancy is opinion about inequality can easily degenerate into partisan diseven longer. putes over whether poor people deserve help and sympathy or
July - August 200 8

But the coincidences are intriguing indeed. Research indicates that high inequality reverberates through societies on multiple levels, correlating with, if not causing, more crime, less happiness, poorer mental and physical health, less racial harmony, and less civic and political participation. Tax policy and social-welfare programs, then, take on importance far beyond determining how much income people hold onto. The level of inequality we allow represents our answer to a very important question, says Nancy Krieger, professor of society, human development, and health at HSPH: What kind of society do we want to live in?

KEEPING UP WITH THE JONESES

The united states is becoming even more unequal as income becomes more concentrated among the most auent Americans. InLawrence F. Katz come inequality has been rising since the late 1970s, and now rests at a level not seen since the Gilded Ageroughly 1870 to 1900, a period in U.S. history dened by the contrast between the excesses of the super-rich and the squalor of the poor. Early in the twentieth century, the share of total national income drawn by the top 1 percent of U.S. earners hovered around 18 percent. That share hit an all-time high in 1928when top earners took home 21.1 percent of all income, including capital gainsthen dropped steadily through the next three decades. Amid the post-World War II boom in higher education, and overall economic growth, the American middle class swelled and prospered, and the top 1 percent of earners took home less than 10 percent of all income through the 1960s and 1970s. Since then, the topmost 1 percent have seen their share rise again: it shot past 15 percent in 1996 and crested at 20.3 percent in 2006, the most recent year for which numbers are available. To describe the distribution of income inequality in the United States, Allison professor of economics Lawrence F. Katz likes to use the analogy of an apartment building. Over the last 25 years, he says, the penthouse has gotten really, really nice. All sorts of new gadgets have been put in. The units just below the penthouse have also improved a lot. The units in the middle have stayed about the same. The basement apartment used to be OK, but now its gotten infested with cockroaches and its been ooding. (See graph, page 26.) The argument that none of this matters as long as the overall economy is growingthat a rising tide lifts all boats, as President John F. Kennedy famously saidis the subject of vigorous academic review, with mixed results, but it may not be the most important question. Picture a buoyant luxury cruise ship surrounded by dilapidated dinghies, full of holes and on the verge of
FRED FIELD

sinking. The fact that the tide has lifted them does not mean they are doing well. This is a concept social scientists call relative deprivation. The idea is that, even when we have enough money to cover basic needs, it may harm us psychologically to see that other people have more. When British economist Peter Townsend developed his relative deprivation index in 1979, the concept was not new. Seneca wrote that to be poor in the midst of riches is the worst of poverties; Karl Marx wrote, A house may be large or small; as long as the neighboring houses are likewise small, it satises all social requirement for a residence. But let there arise next to the little house a palace, and the little house shrinks to a hut. Investigating whether relative deprivation and the negative emotions it engenders help explain why the poor have worse health than the rich in most societies began with epidemiologist Michael Marmots study of British civil servants in the 1960s and 1970s. Marmot found that the lower-ranking bureaucrats had elevated levels of stress hormones compared to their high-status coworkers, even though the low-ranking workers still had job security, a living wage, decent hours, and benets. Others have found similar links. Examining health outcomes for identical twins raised togetherpairs that shared genes and environmentNancy Krieger found that when the twins became adults, if one was working class and the other professional, the working-class twins health was, on average, worse. There is little question that it is bad for ones health to be poor. Americans at the 95th income percentile or higher can expect to live nine years longer than those at the 10th percentile or lower. The poor are more likely to develop illnesses such as diabetes, hypertension, heart disease, and cancer, and there is evidence that relaHarvard Magazine 23

tive deprivation and the stress it engenders are involved. When high inequality and rising top incomes shift societys accepted standards of living upward, it seems that people experience deprivation even when they have adequate food, clothing, and shelter. The ocial U.S. poverty rate12.3 percent in 2006is relatively low, but scholars agree that number is essentially meaningless. The poverty threshold was developed in 1965 based on the cost of a grocery budget for temporary or emergency use when funds are low, multiplied by three. It was arbitrary, says Wiener professor of social policy Christopher Jencks, but once it was adopted, it was politically impossible to change it. That threshold has been adjusted for ination, but does not take into account the fact that housing prices, energy prices, and certain other costs have grown faster than the consumer price index (CPI). Going to movies, eating out at restaurants, going on occasional vacations, having Internet access and a cell phone none of these things are in the federal poverty level, says Ichiro Kawachi, professor of social epidemiology at HSPH and associate professor of medicine at Harvard Medical School (HMS). What matters for functioning in society is what the average person is able to do. During the same period, the Gallup Poll denition of the poverty linebased on asking people how much income they need not to feel deprivedhas risen much more steeply than the CPI. Kawachi, who grew up in Japan, believes a predominant consumption culture in the United States exacerbates relative deprivation. The Japanese have a very strong culture against conspicuous displays of auence, he says. When I was a child growing up in suburban Tokyo, it was very dicult to distinguish, by dress or anything else, rich kids from poor kidswhereas in America, bring it on!

As further evidence of a correlation between inequality and consumption culture, he points to national spending on advertising as a percentage of gross domestic product (GDP). The top-ranked countries on this measure, according to United Nations (UN) data, are Colombia, Brazil, and Venezuelacountries with inequality levels among the highest in the worldbut also Australia, New Zealand, the United Kingdom (U.K.), and the United States, countries with higher inequality than similarly prosperous peers. Japan comes second only to Denmark in terms of equal-income distribution among its inhabitants, according to United Nations data. And life expectancy at birth for the Japanese is 82.3 years, compared to Americans 77.9 years, even though per-capita GDP in the United States is about $10,000 more than in Japan. Its pretty clear that an egalitarian ethos runs along with the idea of having strong safety nets and protecting the health of the most vulnerable, says Kawachi, who also directs HSPHs Center for Society and Health. And thats reected in national health statistics. The United States ranks twenty-rst among the 30 nations in the Organization for Economic Cooperation and Development (OECD) in terms of life expectancy, and twenty-fth in terms of infant mortality. Kawachi and others have found that the U.S. counties with the most income inequality stack up poorly on health measures, and as mortality rates have fallen nationwide, they have fallen most slowly in states where income inequality increased the mosta cause for concern, whatever the explanation.

AMERICAN EXCEPTION?
One widely used measure of inequality is the Gini coecient, named for Italian statistician Corrado Gini, who rst articulated the concept in 1912. The coecient measures income distribution on a scale from zero (where income is perfectly equally distributed among all members of a society) to one (where a single person possesses all the income). For the United States, the Gini coecient has risen from .35 in 1965 to .44 today. On the per-capita GDP scale, our neighbors are Sweden, Switzerland, and the U.K.; on the Gini scale, our neighbors include Sri Lanka, Mali, and Russia. (Even with this basic measure of inequality, it is dicult to get comparable data for all countries, and some other sources nd a much wider gap between the United States and Russia. For instance, the Luxembourg Income Study ranks Russia at .43 and the United States at .37, and does not even list Sri Lanka and Mali.) The recent increase in inequality reects a migration of money upward as salaries have ballooned at the top. In 1965, the average salary for a CEO of a major U.S. company was 25 times the salary of the average worker. Today, the average CEOs pay is more than 250 times the average workers. At the same time, the government is doing less to redistribute income than it has at times in the past. The current top marginal tax rate35 percentis not the lowest its beenthere was no federal income tax at all until 1913but it is far lower than the 91-percent tax levied on top earners from 1951 to 1963. Meanwhile, forces such as immigration and trade policy have put pressure on wages at the bottom. Tax policies and employer-pay practices aect income distribution directly. But what governs these pay practices, and why have American voters and politicians chosen the tax policies they have? One answer lies in Americans unique attitudes toward inequality.

Lisa Berkman

SUZANNE CAMARATA/HARVARD SCHOOL OF PUBLIC HEALTH

Quantifying Inequality
.743 The Gini coefficient measures the distribution of income on a scale from zero (where income is perfectly equally distributed among all members of a society) to one (where a single person possesses all the income). .332 .337 .346 .360 United Kingdom .399 .401 .402 .408 .586 .592

.570 .461 .482

United States

.250 Sweden

.297

.300

Switerland

Venezuela

Colombia

Sri Lanka

Moldova

Source: United Nations Human Development Report, 2007/08

Asked by the International Social Survey Programme whether they agreed or disagreed with the statement that income dierences in their home country are too large, 62 percent of Americans agreed; the median response for all 43 countries surveyed some with a much lower degree of inequalitywas 85 percent. Americans and Europeans also tend to disagree about the causes of poverty. In a dierent surveythe World Values Survey, including 40 countriesAmerican respondents were much more likely than European respondents (71 percent versus 40

left-leaning groups have succeeded at writing in change. By and large, Alesina and Glaeser write, the U.S. Constitution is still the same document approved by a minority of wealthy white men in 1776. And the vestiges of feudalism in European society make leftist arguments appealing there, whereas American politicians rhetoric has emphasized individual agency since the time of George Washington (who wrote in 1783 that if citizens should not be completely free and happy, the fault will be intirely their own). The authors cite a 1980s history curriculum for public

Americans are much more likely than Europeans to agree with the statement that the poor could escape poverty if they worked hard enough.
percent) to agree with the statement that the poor could escape poverty if they worked hard enough. Conversely, 54 percent of European respondents, but only 30 percent of American respondents, agreed with the statement that luck determines income. It makes intuitive sense that those who view poverty as a personal failing dont feel compelled to redistribute money from the rich to the poor. Indeed, Ropes professor of political economy Alberto Alesina and Glimp professor of economics Edward L. Glaeser nd a strong link between beliefs and tax policy: they nd that a 10-percent increase in the share of the population that believes luck determines income is associated with a 3.5percent increase in the share of GDP a given nations government spends on redistribution (see Down and Out in Paris and Boston, January-February 2005, page 14). These attitudes, in turn, are rooted in U.S. history, says Christopher Jencks, whose 1973 book Inequality examined social mobility in the United States. Jencks has been studying inequality and social class since the 1960s, and has written dozens of journal articles, essays, and book chapters, as well as four more books, on the subject. He looks back to the Constitutions framers, who enshrined property rights as sacred and checked the governments ability to control the national economy. The founding fathers didnt want the government to do that much, he says. The Constitution is structured in such a way that it is harder to change than the constitutions of Europes welfare states, where schools in California (hardly the most right-wing of states, they note) that instructed, A course should assess the role of optimism and opportunity in a land of work: the belief that energy, initiative, and inventiveness will continue to provide a promising future. An alternative, and possibly complementary, explanation points to the United Statess particular place in geography and history. Jencks also nds this persuasive. The highest levels of inequality are found in the New World and not the Old, for reasons we dont understand, he says (see chart above). Societies with higher inequality also tend to have higher crime rates, although its not clear which way the causal arrow runs, or if it exists. These are societies built on conquest, many of them on slavery, Jencks adds. A lot of the inequality may just be the legacy of those things. Former colonies such as Haiti and Namibia inhabit the top end of the Gini scale, with coecients of .59 and .74, respectively. But there are exceptions to the pattern: the low end of the scale includes transitional economies that are far from rich (Belarus and Moldova, with coecients of .30 and .33), and former colonies (Ethiopia and Laos, with coecients of .30 and .35). For all the scholarly study, consensus on whether the Gini coecient can, in and of itself, say something good or something bad about a country is still lacking. Still, scholars are using what evidence does exist to ask, and test, whether the United States has things in common with Sri Lanka, Mali, and Russia, as it
Harvard Magazine 25

Namibia

Ethiopia

Mexico

Belarus

Russia

Brazil

Haiti

Laos

Mali

undoubtedly does with Sweden, Switzerland, and the U.K. The excesses of the Gilded Age led, in the decades that followed, to a backlash in the form of the minimum wage and other labor laws to protect workers, business and nancial-market regulation to protect consumers, social safety-net programs Social Security, Medicare, Medicaidand infrastructure investment to benet all. But as the United States moves from a period of relatively balanced income distribution back into higher in-

not necessarily the tactics that lift team performanceas opposed to, say, practicing great defense. This gets at the ways inequality may aect the fabric of society. Perhaps motivated by inequality and the prospect of getting ahead, Americans work longer hours than their European counterpartsabout 200 more hours per year, on average, than the British, and 400 more hours per year than the Swedes. Again, there are counter-examples (the Japanese work almost as much

Inequality may act on the human psyche to elicit hard work and high achievementbut it also may make us more individualistic.
equality, it remains to be seen whether these twentieth-century developments will enable the country to escape the problems that often accompany high inequality. as Americans do, just 50 hours less a year), but in any case, time spent at work is time not spent with friends or family, and this has its own implications for health. As an outreach worker in San Francisco in the 1970s, Lisa Berkman noticed that her clients in the North Beach and Chinatown neighborhoodspoor or working-class, but with the strong social connections typical of immigrant communities had far better health than her clients in the gritty Tenderloin district, who were much more socially isolated and disconnected from one another. The link between social integration and mortality risk became the subject of Berkmans dissertation at Berkeley, where she earned her Ph.D. in 1977. At the time, the idea that social ties could protect health was radical. Now it is accepted wisdomand a factor that, Berkman believes, helps to explain the extraordinarily high life expectancy in Spain and Italy. But the danger of disconnectedness may go beyond being less happy or even less healthy. Kawachi and Kennedy cited a wealth of evidence that increasing income inequality goes hand in hand with a decrease in social capital, a concept akin to community involvement that incorporates, among other things, social relationships, trust, reciprocity among friends and neighbors, and civic engagement. (Malkin professor of public policy Robert Putnam made a similar argument in his seminal 2000 book Bowling Alone.) Letting social capital atrophy means a less cohesive

LEFT OUT AT THE BOTTOM

An argument commonly made in inequalitys defense is that it
serves to motivate. Here, Kawachi cites evidence from the sports world. A 1990 study of golfers found that they performed best in professional tournaments, where the spread in the size of the prize money is widest. Similarly, a study of professional auto racers found that performance improved as the spread in the size of the various prizes widened. So inequality may act on the human psyche to elicit hard work and high achievementbut it also may make us more individualistic. In a study of baseball players, teams with wider pay dispersion performed more poorlyand so did individual players within those teams. In a world in which each individual is looking out for themselves, players will tend to concentrate on improving their own performance to the exclusion of team goals, since their own performance is what matters for moving up the pay scale, Kawachi and Bruce P. Kennedy (a former HSPH professor who passed away this year) wrote in The Health of Nations: Why Inequality Is Harmful to Your Health. Concentrating on trying to hit more home runs or improving ones own hitting average are

Growing Apart
.475 0.45 .425

0.4
.375 0.35 .325

Indexed Real Income, 1973=100

The growth in the Gini coefficient for U.S. family income indicates increasing inequality in recent decades.

150

Trends in family income show those at the top pulling away from those at the middle and the bottom.

100 95th percentile 50th percentile 50 20th percentile

1945

1955

1965

1975

1985

1995

2005

1945

1955

1965

1975

1985

1995

2005

Source: The Race between Education and Technology, by Lawrence F. Katz and Claudia Goldin (Harvard University Press, 2008)

26

July - August 200 8

populace that, at the extreme, leaves entire classes of people disadvantaged and excluded. The big worry, says Lawrence Katz, is creating something like a caste society. As American neighborhoods have become more integrated along racial lines, they have become more segregated along income lines and, some research indicates, with regard to all manner of other factors, including political and religious beliefs. (The Big Sort, a new book by journalist Bill Bishop, examines this evidence.) Whats more, even along racial lines, American society is still far from integrated. Sociologist David R. Williams, Norman professor of public health and Ichiro Kawachi professor of African and African American studies, has examined racial discrimination and health in the United States and elsewhere, including South Africa, where in 1991, under apartheid, the segregation index was 90, meaning that 90 percent of blacks would have had to move to make the distribution even. In the year 2000, says Williams, in most of Americas larger citiesNew York City, Detroit, Chicago, Milwaukeethe segregation index was over 80. Only slightly lower, that is, than under legally sanctioned apartheid. When a society is starkly divided along racial or ethnic lines, the auent are less likely to take care of the poor, Glaeser and Alesina have found. Internationally, welfare systems are least generous in countries that are the most ethnically heterogeneous. Those U.S. states with the largest black populations have the least generous welfare systems. And in a nationwide study of peoples preferences for redistribution, Erzo F.P. Luttmer, associate professor of public policy at the Harvard Kennedy School (HKS), found strong evidence for racial loyalty: people who lived near poor people of the same race were likely to support redistribution, and people who lived near poor people of a dierent race were less likely to do so. Dierences in skin color seem to encourage the wealthy to view the poor as fundamentally dierent, serving as a visual cue against thinking, There but for the grace of God go I. Alesinas work investigates this cognitive process as an explanation for the high crime rates in less equal societies. Rather than following the common-sense explanation that the poor see what the rich have and covet it, leading to burglary and violent crime, Alesina argues that as the incomes of the rich and poor diverge, so do their interests. Members of a relatively equal society nd it relatively easy to reach agreement about what the purpose and priorities of a legal system should be. But if the rich favor protecting property, while the poor care more about preventing and
FRED FIELD

punishing interpersonal violent crime, the lack of consensus will produce a weak system that fails to meet the desires of either group. In one essay, his colleague Glaeser oers this apocalyptic prediction: Great gaps between rich and poor mayhurt democracy and rule of law if elites prefer dictators who will protect their interests, or if the disadvantaged turn to a dictator who promises to ignore property rights. This doesnt seem possible in a democracy such as the United States, where each citizens vote carries the same weight regardless of income (the electoral-college system notwithstanding). In fact, given the shape of the income distribution, it seems that Americans would elect leaders whose policies favor the poor and middle class. Mean household income in 2004 was $60,528, but median household income was only $43,389. More than half of households make less money than average, so, broadly speaking, more than half of voters should favor policies that redistribute income from the top down. Instead, though, nationsand individual stateswith high inequality levels tend to favor policies that allow the auent to hang onto their money. Filipe R. Campante, an assistant professor of public policy at HKS and a former student of Alesinas, thinks hes discovered why. After investigating what drives candidates platforms and policy decisions, Campante has concluded that donations are at least as inuential a mode of political participation as votes are. Previous research has shown that voter turnout is low, particularly at the low end of the income spectrum, in societies with high inequality. Again, this is counterintuitive: in unequal places, poor people unhappy with government policies might be expected to turn out en masse to vote, but instead they stay home. Campaign contributions may provide the missing link. Candidates, naturally, target voters with money because they
Harvard Magazine 27

David R. Williams

need funds for their campaigns. And since the poor gravitate toward parties that favor redistribution and the wealthy align themselves with parties that do not, campaign contributions end up beneting primarily parties and candidates whose platforms do not include redistribution. By the time the election comes around, the only candidates left in the race are those whove shaped their platforms to maximize fundraising; poor voters, says Campante, have already been left out. In a study of campaign contributions in the 2000 U.S. presidential election, he found that higher income inequality at the county level was associated with fewer people contributing to campaigns, but contributing a larger amount on averageso the haves participated, and the have-nots did not. The solution, he says, is not to scrap the system altogether in favor of full public nancing, but to enact contribution limits strict enough to level the playing eld. He views contributions not as bribery or buying policy, but as a legitimate form of civic engagement. The ideal system, he says, would be a system where you have a really broad base of contributors that are contributing relatively small amounts.You want parties to be responsive to voters. Donations are a way in which parties are made responsive to voters.

BUFFERS AGAINST INEQUALITY

The effects of relative deprivation can come in a form more tangible than stress or low self-esteem. Krieger uses the example of a job interview. In a society where the average person has a cell phone, it can hurt ones job chances not to have one. Wearing old clothes to a job interview might be interpreted as a sign of not
28 July - August 200 8

taking the interview seriously, when in fact the problem is inability to aord a new outt. Bad teeth, which require money to x, can trigger disgust in prospective employers and even hold people back from making friends. Your income, Krieger says, can decline to a point where youre no longer able to participate meaningfully in society. Stress can also make people behave in ways they otherwise wouldnt. David Williams believes that the hierarchy of needs framework helps explain why, the poorer people are, the less likely they are to take care of their health. The framework, developed in 1943 by psychologist Abraham Maslow, denes the needs that motivate human behavior and the priority people assign to those needs. Physiological needs (eating, sleeping, breathing) form the foundation; not until those needs are met can people pursue needs in the higher categories (in succession: safety, love/belonging, esteem, and self-actualization). If people are worried about their basic needs of survival and security and food and shelter, says Williams, they cannot worry about the fact that a cigarette, which is providing relief from stress now, is going to cause lung cancer 20 years from now. If you can address the basic needs so people are no longer worried about them, you free them to consider those larger, higher-level needs that have long-term consequences for their well-being. Lisa Berkmans latest project aims to let low-wage workers focus on such higher-order needs. In a study of nursing-home employees, Berkman found that nursing assistants, janitors, and kitchen workers had far less exibility than higher-status workers in terms of being able to leave work if a family member fell ill, and that this lack of exibility was related to increased risk of heart disease and chronic sleep problems. Now she is following nursing homes and retail establishments to see what happens when they implement more exible policies. If workers in highdemand, low-wage jobs can spend more time with their families and stop worrying about getting red if they need to handle an emergency, she says, workplace policies may have a profound eect on health. Improving living conditions in poor neighborhoods is another way to alleviate povertys ill eects even in the absence of income redistribution, says Williams. The poor are more likely to smoke, to eat poorly, and to lead sedentary lives. These are personal choicesbut every choice is made in context, and ones surroundings aect the choices one makes. When people live in areas where there arent supermarkets that sell fresh fruits and vegetables, their intake of fresh fruits and vegetables is dramatically lower, he says. If people live in areas where there arent sidewalks, where there arent safe bike paths and places to walk and playgrounds, or where the rate of crime is so high that its not safe to go outside, then their level of exercise is much lower and their rates of obesity are higher. Building parks and sidewalks and bringing farmers markets to poor neighborhoods, then, makes it easier for residents to make healthy choices. Another category of initiatives aims at improving living conditions for poor people by giving them vouchers to move to better neighborhoods, but the details are important, says Dolores Acevedo-Garcia, an HSPH associate professor of society, human development, and health. She is helping design the publichealth component of one such program. Stemming from a landmark 2005 desegregation court case, it has already enabled about 1,300 former tenants of Baltimore public housing to move

JIM HARRISON

to suburban communities. What people are expecting, she says, is that if people move to a new neighborhood, theyre automatically going to do better. Well, in fact, a lot of this is about connecting people to resources: for example, helping them nd landlords who will rent to themnot the easiest thing in an unfamiliar neighborhood. The aid doesnt stop there. Many doctors in auent communities dont accept Medicaid; Acevedo-Garcias proposal would have case workers help clients nd doctors who do, and in some cases persuade doctors to start. People may be used to doing their shopping at a convenience store or a liquor store, she says; case workers will tell them which grocery store has good produce at low prices, and where to catch the bus that will take them there. Something as simple as taking the new residents to a park can make a dierence, she says: They may not be used to the idea of exercising outside if they came from a neighborhood that was not safe.

UNEQUAL CHANCES
Adults economic status is positively correlated with their parents economic status in every society for which we have data, write Christopher Jencks and Laura Tach, a doctoral student in sociology and social policy, but no democratic society is entirely comfortable with this fact. The prospect of upward mobility forms the very bedrock of the American dream, but analyses indicate that intergenerational mobility is no higher in the

on a hill. But if anything, Alesina and Glaeser write, the American poor seem to be much more trapped than their European counterparts, in the sense that fewer people who start life in the bottom quintile ever make it out. This is puzzling given American societys emphasis on fairness and openness. Lee professor of economics Claudia Goldin and Katz detect an explanation in the increasing cost of college tuition. In 1950, the average tuition price at a private college was roughly 14 percent of the U.S. median family income; public college tuition was even lower (only 4 percent). Percentages for both types of institutions fell further in the ensuing decades, bottoming out around 1980, but then rising steeply ever since. In 2005, the cost of attending the average public college was 11 percent of median family income; for private colleges, the average was 45 percent. There is nancial aid, but not enough, and the system can be harder to crack than Fort Knox, Katz and Goldin write in their new book, The Race between Education and Technology. For most of the twentieth century, the average American exceeded his parents education level by a signicant margin: between 1900 and 1975, the average Americans educational attainment grew by 6.2 years, or about 10 months per decade. Then, between 1975 and 1990, the authors nd that there was almost no increase at all; from 1990 to 2000, there was a gain of just six months. Although college graduation rates for women are still rising steadily, for men they have barely increased since the days of the Vietnam draft.

If people are worried about their basic needs of survival and security and food and shelter, they cannot worry about the fact that a cigarette is going to cause lung cancer 20 years from now.
United States than in other developed democracies. In fact, a recent Brookings Institution report cites ndings that intergenerational mobility is actually signicantly higher in Norway, Finland, and Denmarklow-inequality countries where birth should be destiny if inequality, as some argue, fuels mobility. In the United States, the correlation between parents income and childrens income is higher than chance: 42 percent of children born to parents in the bottom income quintile were still in the bottom quintile as adults, and 39 percent of children born to parents in the top quintile remained in the top quintile as adults, according to the Brookings analysis. But it is dicult to see whether mobility is increasing or decreasing, because it would require comparing specic individuals incomes to their parents incomes, against the wider backdrop of income distribution across society at that time. Because data with that level of detail do not exist for earlier periods, scholars cant say with certainty whether the results represent an increase or a decrease in mobility from other periods in American history. Americans steadfast belief in mobility probably stems from increases in absolute, rather than relative, mobility. As the overall economy mushroomed throughout the nations history, the majority of people exceeded their parents income. Recall Katzs apartment building analogy; rather than tenants moving from one oor to another, the entire building was shifting ever higher At the same time, the college wage premium has also increased. In 1975, the average college graduates hourly wage was 24 percent higher than the average high-school graduates. By 2002, that number had risen to 43 percent. Katz and Goldin say this increase indicates higher demand for workers with college degrees, even as computers have eliminated the type of jobs a high-school-diploma recipient or mediocre college graduate would have done 25 years ago: clerical work, basic accounting, middle management. Technology has exerted downward pressure on those workers pay, explaining stagnating wages at the middle and bottom of the income distribution. The United States once led the world in the rate at which its citizens nished college; it now falls in the middle of the OECD pack. It could lead again if Americans made a decision to fund higher education the way they chose to fund universal public high-school education early in the last century. If you had made people borrow money to go to high school in the early twentieth century, says Katz, you wouldnt have seen the same sort of expansion. But as technology continues to advance, if Americans do not break down barriers to higher education, the authors foresee an even more acute shortage of highly trained workers and, other things being equal, a further increase in inequality. Elizabeth Gudrais 01 is associate editor of this magazine.
Harvard Magazine 29

posttest evaluations, Objective Structured Clinical Exams (OSCE), and videotaped or audio-taped clinical encounters.24 Attempts to rigorously evaluate the effectiveness of these programs will, expectantly, lead to a consensus regarding core concepts as well as a standardized approach to instruction. However, many in the field believe it is also important not to hold cultural competence curricula to unfair evaluation standards. Much of what is standard in undergraduate medical education has no proven link to improved health outcomes.14
Other Sections

Conclusion
This prescription for cultural competence calls for a more active approach, integrated across all levels of medical education. Our recommendations are founded in adult learning theory and the diffusion of innovations model and are concordant with other published guides.24,46,6769 Though it may be more challenging to implement multilevel curricula, achieve support from senior administrators, and tap the hidden curriculum, this approach is more likely to yield long-term results than are isolated workshops. As the field continues to move forward, outcomes-based research efforts will be essential in determining the value of these strategies and others as we seek to improve health care and reduce health disparities.!

8. Develop a Cadre of Dedicated Faculty

To augment the efforts of the physician champion(s), training an additional group of faculty in cultural competence will begin to build an early majority of supporters. This group is critical for the spread of new programs beyond the early adopters.57 As noted above, cultural competence education should not take place only in workshops, and the teaching should not be done by 1 or 2 physician champions alone. In addition, by equipping an early majority of committed faculty physicians with the skills necessary to routinely discuss multicultural issues as part of patient care, teachable moments around culture will be explored more often. The regular discussion of cultural issues during rounds will influence the hidden curriculum, the informal part of medical education delivered through role modeling and other sometimes subconscious activities.61,62

9. Make it a Real Science

Training in cross-cultural communication may be viewed as a softscience, where discussions of explanatory models and empathy contrast against the fact-oriented majority of medical education.24 To satisfy medical trainees' demand for scientific evidence, curricula should emphasize the wealth of research on health disparities, the importance of culture in patient care, and the demonstrated value of cultural competence education.14,28,63,64 A recent systematic review by Beach et al.14 concluded that cultural competence education improves the knowledge, attitudes, and skills of health professionals, as well as patient satisfaction. However, little consistency exists among training programs, and it is difficult to conclude what method and duration of training is most effective. Evaluation studies are often limited to a single institution, with homogenous learner populations and unique evaluation strategies. As the field moves forward, multicenter studies using standardized and validated measures will be critical to more fully assess the impact of training on physician's attitudes, skills, and knowledge, as well as the effect of newly acquired skills on patient outcomes.14,27,65,66 In this respect, it is important that programs in cultural competence include an evaluation component. Several appropriate methods are available to demonstrate the impact of curricula on physician-patient communication and other processes of care, such as pretest and

by serving as role models for minority students who may consider a career in medicine, and medical schools must continue recruiting diverse student bodies.46 Societal efforts are also needed to improve the education of minority students at all stages of schooling. Although minority physicians provide a disproportionate amount of care to underserved populations,5456 it should be noted that they are not necessarily more culturally competent than the majority group. Therefore, while it is important to increase the diversity of medical schools and health care systems, minority physicians should also receive cultural competency training in order to maximize their ability to relate to patients of a different background. Cultural competence training is for everyone regardless of cultural background.

7. Involve an Opinion Leader as the Physician Champion

According to the diffusion of innovations theory, innovation is communicated over time, through particular channels, among the members of a social system.57 This theory predicts that the uptake of cultural competence education will follow an S-shaped curve, spreading in sequence through groups of individuals categorized as innovators, early adopters, early majority, late majority, or laggards.57 Opinion leaders comprise a key segment of the early adopters group. An opinion leader is an educationally influential colleague who models appropriate behaviors.58 The social influences model of behavior change suggests that an opinion leader plays a vital role in the diffusion of the educational program to the larger community, by lending credibility and serving as a physician champion.59,60 In cultural competence education, while it is often a minority faculty member who plays the role of a physician champion, it may actually be more difficult for a single minority instructor to try to convince a majority group of the topic's importance. He or she may not be seen as an opinion leader, but rather as someone with a personal agenda or an advocate for a particular group. In cases where the primary champion of cultural competence education is a minority physician and that individual is not widely seen as an opinion leader, it may be helpful to include additional minority faculty or an influential member of the majority racial/ethnic group. Such a strategy should help appropriately sanction the activity and speed its dissemination.

formal cultural competence training will facilitate complete integration of cultural competence training in medical education. Such a partnership has been formed at some institutions. At Wake Forest School of Medicine, for example, the Dean of Medical Education created a Cultural Competency Theme Team (CCTT), which is composed of individuals who direct curriculum components. The CCTT is charged with integrating culturally relevant activities throughout the 4 years of medical education.27 In order to serve as effective partners and advocates, some top-level administrators and educators themselves may benefit from core cultural competence training, particularly if they attended medical school within the context of a less diverse society. Their personal participation will also deliver the message that cultural competence education is important. The American College of Physicians supports cultural training for health care providers and administrators at all levels.46 If medical schools plan to shape the practices of future physicians within the context of cultural competence, it is of vital importance for the medical profession to reassess traditionally conceived values, beliefs, and biases, which may not be in agreement with current social diversity.

6. Promote Cultural Diversity Among Medical Students and at All Levels of the Medical School Faculty
The diversity of the U.S. population is not reflected in the composition of the student body or faculty at most medical schools, or in the community of practicing physicians. Although minority groups comprise 30% of the U.S. population, only 13% of medical students, 6% of practicing physicians, and 3% of medical school faculty are members of an underrepresented minority group.4750 Minority faculty are less likely to be promoted,51 and less than 2% of senior leaders in health care management are nonwhite.52 Such lack of diversity in the health care workforce and leadership may limit the ability of health care systems to adapt to the changing demands of an increasingly diverse population.50 Racial discordance between physician and patient is also associated with lower patient satisfaction and less participatory clinical interactions.53 Efforts to promote diversity among health care providers should begin early and occur at many levels. Physicians can play an important role

3. Provide Direct Faculty Observation and Feedback

In addition to trainees receiving feedback from standardized patients and their peers, direct observation and feedback from faculty members who have cultural competence training can provide a memorable and useful experience. An individual's culture is shaped by innumerable elements such as education, religion, economic status, immigration history, age, and places traveled.33 Therefore, a teachable moment in cultural competence may exist in virtually any physician-patient interaction.40 For example, a clinic preceptor could provide feedback on the student's ability to perform one of the skills described above, such as eliciting the patient's understanding of the illness etiology.3 Alternatively, a faculty physician could review with the trainee a videotape of the trainee's standardized patient encounter, an actual patient encounter, or a prerecorded trigger tape.4144

4. Discuss Cultural Competence Throughout Clinical Education, Rather Than in Isolated Workshops
Becoming culturally competent is a complex, life-long process. However, most cultural competence education for medical students has a total contact time of less than 1 week,14 a duration that is unlikely to lead to long-term behavior change.45 To reinforce culturally-relevant knowledge and skills, cultural competence training should be infused throughout students' clinical education. There are many opportunities to discuss cultural issues with our learners. Whether treating a patient from another culture or simply one who does not share the Western biomedical view of disease, the discussion during medical rounds should be broad and include the patient's cultural background and its impact on disease and health behavior, in addition to teaching about pathophysiology and management.

5. Get Buy-in From the Top

The Liason Committee on Medical Education requires that medical school faculty and students must demonstrate an understanding of the manner in which people of diverse cultures and belief systems perceive health and illness and respond to various symptoms, diseases, and treatments.19 On these grounds, seeking the support of medical school deans and a commitment from course directors to

behaviors based on culture. This oversimplified practice fails to acknowledge diversity within groups and emphasizes differences between groups, potentially reinforcing stereotyping behavior.14 Instead, cultural competence programs should acknowledge heterogeneity within cultural groups and teach medical trainees how to apply knowledge of socio-cultural issues at the individual level. Even within a family unit, clinicians find different behavioral patterns and health beliefs, based on individual experiences, preferences, and acculturation level.30,31 Skills that are generalizable across patients and cultures include the ability to elicit an individual's perceptions of health and illness, as well as his or her explanatory model and preferences for treatment. Berlin and Fowkes' LEARN guideline provides a framework for this interaction, suggesting that physicians listen to the patient's perception of the problem, explain their own opinion, acknowledge and discuss differences and similarities, recommend treatment, and negotiate an agreement.32 Kleinman's questions for eliciting a patient's explanatory model of illness also serve as a useful guide.3 A third framework, RISK, calls on clinicians to assess a patient's Resources, Identity, Skills, and Knowledge to gain a better understanding of the level of cultural influence on an individual patient's perceptions and behaviors.33

2. Use Interactive Educational Methods, Such as Standardized Patient Encounters, Role-play, and Selfreflective Journal Assignments
In order to effectively teach practical skills, it is important to use interactive educational methods that correspond with principles of adult learning.34 Standardized patient encounters using patient actors can create realistic clinical scenarios in which students and residents may practice new communication skills and receive direct feedback from the trained actor.35 Role-play exercises serve a similar purpose.36 During role-play, the opportunity to provide feedback to a colleague may give trainees more insight into their own behaviors. Finally, narrative writing helps trainees openly reflect on their own values, beliefs, and biases,37,38 and encourages them to consider their personal experiences with prejudice, discrimination, challenging patient encounters, and prior mistakes.39 This exercise could help facilitate attitude change and promote awareness.

classroom lectures, workshops, electives, standardized patient exercises, clinical clerkships, language training, immersion programs, and other interactive exercises.5,25 Most training occurs during the first or second years of medical school, commonly in a case-based or didactic format.5,26 However, schools generally dedicate only a small portion of their curricular time to discussion of cultural competence as it relates to patient care.14 Moreover, little attention is given to cross-cultural issues during students' clinical rotations,5 a time when students have a valuable opportunity to experience, practice, and internalize multicultural communication skills. While one large survey noted that cultural competence education is generally incorporated into larger courses,5 it is unclear to what extent schools have succeeded in integrating cultural competence into the overall curriculum. Cultural competence training appears to primarily occur in the form of an occasional lecture, case study, or workshop.5,25

Evidence of Impact
There is currently no consensus on how cultural competence should be taught in medical school curricula, and therefore considerable variability exists in the design and implementation of cross-cultural education. Emerging evidence shows that many of these varied cultural competence interventions can have an impact on the knowledge, attitudes, and skills of health professionals, as well as on patient satisfaction.14,27 However, there is limited evidence demonstrating that the current models of education lend themselves to positive outcomes and implementation in clinical practice.28 In fact, 20% to 25% of recent medical school graduates feel unprepared to provide specific components of cross-cultural care.29
Other Sections

Prescription for Success in Cultural Competence Education 1. Teach Practical Skills

Cultural competence programs have traditionally followed a knowledge-based approach.24 Such curricula often include lists of preferred words, images, or approaches for treating minority groups, portraying each group as having particular values, beliefs, and

programs.19,20 Their call to action is bolstered by the U.S. Department of Health and Human Services National Standards for Culturally and Linguistically Appropriate Services (CLAS) in Health Care, which provides a framework for implementing culturally and linguistically competent health services.21 Despite the availability of guidelines and model programs,22 not all cultural competence education is effective in improving the attitudes and skills of health professionals.14,23 In this manuscript, we describe current approaches to cultural competence education. Based on features of successful programs and drawing on established educational principles, we propose several elements that may improve cultural competence training in medical education.
Other Sections

Current Approaches to Cultural Competence Education Conceptual Approach

Three major conceptual approaches have emerged for teaching cultural competence, focusing on knowledge, attitudes, and skills, respectively.24Knowledge-based programs (the multicultural/categorical approach) focus on information, such as definitions about culture and related concepts, social determinants of health, and variations in disease incidence and prevalence. These programs may also identify common ethno-medical beliefs and practices thought to influence the patient-physician relationship and medical outcomes. Attitude-based curricula (the cultural sensitivity/awareness approach) seek to improve provider awareness of the impact of socio-cultural factors on patients' values and behaviors and how these factors may ultimately impact clinical outcomes. These curricula often use self-reflection and explore issues of bias, racism, and gender disparities. Skill-building educational programs (the cross-cultural approach) focus on learning communication skills, such as how to elicit the patient's explanatory model of illness and social context, and how to use an interpreter or cultural liaison. These skills are applied to negotiate the patient's participation in decisions and treatment.

Training Format
Different components of cultural competence can be taught through

Cultural competence programs have proliferated in U.S. medical schools in response to increasing national diversity, as well as mandates from accrediting bodies. Although such training programs share common goals of improving physician-patient communication and reducing health disparities, they often differ in their content, emphasis, setting, and duration. Moreover, training in cross-cultural medicine may be absent from students' clinical rotations, when it might be most relevant and memorable. In this article, the authors recommend a number of elements to strengthen cultural competency education in medical schools. This prescription for cultural competence is intended to promote an active and integrated approach to multicultural issues throughout medical school training. Keywords: cultural competency, medical education, disparities
Other Sections

Medical education has witnessed a steady increase in efforts to train physicians to provide high-quality, culturally competent care. Training in cultural competence has risen to the forefront of medical education in part because the United States is becoming increasingly diverse. Ethnic minorities now comprise about 30% of the population, and demographic trends show that they will become the majority by the year 2050.1,2 In addition, greater appreciation exists for the impact of culture on health care and health disparities.3,4 Health seeking behaviors are affected by cultural mores. Some patients may delay seeking care due to perceived cultural insensitivity,5 concern that they will receive a lower quality of care,6 or the perception that they have been treated unfairly because of race or ethnic background.7,8 Furthermore, health disparities have been widely recognized, with racial differences in treatment persisting after adjustment for insurance status, income level, and health status.913 Published research suggests cultural competence may improve physician-patient communication and collaboration, increase patient satisfaction, and enhance adherence, thereby improving clinical outcomes and reducing health disparities.3,4,7,8,1418 The Liaison Committee for Medical Education and the Accreditation Council for Graduate Medical Education emphasize the need for training in cultural competence in medical schools and other post-graduate

 RACIAL/ETHNIC BIAS AND HEALTH

21. Landrine H, Klonoff E. The Schedule of Racist Events: a measure of racial discrimination and a study of its negative physical and mental health consequences. J Black Psychol. 1996;22: 144168. 22. Wiggins JS. Personality and Prediction: Principles of Personality Assessment. Boston, Mass: Addison-Wesley Publishing Co; 1973. 23. Armstead CA, Lawler KA, Gorden G, Cross J, Gibbons J. Relationship of racial stressors to blood pressure and anger expression in black college students. Health Psychol. 1989;8:541556. 24. McNeilly MD, Robinson EL, Anderson NB, et al. Effects of racist provocation and social support on cardiovascular reactivity in African American women. Int J Behav Med. 1995;2: 321338.

25. Blascovich J, Spencer SJ, Quinn D, Steele C. African Americans and high blood pressure: the role of stereotype threat. Psychol Sci. 2001;12:225229. 26. Sutherland ME, Harrell JP. Individual differences in physiological responses to fearful, racially noxious and neutral imagery. Imagination Cogn Pers. 1986;6:133150. 27. Morris-Prather CE, Harrell JP, Collins R, Jeffries Leonard KL, Boss M, Lee JW. Gender differences in mood and cardiovascular responses to socially stressful stimuli. Ethn Dis. 1996;6: 109122. 28. Kinzie JD, Denney D, Riley C, Boehnlein J, McFarland B, Leung P. A cross-cultural study of reactivation of posttraumatic stress disorder symptoms: American and Cambodian psychophysiological responses to viewing traumatic

video scenes. J Nerv Ment Dis. 1998; 186:670676. 29. Fang CY, Myers HF. The effects of racial stressors and hostility on cardiovascular reactivity in African American and Caucasian men. Health Psychol. 2001;20:6470. 30. Jones DR, Harrell JP, MorrisPrather CE, Thomas J, Omowale N. Affective and physiological responses to racism: the roles of Afrocentrism and mode of presentation. Ethn Dis. 1996;6: 109123. 31. Torres A, Bowens L. Correlations between the internalization theme of the Racial Identity Attitude Survey-B and systolic blood pressure. Ethn Dis. 2000;10:375383. 32. Harrell JP. Psychological factors and hypertension: a status report. Psychol Bull. 1980;87:482501.

33. Berntson GG, Sarter M, Cacioppo JT. Anxiety and cardiovascular reactivity: the basal forebrain cholinergic link. Behav Brain Res. 1998;94:225248. 34. Berntson GG, Cacioppo JT. From homeostasis to allodynamic regulation. In: Cacioppo JT, Tassinary LG, Berntson GG, eds. Handbook of Psychophysiology. 2nd ed. Cambridge, England: Cambridge University Press; 2000. 35. Clark R, Anderson NB, Clark VR, Williams DR. Racism as a stressor for African Americans: a biopsychosocial model. Am Psychol. 1999;54:805816. 36. Henry JP, Liu J, Meehan WP. Psychosocial stress and experimental hypertension. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis and Management. New York, NY: Raven Press; 1995:905921.

Paved With Good Intentions: Do Public Health and Human Service Providers Contribute to Racial/Ethnic Disparities in Health?
There is extensive evidence of racial/ethnic disparities in receipt of health care. The potential contribution of provider behavior to such disparities has remained largely unexplored. Do health and human service providers behave in ways that contribute to systematic inequities in care and outcomes? If so, why does this occur? The authors build on existing evidence to provide an integrated, coherent, and sound approach to research on providers contributions to racial/ ethnic disparities. They review the evidence regarding provider contributions to disparities in outcomes and describe a causal model representing an integrated set of hypothesized mechanisms through which health care providers behaviors may contribute to these disparities. (Am J Public Health. 2003;93:248255)

| Michelle van Ryn, PhD, MPH, and Steven S. Fu, MD, MSCE
THERE IS EXTENSIVE EVIDENCE of racial/ethnic disparities in receipt of a wide range of health and social services. The past decade has seen an explosion of empirical literature documenting racial/ethnic disparities in medical care received, independent of clinical appropriateness, insurance status, treatment site, and other clinical and socioeconomic status (SES) correlates.16 These findings parallel documentations of disparities in the justice, child welfare, education, labor, and housing sectors. This deluge of evidence leads us to some painful questions. Because institutional racism (differential processes or outcomes according to race/ethnicity) is the result of the sum total of policies and procedures created and enforced, and the behaviors engaged in, by institutional members, we must ask whether health and human service providers directly contribute to these racial/ ethnic disparities in care and health outcomes. If so, how does this occur? Our intent here is to begin to answer such questions by applying existing theory and evidence to a framework designed to guide future inquiry and intervention. First, we provide a brief summary of the empirical evidence supporting the hypothesis that the behavior of health and human service providers contributes to race/ethnicity differences in care and, thus, institutional discrimination. Second, we integrate the empirical evidence regarding social cognition, unintentional bias, and provider behavior into a hypothesized causal model. Social cognition is a subfield of social psychology that studies how we make sense of other people, that is, the mental representations and processes that underlie social perception, social judgment, social interaction, and social influence. Social psychologists focusing on social cognition have been studying the ways in which group characteristics influence person perception and interpersonal processes for several decades, resulting in a massive body of evidence with significant implications for understanding how race/ethnicity influences provider behavior. Finally, we briefly discuss implications for future directions and interventions.

EVIDENCE OF CONTRIBUTIONS TO DISPARITIES

Public health, medical care, and human service providers

248 | Racial/Ethnic Bias and Health | Peer Reviewed | van Ryn and Fu

American Journal of Public Health | February 2003, Vol 93, No. 2

 RACIAL/ETHNIC BIAS AND HEALTH

may influence race/ethnicity and class health disparities in several interconnected ways. First, providers may influence help seekers views of themselves and their relation to the world (society, culture, community). For example, providers may intentionally or unintentionally reflect and reinforce societal messages regarding help seekers fundamental value, self-reliance, competence, and deservingness.79 Providers may both have and intentionally or unintentionally communicate lower expectations for patients in disadvantaged social positions (owing to their race/ethnicity, income, education, class, or any stigmatized characteristic) than for their more advantaged counterparts. In this way, providers can influence help seekers expectations for the future, the degree to which they expect to obtain the resources and help they need, and their expectations for improvements in their situations or conditions,1016 which in turn may account for some of the disparities observed in outcomes and health status.14,17,18 Second, the potential influence of provider communications on help seekers health-related cognition and behavior has been well documented.7,1932 Thus, disparities in communications regarding health promotion and disease prevention behavior and services may account for an additional portion of documented health disparities. Third, providers are powerful gatekeepers and may influence health disparities via such mechanisms as differential access to treatments or services and loss of benefits and rights. This is the mechanism that has received the most research attention and thus is a significant part of our focus here.

Research focusing specifically on provider contributions to racial/ethnic disparities in care is in its infancy. However, a growing body of evidence is reviewed briefly in the sections to follow (for a more extensive review, see van Ryn33).

lated pain,43,44 and evaluation of chest pain emergencies.45 These studies suggest that both nurses and physicians may contribute to racial/ethnic disparities in care, although in general the role of nurses has been understudied.

Mental Health Services

It has been shown that, independent of clinical factors, both US and UK psychiatrists are more likely to prescribe antipsychotic medications to non-Whites than to their White counterparts; also, these individuals are more likely to be involuntarily hospitalized and to be placed in seclusion once hospitalized.4663 In addition, African Americans and Latinos have been found to be less likely than Whites to receive guideline-adherent treatment64,65 and follow-up.66

Specialty Care
There is good evidence of physician contributions to racial/ethnic disparities in both kidney transplant rates and cardiac procedures.3436 For example, one study showed that Black dialysis patients were less likely than their clinically similar White counterparts to be told about transplantation, obtain all of the medical information they desire, discuss with a physician the possibility of receiving a kidney from a family member, and report that a physician had recommended a transplant. Among patients who were certain they wanted a transplant, Blacks were less likely than Whites to be referred for evaluation and to be placed on a waiting list.35 Similarly, racial/ethnic disparities have been extensively documented in relation to cardiac tests, diagnoses, and procedures,2,37 with several studies revealing significant provider contributions to these disparities via treatment recommendations and clinical decisionmaking.36,38,39

in health care and outcomes, these studies do not provide sufficient insight into why providers behavior varies by race/ethnicity to allow for meaningful conclusions or intervention directions. Thus, the following section is intended to help guide research and intervention directions by proposing a causal model representing a set of hypotheses regarding the effect of help seekers race/ethnicity on providers beliefs and behaviors. These hypotheses result from an integration of the social cognition literature and the empirical literature on providerpatient interactions.

IMPLICATIONS REGARDING SOCIAL COGNITION

Figure 1 represents an integration of the social cognition and provider behavior research into a hypothesized model of the effects of help seeker race/ethnicity on provider behaviors and, in turn, the effects of these behaviors on help seeker outcomes (care received, case disposition, or service provided). It is important to note here that the effect of social class is likely to be as powerful as that of race/ethnicity and that there are likely to be significant interactions between class and race/ethnicity. In addition, the concepts represented in Figure 1 are clearly influenced by economic, social, and setting factors. However, these important structural factors, as well as the independent and interaction effects of class, have received little research attention and are beyond our scope here. Arrow A in Figure 1 reflects the hypothesis that the primary cognitive mediator of the effect of help seeker race/ethnicity on provider behavior is providers

Child Welfare and At-Risk Youths

African American children are more likely to be placed in foster homes than White children and, along with Hispanic and American Indian children, are overrepresented in child abuse and neglect reports.67,68 These disparities persist even though 3 national incidence studies conducted by the US Department of Health and Human Services indicate that child maltreatment does not vary according to race/ethnicity.69 Similarly, in regard to every offense category, Black youths are more likely than White youths to be incarcerated. Among those charged with drug offenses, Black youths are 48 times more likely than White youths to be sentenced to juvenile prison.70 Although this body of evidence provides support for the hypothesis that providers contribute to at least a portion of the observed racial/ethnic disparities

Pain Assessment and Treatment

Non-Whites have been found to be at significantly higher risk for inadequate or no pain assessment or pain control than their White counterparts in a variety of situations, including emergency department treatment of long bone fractures,40,41 nonmalignant pain in a nursing home,42 treatments for cancer-re-

February 2003, Vol 93, No. 2 | American Journal of Public Health

van Ryn and Fu | Peer Reviewed | Racial/Ethnic Bias and Health | 249

 RACIAL/ETHNIC BIAS AND HEALTH

Note. HS = help seeker.

FIGURE 1Proposed mechanisms through which health and human service providers can influence race/ethnicity disparities in treatment.
conscious and unconscious beliefs about the help seeker. We expect providers to conduct encounters, make assessments, and recommend courses of action in a way that it is unaffected or unbiased by the sociodemographic characteristics of the people they serve. In addition, they are expected to be attuned to cultural differences and to be culturally sensitive as they work, in an unbiased manner, with various populations. Finally, they are expected to engage in these practices with little specific training or support. Unfortunately, there is a massive body of research on social categorization and stereotyping demonstrating that humans universally apply stereotypes when making sense of other people.7175 All humans share the highly adaptive cognitive strategy of making the world more manageable by using categorizing and generalizing techniques to simplify the massive amounts of complex information and stimuli to which they are exposed.73,76,77 This process is efficient. It simplifies cognitive processing, reduces effort, and frees cognitive resources to meet other demands.78 In applying this process to the social world, people develop beliefs and expectations about categories or groups of people and generalize these beliefs and expectations to all individuals mentally assigned to a given category or group.72,73,7680 There is substantial evidence that when people mentally assign an individual to a particular class or group, they unconsciously and automatically81 assign the characteristics of that group to the individual in question, a process referred to as stereotype application.73,8284 Although most people do not believe that they engage in stereotyping, almost all of us are able to quickly list the characteristics and describe the images that automatically come to mind when we hear words such as librarian, welfare mother, plumber, Navajo, Minnesotan, or New Yorker. It is both difficult and painful for many of us to accept the massive evidence that social categories automatically and unconsciously influence the way we perceive people and, in turn, influence the way in which we interpret their behavior and behave toward them. However, given that this type of strategy is common to all humans in all cultures and is more likely to be used in situations that tax cognitive resources (e.g., time pressure),78,82,8588 the expectation that providers will be immune is unrealistic.

There is substantial evidence that patients sex, age, diagnosis, marital status, sexual orientation, type of illness,29,30,78,89105 and, more recently, race/ethnicity58,71,106108 can influence providers beliefs and expectations. For example, one study revealed that, independent of physician characteristics and patient personality, clinical, and sociodemographic characteristics, cardiac patients race/ethnicity and socioeconomic status negatively influenced physicians ratings of their personality, education, intelligence, career demands, and likely treatment adherence.71 However, existing research is limited and unable to predict circumstances in which providers perceptions will or will not be influenced by help seeker characteristics, nor can we predict the specific perceptions that will be influenced. Arrow B reflects the hypothesis that providers beliefs about help seekers influence their interpretation of the problems or symptoms of these individuals. There is ample evidence that we interpret information about others through a screen, filter, or framework of beliefs created by the way in which we cognitively classify them.72,73,77,78,90,109 A number of studies have shown that observers assign different meanings to the same behavior depending on the race, class, or other demographic characteristics of the individual involved.73,74,77,110114 The effect of race/ethnicity on interpretations of behavior is exacerbated when the behavior is ambiguous or open to multiple interpretations.74,115 For example, Burk and Sher found that mental health workers diagnoses of a videotaped adolescent varied significantly according to whether they were

250 | Racial/Ethnic Bias and Health | Peer Reviewed | van Ryn and Fu

American Journal of Public Health | February 2003, Vol 93, No. 2

 RACIAL/ETHNIC BIAS AND HEALTH

told or not told that they were watching the child of an alcoholic.116 Another example of this phenomenon can be found in a study in which one group of 20 psychotherapists was presented with a scenario involving a White adolescent and a second group of 20 psychotherapists was presented with an identical scenario involving a Black adolescent. Overall, the behaviors of the Black adolescent were rated as less clinically significant than the behaviors of the White adolescent.117 Finally, Stern et al. found that medical students and Israeli providers assessments of normal children were negatively influenced by whether or not they were told the children had been born prematurely.96,97 Arrow C reflects the hypothesis that providers beliefs about help seekers social and behavioral characteristics directly influence their professional or clinical decisionmaking. This may occur as a result of either of 2 kinds of interconnected reasoning: moral rationing or appropriateness. In the case of moral rationing, the provider believes that a patients characteristics make her or him more or less deserving of treatment. For example, a cardiac surgeon told one of the authors that he was not going to treat cardiac patients who were just going to go out and do drugs. In the case of appropriateness, the provider believes that a social or behavioral characteristic of the help seeker makes her or him more or less appropriate for a particular treatment or more or less likely to benefit from a service or procedure. To date, only 2 studies of which we are aware have tested this hypothesized cause of racial/ethnic disparities in treatment. Bogart and colleagues

found that physicians were more likely to provide highly active antiretroviral therapy to HIV/AIDS patients when they were perceived as likely to be adherent to treatment.118 They then examined patient characteristics associated with physician predictions of adherence by randomly assigning physicians to review patient vignettes varying only in terms of the patients sex, disease severity, ethnicity, and risk group. Results revealed that, independent of other factors, the patient depicted in the African American vignette was more likely to be rated as nonadherent.119 Similarly, van Ryn et al. found that physicians ratings of patients likelihood of having adequate social support or participating in cardiac rehabilitation predicted physicians recommendations for revascularization, independent of the clinical appropriateness of revascularization and patient demographic characteristics. In turn, these physicians was more likely to rate African American patients as lacking in social support and as more unlikely to participate in cardiac rehabilitation than White patients.120 It is possible that these findings are due to providers overapplication of population statistics to individual help seekers. There is evidence that data on population rates may be incorporated into providers general belief systems and professional decisionmaking such that group probabilities are overapplied to decisionmaking about individuals.121,122 In this way, providers may fail to correctly incorporate individual data, instead being swayed by their beliefs regarding the probability of individuals in a particular sociodemographic category possessing a given charac-

teristic.123 This is a disturbing possibility because it suggests that statistical data on marginalized populations may be inadvertently incorporated and applied by providers in a way that reinforces marginalization. Some of the data that reinforce bias in fact result from bias, compounding this situation. For example, as mentioned earlier, although 3 national incidence studies indicate that child maltreatment does not vary according to race,69 minority groups are heavily overrepresented in child neglect and abuse reports; as a result, in publications based on reported cases of abuse, authors state that victimization varies by race.68,124 Arrow D reflects the hypothesis that providers conscious beliefs about and unconscious stereotypes regarding help seekers influence their interpersonal behavior. There is considerable evidence that stereotypes activated outside of consciousness influence our behavior toward other individuals and thus affect their reactions and behaviors, as represented in arrow I.74,125 For example, one study revealed that subliminal exposure to photographs of African Americans as opposed to photographs of Whites caused naive participants to unknowingly behave in a more hostile manner in a subsequent word-guessing game.125 The partners of these participants, who were also unaware of the experiments design, in turn behaved in a more hostile manner. These findings illustrate the way in which the well-documented selffulfilling prophecy can be produced. Provider behavior may be influenced by help seeker race/ ethnicity (or another characteristic), and this behavior influences help seekers behavior in turn.77,82,125127

Several studies directly support the hypothesis that help seekers characteristics influence provider behavior. In wellcontrolled studies, physicians have been found to have a less participatory decisionmaking style128,129 and to adopt a more narrowly biomedical communication pattern (characterized by low patient control of communication and psychosocial talk and high levels of physician biomedical information provision and closed-ended question asking) with non-White than with White patients.130 In other studies, patients characteristics have been found to be associated with provider communication effectiveness131133 and physician interpersonal behaviors such as nonverbal attention, empathy, courtesy, and information giving.134 Arrows E and F represent the hypotheses that providers interpersonal behavior influences help seekers satisfaction as well as cognitive factors such as attitudes, self-efficacy, trust, and behavioral intentions, while arrow G represents the well-established causal relationship between cognitive factors and behavior. There is a substantial body of evidence indicating a strong relationship between provider interpersonal behavior and patient satisfaction, adherence, utilization, and outcomes.7,19,20,2932 A more participatory decisionmaking style on the part of providers, which can be defined as increased help seeker engagement in medical care through information sharing, negotiation, and consensus seeking, has been shown to be positively associated with satisfaction,29,30,128130 disclosure,135 successful self-management, adherence, lower pain levels, and

February 2003, Vol 93, No. 2 | American Journal of Public Health

van Ryn and Fu | Peer Reviewed | Racial/Ethnic Bias and Health | 251

 RACIAL/ETHNIC BIAS AND HEALTH

symptom recovery20,21,25,136139 among patients. In addition, there is evidence that help seekers race/ethnicity and SES influence the substantive content of encounters. For example, 4 separate studies involving patients with advanced illnesses showed that non-White patients were less likely to report that their provider discussed endof-life care with them than their White counterparts, despite their equal or stronger desire to have such a discussion.133,140142 Similarly, African Americans and low-income patients have been found to be less likely to report receiving advice to quit smoking during their primary care visit,143 less likely to discuss diet and exercise with their physician,144 less likely to receive a recommendation for mammography,144146 and less likely to receive prenatal preventive care advice regarding smoking cessation, alcohol use, and breastfeeding.147 Arrow H represents the hypothesis that providers beliefs about help seekers are influenced by help seekers behaviors during the encounter. In one study testing this hypothesis, physicians viewed randomly assigned videotapes of women seeking care for breast cancer; the videotapes varied only in regard to the patients sociodemographic characteristics, general health status, and assertiveness. Results revealed that physician decisions were affected by the interaction of patient assertiveness with race and SES, in that assertive behavior increased the likelihood of full tumor staging among Black and low-SES patients but not White or upperSES patients.148 This study suggests the possibility that providers beliefs and behaviors are influenced by complex inter-

actions between patients race/ ethnicity and behavior. While not a primary focus of this review, it is important to bear in mind that a variety of class and cultural factors contribute to help seeker variations in health beliefs, beliefs regarding the nature of care and cure, and past experiences with and attitudes toward providers. These factors, in turn, influence help seekers interactions with providers.149151 Further research is necessary to examine the complex interactions between the expectations, beliefs, and behaviors that help seekers and providers bring to their encounters.

CONCLUSIONS
In conclusion, there is sufficient evidence for the hypothesis that provider behaviors contribute to racial/ethnic disparities in care to warrant further study and intensive efforts to develop evidence-based interventions. There have been insufficient resources devoted to this line of inquiry, perhaps reflecting the fact that bias on the part of providers is very uncomfortable and disturbing to contemplate. Many providers, devoted to justice and equity, may understandably find the hypothesis that they and their professional peers contribute to systematic inequities in health difficult to believe. Yet, we would argue that resistance to exploring the ways in which providers may contribute to health disparities reflects a lack of understanding of the automatic, unconscious, and ubiquitous nature of fundamental social cognition processes. In addition, lack of attention to the interpersonal mechanisms or mediators of institutionalized discrimination may undermine the

effectiveness of our strongest policy- and organizational-level strategies. Laws, incentives, mandates, court-based remedies, reimbursement methodologies, sanctions, and reorganizations may not create the desired effects if the fundamental human information-processing, social cognition, and social interaction processes that contribute to institutional discrimination are not addressed. Furthermore, the current lack of research in this area creates a barrier to the development of evidence-based interventions addressing racial/ethnic disparities in care. Our ability to develop informed interventions is severely limited by our lack of knowledge regarding the effects of providers education, socialization, setting, and other characteristics on the degree to which their behaviors are influenced by help seekers race/ethnicity. Two overarching categories of research are suggested by the evidence gathered to date: (1) multimethod studies (e.g., combining clinical, survey, and qualitative data) intended to test and refine the hypotheses presented here and provide greater knowledge of the conditions under which provider behaviors do and do not influence racial/ethnic disparities and (2) theory- and evidence-based intervention research intended to: develop and test methods that influence racial/ethnic variations in providers perceptions of help seekers; reduce disparities in providers and help seekers behaviors during their encounters; and examine the resultant effects on racial/ethnic variations in care. Finally, although social categorization is an automatic and dominant information-processing strategy, recent reviews of the evidence suggest that some of

the associated processes may be controllable in certain circumstances.82,87 These reviews point to the crucial influence of 3 broad categories of factors that may influence automatic activation of stereotypes: public health providers acceptance and awareness of the automatic nature of these processes, providers motivation to detect and inhibit stereotypes, and setting and system characteristics that allow providers to have sufficient cognitive resources to overcome and replace automatic cognitive processes. This evidence clearly suggests that while awareness and motivation are necessary conditions for individuals to exert control over stereotypes, they are not sufficient. Considerable reorganization of our service delivery and reimbursement systems may be needed if providers are to have the time and cognitive resources needed to overcome unconscious bias.82,87

About the Authors

Michelle van Ryn is with the Center for Chronic Disease Outcomes Research, Minneapolis Veterans Affairs Medical Center, and the Department of Epidemiology, University of Minnesota School of Public Health, Minneapolis. Steven S. Fu is with the Center for Chronic Disease Outcomes Research, Section of General Internal Medicine, Minneapolis Veterans Affairs Medical Center, and the Department of Medicine, University of Minnesota. Requests for reprints should be sent to Michelle van Ryn, PhD, MPH, Minneapolis VAMC/Center for Chronic Disease Outcomes Research, One Veterans Dr (152/2E), Minneapolis, MN 55417 (e-mail: vanryn@epi.umn.edu). This article was accepted October 22, 2002. Note. The findings and conclusions presented here are solely the responsibility of the authors.

Acknowledgments
The present work was supported in part by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development

252 | Racial/Ethnic Bias and Health | Peer Reviewed | van Ryn and Fu

American Journal of Public Health | February 2003, Vol 93, No. 2

 RACIAL/ETHNIC BIAS AND HEALTH

Service, and the National Institutes of Health, Office of Behavioral and Social Sciences. Portions of this research were presented at the National Institutes of Health Conference on Racial Bias and Health, April 2002; the Health Services Research and Development Service 19th Annual Meeting, February 2001, Washington, DC; and the 2001 Academy for Health Services Research meeting in Atlanta, Ga. We wish to thank H. Jack Geiger and Joseph Betancourt for helpful feedback.

Hope in nursing research: a metaanalysis of the ontological and epistemological foundations of research on hope. J Adv Nurs. 1997;25:364371. 13. Staudenmayer H. Clinical consequences of the EI/MCS diagnosis: two paths. Regul Toxicol Pharmacol. 1996; 24:S96S110. 14. Gilliss CL, Gortner SR, Hauck WW, Shinn JA, Sparacino PA, Tompkins C. A randomized clinical trial of nursing care for recovery from cardiac surgery. Heart Lung. 1993;22:125133. 15. Olsson B, Tibblin G. Effect of patients expectations on recovery from acute tonsillitis. Fam Pract. 1989;6: 188192. 16. Heaney C, van Ryn M. The implication of status, class, gender, and cultural diversity for health education practice: the case of worksite stress reduction programs. Health Educ Res Theory Pract. 1996;11:5770. 17. Mondloch MV, Cole DC, Frank JW. Does how you do depend on how you think youll do? A systematic review of the evidence for a relation between patients recovery expectations and health outcomes. Can Med Assoc J. 2001;165: 174179. 18. Borkan JM, Quirk M. Expectations and outcomes after hip fracture among the elderly. Int J Aging Hum Dev. 1992; 34:339350. 19. Di Blasi Z, Harkness E, Ernst E, Georgiou A, Kleijnen J. Influence of context effects on health outcomes: a systematic review. Lancet. 2001;357: 757762. 20. Stewart M, Brown JB, Donner A, et al. The impact of patient-centered care on outcomes. J Fam Pract. 2000; 49:796804. 21. Greenlund KJ, Giles WH, Keenan NL, Croft JB, Mensah GA. Physician advice, patient actions, and health-related quality of life in secondary prevention of stroke through diet and exercise. Stroke. 2002;33:565571. 22. Glasgow RE, Eakin EG, Fisher EB, Bacak SJ, Brownson RC. Physician advice and support for physical activity: results from a national survey. Am J Prev Med. 2001;21:189196. 23. Lane DS, Zapka J, Breen N, Messina CR, Fotheringham DJ. A systems model of clinical preventive care: the case of breast cancer screening among older women. Prev Med. 2000; 31:481493. 24. Brenes GA, Paskett ED. Predictors of stage of adoption for colorectal cancer screening. Prev Med. 2000;31: 410416. 25. Greenlund KJ, Keenan NL, Ander-

son LA, Mandelson MT, Newton KM, LaCroix AZ. Does provider prevention orientation influence female patients preventive practices? Am J Prev Med. 2000;19:104110. 26. May DS, Kiefe CI, Funkhouser E, Fouad MN. Compliance with mammography guidelines: physician recommendation and patient adherence. Prev Med. 1999;28:386394. 27. Li VC, Coates TJ, Ewart CK, Kim YJ. The effectiveness of smoking cessation advice given during routine medical care: physicians can make a difference. Am J Prev Med. 1987;3:8186. 28. Nichol KL, MacDonald R, Hauge M. Factors associated with influenza and pneumococcal vaccination behavior among high-risk adults. J Gen Intern Med. 1996;11:673677. 29. Bertakis KD, Callahan EJ, Helms LJ, Azari R, Robbins JA, Miller J. Physician practice styles and patient outcomes: differences between family practice and general internal medicine. Med Care. 1998;36:879891. 30. Bertakis KD, Roter D, Putnam SM. The relationship of physician medical interview style to patient satisfaction. J Fam Pract. 1991;32:175181. 31. Tates K, Meeuwesen L. Doctor-parent-child communication: a review of the literature. Soc Sci Med. 2001;52: 839851. 32. Rao JK, Weinberger M, Kroenke K. Visit-specific expectations and patientcentered outcomes: a literature review. Arch Fam Med. 2000;9:11481155. 33. van Ryn M. Research on the provider contribution to race/ethnicity disparities in medical care. Med Care. 2002;40(suppl 1):I140I151. 34. Thamer M, Hwang W, Fink NE, et al. U.S. nephrologists attitudes towards renal transplantation: results from a national survey. Transplantation. 2001;71: 281288. 35. Ayanian JZ, Cleary PD, Weissman JS, Epstein AM. The effect of patients preferences on racial differences in access to renal transplantation. N Engl J Med. 1999;341:16611669. 36. Hannan EL, van Ryn M, Burke J, et al. Access to coronary artery bypass surgery by race/ethnicity and gender among patients who are appropriate for surgery. Med Care. 1999;37:6877. 37. Kressin NR, Petersen LA. Racial differences in the use of invasive cardiovascular procedures: review of the literature and prescription for future research. Ann Intern Med. 2001;135: 352366. 38. Maynard C, Fisher LD, Passamani ER, Pullum T. Blacks in the Coronary

Artery Surgery Study (CASS): race and clinical decision making. Am J Public Health. 1986;76:14461448. 39. Naumburg EH, Franks P, Bell B, Gold M, Engerman J. Racial differentials in the identification of hypercholesterolemia. J Fam Pract. 1993;36:425430. 40. Todd KH, Deaton C, DAdamo AP, Goe L. Ethnicity and analgesic practice. Ann Emerg Med. 2000;35:1116. 41. Todd KH, Samaroo N, Hoffman JR. Ethnicity as a risk factor for inadequate emergency department analgesia. JAMA. 1993;269:15371539. 42. Won A, Lapane K, Gambassi G, Bernabei R, Mor V, Lipsitz LA. Correlates and management of nonmalignant pain in the nursing home. J Am Geriatr Soc. 1999;47:936942. 43. Anderson KO, Mendoza TR, Valero V, et al. Minority cancer patients and their providers: pain management attitudes and practice. Cancer. 2000;88: 19291938. 44. Cleeland CS, Gonin R, Baez L, Loehrer P, Pandya KJ. Pain and treatment of pain in minority patients with cancer. Ann Intern Med. 1997;127: 813816. 45. Bell PD, Hudson S. Equity in the diagnosis of chest pain: race and gender. Am J Health Behav. 2001;25: 6071. 46. Cuffe SP, Waller JL, Cuccaro ML, Pumariega AJ, Garrison CZ. Race and gender differences in the treatment of psychiatric disorders in young adolescents. J Am Acad Child Adolesc Psychiatry. 1995;34:15361543. 47. DelBello MP, Lopez-Larson MP, Soutullo CA, Strakowski SM. Effects of race on psychiatric diagnosis of hospitalized adolescents: a retrospective chart review. J Child Adolesc Psychopharmacol. 2001;11:95103. 48. Kales HC, Blow FC, Bingham CR, Roberts JS, Copeland LA, Mellow AM. Race, psychiatric diagnosis, and mental health care utilization in older patients. Am J Geriatr Psychiatry. 2000;8: 301309. 49. Kales HC, Blow FC, Bingham CR, Copeland LA, Mellow AM. Race and inpatient psychiatric diagnoses among elderly veterans. Psychiatr Serv. 2000;51: 795800. 50. Strakowski SM, Shelton RC, Kolbrener ML. The effects of race and comorbidity on clinical diagnosis in patients with psychosis. J Clin Psychiatry. 1993;54:96102. 51. Strakowski SM, Lonczak HS, Sax KW, et al. The effects of race on diagnosis and disposition from a psychiatric emergency service. J Clin Psychiatry. 1995;56:101107.

References
1. Smedley BD, Stith AY, Nelson AR, eds. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academy Press; 2002. 2. Einbinder LC, Schulman KA. The effect of race on the referral process for invasive cardiac procedures. Med Care Res Rev. 2000;57(suppl 1):162180. 3. Fiscella K, Franks P, Gold MR, Clancy CM. Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA. 2000;283:25792584. 4. Fiscella K, Franks P, Doescher MP, Saver BG. Disparities in health care by race, ethnicity, and language among the insured: findings from a national sample. Med Care. 2002;40:5259. 5. Etchason J, Armour B, Ofili E, et al. Racial and ethnic disparities in health care. JAMA. 2001;285:883. 6. Oddone EZ, Petersen LA, Weinberger M, Freedman J, Kressin NR. Contribution of the Veterans Health Administration in understanding racial disparities in access and utilization of health care: a spirit of inquiry. Med Care. 2002;40(suppl 1):I3I13. 7. Roter D. The medical visit context of treatment decision-making and the therapeutic relationship. Health Expectations. 2000;3:1725. 8. Roter DL, Hall JA. Health education theory: an application to the process of patient-provider communication. Health Educ Res. 1991;6:185193. 9. van Ryn M, Heaney CA. Developing effective helping relationships in health education practice. Health Educ Behav. 1997;24:683702. 10. Kylma J, Vehvilainen-Julkunen K, Lahdevirta J. Hope, despair and hopelessness in living with HIV/AIDS: a grounded theory study. J Adv Nurs. 2001;33:764775. 11. Centers LC. Beyond denial and despair: ALS and our heroic potential for hope. J Palliat Care. 2001;17:259264. 12. Kylma J, Vehvilainen-Julkunen K.

February 2003, Vol 93, No. 2 | American Journal of Public Health

van Ryn and Fu | Peer Reviewed | Racial/Ethnic Bias and Health | 253

 RACIAL/ETHNIC BIAS AND HEALTH

52. Strakowski SM, Hawkins JM, Keck PE Jr., et al. The effects of race and information variance on disagreement between psychiatric emergency service and research diagnoses in first-episode psychosis. J Clin Psychiatry. 1997;58: 457463. 53. Crawford K, Fisher WH, McDermeit M. Racial/ethnic disparities in admissions to public and private psychiatric inpatient settings: the effect of managed care. Adm Policy Ment Health. 1998;26:101109. 54. Davies S, Thornicroft G, Leese M, Higgingbotham A, Phelan M. Ethnic differences in risk of compulsory psychiatric admission among representative cases of psychosis in London. BMJ. 1996;312:533537. 55. Dixon L, Green-Paden L, Delahanty J, Lucksted A, Postrado L, Hall J. Variables associated with disparities in treatment of patients with schizophrenia and comorbid mood and anxiety disorders. Psychiatr Serv. 2001;52: 12161222. 56. Kilgus MD, Pumariega AJ, Cuffe SP. Influence of race on diagnosis in adolescent psychiatric inpatients. J Am Acad Child Adolesc Psychiatry. 1995;34: 6772. 57. Lewis DO, Balla DA, Shanok SS. Some evidence of race bias in the diagnosis and treatment of the juvenile offender. Am J Orthopsychiatry. 1979;49: 5361. 58. Lewis G, Croft-Jeffreys C, David A. Are British psychiatrists racist? Br J Psychiatry. 1990;157:410415. 59. Benson PR. Factors associated with antipsychotic drug prescribing by southern psychiatrists. Med Care. 1983;21: 639654. 60. Lawson WB, Hepler N, Holladay J, Cuffel B. Race as a factor in inpatient and outpatient admissions and diagnosis. Hosp Community Psychiatry. 1994; 45:7274. 61. Takei N, Persaud R, Woodruff P, Brockington I, Murray RM. First episodes of psychosis in Afro-Caribbean and White people: an 18-year follow-up population-based study. Br J Psychiatry. 1998;172:147153. 62. Whaley A. Racism in the provision of mental health services: a socialcognitive analysis. Am J Orthopsychiatry. 1998;68:4757. 63. Flaherty JA, Meagher R. Measuring racial bias in inpatient treatment. Am J Psychiatry. 1980;137:679682. 64. Wang PS, Berglund P, Kessler RC. Recent care of common mental disorders in the United States: prevalence and conformance with evidence-based recommendations. J Gen Intern Med. 2000;15:284292.

65. Young AS, Klap R, Sherbourne CD, Wells KB. The quality of care for depressive and anxiety disorders in the United States. Arch Gen Psychiatry. 2001;58:5561. 66. Schneider EC, Zaslavsky AM, Epstein AM. Racial disparities in the quality of care for enrollees in Medicare managed care. JAMA. 2002;287: 12881294. 67. Morton T, Holder W. Decision Making in Childrens Protective Services: Advancing the State of the Art. Atlanta, Ga: Child Welfare Institute; 1998. 68. Drake B, Zuravin S. Bias in child maltreatment reporting: revisiting the myth of classlessness. Am J Orthopsychiatry. 1998;68:295304. 69. Sedlak AJ, Broadhurst DD. Third National Incidence Study of Child Abuse and Neglect. Washington, DC: US Dept of Health and Human Services; 1996. 70. Poe-Yamagata M, Jones E. Juvenile Arrests 1998. Washington, DC: Office of Juvenile Justice and Delinquency Prevention; 2000. 71. van Ryn M, Burke J. The effect of patient race and socio-economic status on physicians perceptions of patients. Soc Sci Med. 2000;50:813828. 72. Moskowitz GB, ed. Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition. Mahwah, NJ: Lawrence Erlbaum Associates; 2001. 73. Stangor C, ed. Stereotypes and Prejudice. Philadelphia, Pa: Psychology Press; 2001. 74. Kunda Z, Sherman-Williams B. Stereotypes and the construal of individuating information. Pers Soc Psychol Bull. 1993;19:9099. 75. Macrae CN, Bodenhausen GV. Social cognition: categorical person perception. Br J Psychol. 2001;92: 239255. 76. Hamilton DL. Cognitive Processes in Stereotyping and Intergroup Behavior. Hillsdale, NJ: Lawrence Erlbaum Associates; 1981. 77. Kunda Z. Social Cognition: Making Sense of People. Cambridge, Mass: MIT Press; 1999. 78. Macrae CN, Milne AB, Bodenhausen GV. Stereotypes as energy-saving devices: a peek inside the cognitive toolbox. J Pers Soc Psychol. 1994;66: 3347. 79. Hamilton DL, Trolier TK. Stereotypes and stereotyping: an overview of the cognitive approach. In: Dovidio JF, Gaetner SL, eds. Prejudice, Discrimination, and Racism. San Diego, Calif: Academic Press Inc; 1986:127163. 80. Andersen SM, Klatzky RL, Murray J. Traits and social stereotypes: effi-

ciency differences in social information processing. J Pers Soc Psychol. 1990;59: 192201. 81. Bargh JA, Chaiken S, Govender R, Pratto F. The generality of the automatic attitude activation effect. J Pers Soc Psychol. 1992;62:893912. 82. Blair I. Implicit stereotypes and prejudice. In: Moskowitz G, ed. Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition. Mahwah, NJ: Lawrence Erlbaum Associates; [[6]] 2001. 83. Vassiliou V, Trandis H, Vassiliou G, McGuire H. Interpersonal contact and stereotyping. In: Triandis H, ed. The Analysis of Subjective Culture. New York, NY: John Wiley & Sons Inc; [[7]] 1972. 84. Lalonde RN, Gardner RC. The intergroup perspective on stereotype organization and processing. Br J Soc Psychol. 1989;28:289303. 85. Gilbert DT, Hixon JG. The trouble of thinking: activation and application of stereotypic beliefs. J Pers Soc Psychol. 1991;60:509517. 86. Macrae CN, Bodenhausen GV, Schloerscheidt AM, Milne AB. Tales of the unexpected: executive function and person perception. J Pers Soc Psychol. 1999;76:200213. 87. Monteith M, Voils C. Control over prejudice. In: Moskowitz G, ed. Cognitive Social Psychology: The Princeton Symposium on the Legacy and Future of Social Cognition. Mahwah, NJ: Lawrence Erlbaum Associates; [[8]] 2001. 88. Pratto F, Bargh JA. Stereotyping based on apparently individuating information: trait and global components of sex stereotypes under attention overload. J Exp Soc Psychol. 1991;27:2647. 89. Gerbert B. Perceived likeability and competence of simulated patients: influence on physicians management plans. Soc Sci Med. 1984;18: 10531059. 90. Tobin JN, Wasserheil-Smoller S, Wexler JP, et al. Sex bias in considering coronary bypass surgery. Ann Intern Med. 1987;107:1925. 91. Hall JA, Epstein AM, DeCiantis ML, McNeil BJ. Physicians liking for their patients: more evidence for the role of affect in medical care. Health Psychol. 1993;12:140146. 92. Kearney N, Miller M, Paul J, Smith K. Oncology healthcare professionals attitudes toward elderly people. Ann Oncol. 2000;11:599601. 93. Like R, Zyzanski SJ. Patient satisfaction with the clinical encounter: social psychological determinants. Soc Sci Med. 1987;24:351357. 94. Revenson TA. Compassionate stereotyping of elderly patients by physicians: revising the social contact hy-

pothesis. Psychol Aging. 1989;4: 230234. 95. Stern M, Arenson E. Childhood cancer stereotype: impact on adult perceptions of children. J Pediatr Psychol. 1989;14:593605. 96. Stern M, Moritzen SK, Carmel S, Olexa-Andrews M. The prematurity stereotype in Israeli health care providers. Med Educ. 2001;35:129133. 97. Stern M, Ross S, Bielass M. Medical students perceptions of children: modifying a childhood cancer stereotype. J Pediatr Psychol. 1991;16:2738. 98. Kelly CE. Bringing homophobia out of the closet: antigay bias within the patient-physician relationship. Pharos. 1992;55:28. 99. Kelly JA, St. Lawrence JS, Smith S, Hood HV, Cook DJ. Medical students attitudes toward AIDS and homosexual patients. J Med Educ. 1987;62: 549556. 100. Kelly JA, St. Lawrence JS, Smith S Jr., Hood HV, Cook DJ. Stigmatization of AIDS patients by physicians. Am J Public Health. 1987;77:789791. 101. Seils DM, Friedman JY, Schulman KA. Sex differences in the referral process for invasive cardiac procedures. J Am Med Womens Assoc. 2001;56: 151154, 160. 102. Ryan CS, Robinson DR, Hausmann LR. Stereotyping among providers and consumers of public mental health services: the role of perceived group variability. Behav Modif. 2001;25: 406442. 103. Ganong LH, Coleman M, Riley C. Nursing students stereotypes of married and unmarried pregnant clients. Res Nurs Health. 1988;11:333342. 104. Bowler IM. Stereotypes of women of Asian descent in midwifery: some evidence. Midwifery. 1993;9:716. 105. Thomson A. The use of stereotypes in the provision of midwifery care. Midwifery. 1993;9:12. 106. Porter JR, Beuf AH. The effect of a racially consonant medical context on adjustment of African-American patients to physical disability. Med Anthropol. 1994;16:116. 107. Schulman KA, Berlin JA, Harless W, et al. The effect of race and sex on physicians recommendations for cardiac catheterization. N Engl J Med. 1999; 340:618626. 108. Rathore SS, Lenert LA, Weinfurt KP, et al. The effects of patient sex and race on medical students ratings of quality of life. Am J Med. 2000;108: 561566. 109. Macrae CN, Bodenhausen GV. Social cognition: thinking categorically about others. Annu Rev Psychol. 2000; 51:93120.

254 | Racial/Ethnic Bias and Health | Peer Reviewed | van Ryn and Fu

American Journal of Public Health | February 2003, Vol 93, No. 2

 RACIAL/ETHNIC BIAS AND HEALTH

110. Duncan B. Differential social perception and attribution. J Pers Soc Psychol. 1976;34:2237. 111. Darley JG. A hypothesis confirming bias in labeling effects. In: Stangor C, ed. Sterotypes and Prejudice. Ann Arbor, Mich: Psychology Press; [[9]] 2000. 112. Lepore L, Brown R. Category and stereotype activation: is prejudice inevitable? J Pers Soc Psychol. 1997;72: 275287. 113. Sagar H, Schofield J. Racial and behavioral cues in black and white childrens perceptions of ambiguously aggressive acts. J Pers Soc Psychol. 1980; 39:590598. 114. Locksley A, Hepburn C, Ortiz V. Social stereotypes and judgements of individuals: an instance of the base-rate fallacy. J Exp Soc Psychol. 1982;18: 2342. 115. Darley JG. A hypothesis confirming bias in labeling effects. J Pers Soc Psychol. 1983;44:2022. 116. Burk JP, Sher KJ. Labeling the child of an alcoholic: negative stereotyping by mental health professionals and peers. J Stud Alcohol. 1990;51: 156163. 117. Martin TW. White therapists differing perceptions of black and white adolescents. Adolescence. 1993;28: 281289. 118. Bogart LM, Kelly JA, Catz SL, Sosman JM. Impact of medical and nonmedical factors on physician decision making for HIV/AIDS antiretroviral treatment. J Acquir Immune Defic Syndr. 2000;23:396404. 119. Bogart LM, Catz SL, Kelly JA, Benotsch EG. Factors influencing physicians judgments of adherence and treatment decisions for patients with HIV disease. Med Decis Making. 2001; 21:2836. 120. van Ryn M, Hannan E, Burke J, Besculides M. An Examination of Factors Associated With Physician Recommendation for Revascularization. Washington, DC: American Public Health Association; 1999. 121. Balsa AI, McGuire TG. Statistical discrimination in health care. J Health Econ. 2001;20:881907. 122. Chin MH, Humikowski CA. When is risk stratification by race or ethnicity justified in medical care? Acad Med. 2002;77:202208. 123. McKinlay JB, Potter DA, Feldman HA. Non-medical influences on medical decision-making. Soc Sci Med. 1996;42: 769776. 124. 10 Years of Reporting: Child Maltreatment, 1999. Washington, DC: US Dept of Health and Human Services; 1999.

125. Chen M, Bargh JA. Nonconscious behavioral confirmation processes: the self-fulfilling consequences of automatic stereotype activation processes. J Exp Soc Psychol. 1997;33:541560. 126. Word C, Zanna M, Cooper J. The nonverbal mediation of self-fulfilling prophesies in interracial interaction. J Exp Soc Psychol. 1974;10:109120. 127. Jussim L. Self-fulfilling prophecies: a theoretical and integrative review. Psychol Rev. 1986;93:429435. 128. Kaplan SH, Gandek B, Greenfield S, Rogers W, Ware JE. Patient and visit characteristics related to physicians participatory decision-making style: results from the Medical Outcomes Study. Med Care. 1995;33:11761187. 129. Cooper-Patrick L, Gallo JJ, Gonzales JJ, et al. Race, gender, and partnership in the patient-physician relationship. JAMA. 1999;282:583589. 130. [[10]] Roter DL, Moira PhD; Putnam, Samuel M. MD; Lipkin, Mack Jr, MD; Stiles, William PhD; Inui, Thomas S. MD. Communication patterns of primary care physicians. JAMA. 1997;277: 350356. 131. Epstein AM, Taylor WC, Seage GR. Effects of patients socioeconomic status and physicians training and practice on patient-doctor communication. Am J Med. 1985;78:101106. 132. Epstein A, Ayanian J. Racial disparities in medical care. N Engl J Med. 2001;344:14711473. 133. Curtis JR, Patrick DL, Caldwell E, Greenlee H, Collier AC. The quality of patient-doctor communication about end-of-life care: a study of patients with advanced AIDS and their primary care clinicians. AIDS. 1999;13:11231131. 134. Hooper EM, Comstock LM, Goodwin JM, Goodwin JS. Patient characteristics that influence physician behavior. Med Care. 1982;20:630638. 135. Wissow LS, Roter DL, Wilson ME. Pediatrician interview style and mothers disclosure of psychosocial issues. Pediatrics. 1994;93:289295. 136. Widdershoven GA. The doctor-patient relationship as a Gadamerian dialogue: a response to Arnason. Med Health Care Philos. 2000;3:2527. 137. Kaplan SH, Greenfield S, Ware JE Jr. Assessing the effects of physicianpatient interactions on the outcomes of chronic disease. Med Care. 1989; 27(suppl 3):S110S127. 138. Greenfield S, Kaplan SH, Ware JE Jr., Yano EM, Frank HJ. Patients participation in medical care: effects on blood sugar control and quality of life in diabetes. J Gen Intern Med. 1988;3: 448457. 139. Greenfield S, Kaplan S, Ware JE Jr.

Expanding patient involvement in care: effects on patient outcomes. Ann Intern Med. 1985;102:520528. 140. Borum ML, Lynn J, Zhong Z. The effects of patient race on outcomes in seriously ill patients in SUPPORT: an overview of economic impact, medical intervention, and end-of-life decisions. J Am Geriatr Soc. 2000;48(suppl 5): S194S198. 141. Mouton C, Teno JM, Mor V, Piette J. Communication of preferences for care among human immunodeficiency virus-infected patients: barriers to informed decisions? Arch Fam Med. 1997; 6:342347. 142. Haas JS, Weissman JS, Cleary PD, Goldberg J, Gatsonis C, Seage GR. Discussion of preferences for life-sustaining care by persons with AIDS: predictors of failure in patient-physician communication. Arch Intern Med. 1993;153: 12411248. 143. Doescher MP, Saver BG. Physicians advice to quit smoking: the glass remains half empty. J Fam Pract. 2000; 49:543547. 144. Taira DA, Safran DG, Seto TB, Rogers WH, Tarlov AR. The relationship between patient income and physician discussion of health risk behaviors. JAMA. 1997;278:14121417. 145. OMalley MS, Earp JA, Harris RP. Race and mammography use in two North Carolina counties. Am J Public Health. 1997;87:782786. 146. OMalley MS, Earp JA, Hawley ST, Schell MJ, Mathews HF, Mitchell J. The association of race/ethnicity, socioeconomic status, and physician recommendation for mammography: who gets the message about breast cancer screening? Am J Public Health. 2001;91:4954. 147. Kogan MD, Kotelchuck M, Alexander GR, Johnson WE. Racial disparities in reported prenatal care advice from health care providers. Am J Public Health. 1994;84:8288. 148. Krupat E, Irish JT, Kasten LE, et al. Patient assertiveness and physician decision-making among older breast cancer patients. Soc Sci Med. 1999;49: 449457. 149. Betancourt L, Fischer R, Giannetta J, Malmud E, Brodsky NL, Hurt H. Problem-solving ability of inner-city children with and without in utero cocaine exposure. J Dev Behav Pediatr. 1999;20:418424. 150. Green AR, Carrillo JE, Betancourt JR. Why the disease-based model of medicine fails our patients. West J Med. 2002;176:141143. 151. Green AR, Betancourt JR, Carrillo JE. Integrating social factors into crosscultural medical education. Acad Med. 2002;77:193197.

February 2003, Vol 93, No. 2 | American Journal of Public Health

van Ryn and Fu | Peer Reviewed | Racial/Ethnic Bias and Health | 255

At the Intersection of Health, Health Care and Policy Cite this article as: David R. Williams and Pamela Braboy Jackson Social Sources Of Racial Disparities In Health Health Affairs, 24, no.2 (2005):325-334 doi: 10.1377/hlthaff.24.2.325

The online version of this article, along with updated information and services, is available at: http://content.healthaffairs.org/content/24/2/325.full.html

For Reprints, Links & Permissions: http://healthaffairs.org/1340_reprints.php E-mail Alerts : http://content.healthaffairs.org/subscriptions/etoc.dtl To Subscribe: http://content.healthaffairs.org/subscriptions/online.shtml

Health Affairs is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600, Bethesda, MD 20814-6133. Copyright 2005 by Project HOPE - The People-to-People Health Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of Health Affairs may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical, including photocopying or by information storage or retrieval systems, without prior written permission from the Publisher. All rights reserved.

Not for commercial use or unauthorized distribution

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

S o c i a l

S o urc e s

Social Sources Of Racial Disparities In Health

Policies in societal domains, far removed from traditional health policy, can have decisive consequences for health.
by David R. Williams and Pamela Braboy Jackson
ABSTRACT: Racial disparities in mortality over time reflect divergent pathways to the current large racial disparities in health. The residential concentration of African Americans is high and distinctive, and the related inequities in neighborhood environments, socioeconomic circumstances, and medical care are important factors in initiating and maintaining racial disparities in health. Efforts are needed to identify and maximize health-enhancing resources that may reduce some of the negative effects of psychosocial factors on health. Health and health disparities are embedded in larger historical, geographic, sociocultural, economic, and political contexts. Changes in a broad range of public policies are likely to be central to effectively addressing racial disparities.

ac i a l d i s pa r i t i e s i n h e a lt h in the United States are substantial. The overall death rate for blacks today is comparable to the rate for whites thirty years ago, with about 100,000 blacks dying each year who would not die if the death rates were equivalent.1 This paper outlines factors in the social environment that can initiate and sustain racial disparities in health. Race is a marker for differential exposure to multiple disease-producing social factors. Thus, racial disparities in health should be understood not only in terms of individual characteristics but also in light of patterned racial inequalities in exposure to societal risks and resources. We illustrate some of these social processes by examining racial differences in mortality from 1950 to 2000 for five causes of death that reveal divergent pathways to current health disparities. Three of these causes of deathhomicide, heart disease, and cancershow wide disparities between black and white populations; two of these causespneumonia and flu, and suicideshow virtually no disparities. Data are available for blacks and whites for the 19502002 time period only. We present both absolute (black-white differences) and relative (black-white ratios) indicators of disparity.

David Williams (wildavid@umich.edu) is the Harold W. Cruse Collegiate Professor of Sociology, a professor of epidemiology, and a senior research scientist at the Institute for Social Research, University of Michigan, in Ann Arbor. Pamela Jackson is an associate professor of sociology at Indiana University in Bloomington.

H E A L T H A F F A I R S ~ Vo l u m e 2 4 , N u m b e r 2
DOI 10.1377/hlthaff.24.2.325 2005 Project HOPEThe People-to-People Health Foundation, Inc.

325

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

O r i gi ns

O f

Dis pa r i t i e s

Persistent Racial Disparities In Health

n Homicide. Exhibit 1 presents national trend data for black-white disparities in homicide, heart disease, and cancer. The homicide rate in 2000 was almost six times greater for African Americans than it was for whites. However, homicide deaths for blacks were almost 30 percent lower in 2000 than in 1950, and the racial gap in homicide death rates, both absolutely and relatively, was smaller in 2000 than in 1950. Homicide makes a small contribution to racial differences in mortality. It is the fifteenth leading U.S. cause of death and is responsible for about 17,000 deaths each year. In contrast, the annual death toll for the three leading causes of death heart disease (700,000), cancer (550,000), and stroke (160,000)are markedly larger. These illnesses and related chronic conditions, such as hypertension, diabetes, and obesity, are the key contributors to excess levels of ill health, premature mortality, and disability among blacks. Heart disease, for example, is the leading U.S. cause of disability and years of life lost for both men and women. n Heart disease. Death rates from coronary heart disease were comparable for blacks and whites in 1950, but by 2000, blacks had a death rate that was 30 percent higher than that for whites (Exhibit 1). Death rates from heart disease declined markedly from 1950 to 2000 for both racial groups, but because the decline for whites (57 percent) was more rapid than for blacks (45 percent), both the relative and absolute racial differences were larger in 2000 than in 1950. n Cancer. Blacks moved from having a lower cancer death rate than whites in 1950 to having a rate that was 30 percent higher in 2000. Cancer death rates for whites have been relatively stable over time, with the mortality rate in 2000 being al-

EXHIBIT 1 Age-Adjusted Death Rates For Blacks And Whites For Three Causes Of Death, And Racial Disparities, 19502000
Cause
Homicide White Black Difference Ratio Heart disease White Black Difference Ratio Cancer White Black Difference Ratio

1950
2.6 28.3 25.7 10.9 584.8 586.7 1.9 1.0 194.6 176.4 18.2 0.9

1960
2.7 26.0 23.3 9.6 559.0 548.3 10.7 1.0 193.1 199.1 6.0 1.0

1970
4.7 44.0 39.3 9.4 492.2 512.0 19.8 1.0 196.7 225.3 28.6 1.2

1980
6.7 39.0 32.3 5.8 409.4 455.3 45.9 1.1 204.2 256.4 52.2 1.3

1990
5.5 36.3 30.8 6.6 317.0 391.5 74.5 1.2 211.6 279.5 67.9 1.3

2000
3.6 20.5 16.9 5.7 253.4 324.8 71.4 1.3 197.2 248.5 51.3 1.3

SOURCE: National Center for Health Statistics, Health, United States, 2003. NOTES: Deaths per 100,000 population. Difference is calculated as black death rates minus white death rates for each cause of death. Ratio refers to the ratio of black deaths to white deaths.

326

M a r c h /A p r i l 2 0 0 5

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

S o c i a l

S o urc e s

most identical to the rate in 1950. In contrast, cancer mortality for blacks has been increasing, with the rate in 2000 being 40 percent higher than in 1950. Over time, lung and ovarian cancer death rates increased for both racial groups, while mortality from colorectal, breast, and prostate cancer markedly increased for blacks but was stable or declined for whites.2

Understanding Racial Differences In Health

Racial differences in socioeconomic status, neighborhood residential conditions, and medical care are important contributors to racial differences in disease. n Socioeconomic status. Whether measured by income, education, or occupation, socioeconomic status (SES) is a strong predictor of variations in health.3 Americans with low SES have levels of illness in their thirties and forties that are not seen in groups with higher SES until three decades of age later.4 All of the indicators of SES are strongly patterned by race, such that racial differences in SES contribute to racial differences in health. Moreover, the differences in health by SES within each racial group are often larger than the overall racial differences in health. Education. Among adults ages 2544, homicide rates are strongly patterned by education.5 The homicide rate for black males who have not completed high school is more than five times that of black males with some college education or more. Similarly, there is a ninefold difference in homicide rates by education for white males, a fourfold difference for black females, and a sixfold difference for white females. At the same time, large racial differences in homicide persist when blacks and whites are compared at similar levels of education. For example, the homicide death rate for African American men with at least some college education is eleven times that of their similarly educated white peers. Strikingly, the homicide rate of black males in the highest education category exceeds that of white males in the lowest education group. Income. Income also plays a role in understanding racial differences in coronary heart disease and cancer mortality. For example, death rates from heart disease are two to three times higher among low-income blacks and whites than among their middle-income peers.6 In addition, for both males and females at every level of income, blacks have higher coronary heart disease death rates than whites. Mortality from heart disease among low- and middle-income black women is 65 percent and 50 percent higher, respectively, than for comparable white women. Health practices. Another pathway underlying the association between race and chronic diseases is the patterning of health practices by race and socioeconomic status.7 Dietary behavior, physical activity, tobacco use, and alcohol abuse are important risk factors for chronic diseases such as coronary heart disease and cancer. Moreover, changes in these health practices over time are patterned by social status. Disadvantaged racial groups and those with low SES are less likely to reduce high-risk behavior or to initiate new health-enhancing practices. For example, people with high SES have been markedly more likely to quit cigarette smoking

H E A L T H A F F A I R S ~ Vo l u m e 2 4 , N u m b e r 2

327

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

O r i gi ns

O f

Dis pa r i t i e s

over the past several decades compared with their lower-SES counterparts. They also have greater health knowledge, are more receptive to new health information, and have greater resources to take advantage of health-enhancing opportunities than their low-SES peers.8 Stress. Exposure to psychosocial stressors may be another pathway linking SES and race to health. Chronic exposure to stress is associated with altered physiological functioning, which may increase risks for a broad range of health conditions.9 People of disadvantaged social status tend to report elevated levels of stress and may be more vulnerable to the negative effects of stressors. In addition, the subjective experience of discrimination is a neglected stressor that can adversely affect the health of African Americans.10 Reports of discrimination are positively related to SES among blacks and may contribute to the elevated risk of disease that is sometimes observed among middle-class blacks. n Residential segregation. The persistence of racial differences in health after individual differences in SES are accounted for may reflect the role that residential segregation and neighborhood quality can play in racial disparities in health.11 Because of segregation, middle-class blacks live in poorer areas than whites of similar economic status, and poor whites live in much better neighborhoods than poor blacks. Other U.S. racial/ethnic minority groups are less segregated than blacks, and although residential segregation is inversely related to income for Latinos and Asians, the segregation of African Americans is high at all levels of income.12 The most affluent African Americans (annual incomes over $50,000) experience higher levels of residential segregation than the poorest Latinos and Asians (incomes under $15,000). Segregation is a neglected but enduring legacy of racism in the United States. Instructively, blacks manifest a higher preference for residing in integrated areas than any other group.13 Impact on income. Residential segregation is a central mechanism by which racial economic inequality has been created and reinforced in the United States.14 It is a key determinant of the observed racial differences in SES because it determines access to education and employment opportunities. For example, an empirical study of the effects of segregation on young African Americans making the transition from school to work found that the elimination of residential segregation would completely erase black-white differences in earnings, high school graduation rates, and employment and would reduce racial differences in single motherhood by two-thirds.15 Violence. In addition, segregation creates health-damaging conditions in both the physical and social environments. Research has identified specific pathways by which neighborhood conditions can encourage violence and create racial differences in homicide.16 Because of its restriction of educational and employment opportunities, residential segregation creates areas with high rates of concentrated poverty and small pools of employable and stably employed males. In turn, high male unemployment and low wage rates for males are associated with high

328

M a r c h /A p r i l 2 0 0 5

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

S o c i a l

S o urc e s

rates of out-of-wedlock births and female-headed households.17 Single-parent households are associated with lower levels of social control and supervision of young males, which, in turn, lead to elevated rates of violent behavior.18 The association between family and neighborhood factors and the risk of violent crime is identical for blacks and whites.19 However, because of residential segregation, blacks are more exposed to these conditions than whites. In the 171 largest U.S. cities, there is not even one in which whites live in socioeconomic conditions that are comparable to those of blacks. As Robert Sampson and William J. Wilson concluded, The worst urban context in which whites reside is considerably better than the average context of black communities.20 Links to disease. Independent of individual SES, factors linked to poor residential environments make an incremental contribution to the risk of a broad range of health outcomes, including heart disease and cancer.21 Multiple characteristics of neighborhoods are conducive to healthy or unhealthy behavioral practices. The perception of neighborhood safety is positively associated with physical exercise, and this association is larger for minority group members than for whites.22 Neighborhoods also differ in the existence and quality of recreational facilities and open, green spaces. The availability and cost of healthy products in grocery stores also vary across residential areas, and the availability of nutritious foods is positively associated with their consumption.23 Also, both the tobacco and alcohol industries heavily market their products to poor minority communities.24 n Medical care. Racial differences in SES contribute to reduced levels of health insurance coverage for African Americans, and limited access to medical care plays a role in racial differences in disease. Moreover, the black-white gap in access to and use of health services did not narrow between 1977 and 1996.25 Also, the racial gap in unemployment, median income, and poverty remained large and fairly stable throughout this period.26 Links to homicide. Medical care is a contributor to homicide and the racial disparities in homicide. Rates of violent crime have increased over time, but homicide rates have been fairly stable. The lethality of violent assaults has declined as advances in emergency medicine and trauma care have reduced the likelihood that a violent assault will end as a homicide.27 However, black assault victims are less likely than their white peers to receive timely emergency transportation and subsequent high-quality medical care.28 The Institute of Medicine (IOM) report Unequal Treatment also found that blacks receive poorer-quality emergency room care than whites.29 It revealed systematic and pervasive racial differences in the quality of care provided across a broad range of medical conditions, including heart disease and cancer. Racial differences in the quality and intensity of treatment persist after SES, insurance status, patient preference, severity of disease, and coexisting medical conditions are taken into account. Links to cancer mortality. African Americans are less likely than whites to receive preventive, screening, diagnostic, treatment, and rehabilitation services for can-

H E A L T H A F F A I R S ~ Vo l u m e 2 4 , N u m b e r 2

329

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

O r i gi ns

O f

Dis pa r i t i e s

cer, and this probably contributes to racial differences in cancer mortality.30 Although blacks have higher cancer mortality than whites, the annual incidence (new cases) of cancer is lower for black than for white women. However, when compared at the same stage of cancer diagnosis, black women have poorer survival rates than their white counterparts. Blacks also are more likely than whites to experience delays in the receipt of care after a positive screening test, delays in the initiation of treatment after a biopsy, the receipt of care from inadequately trained providers, and limited access to appropriate follow-up and rehabilitation services. Impact of segregation. Black Medicare patients are more likely than white ones to reside in areas where medical procedure rates and the quality of care are low.31 In addition, a small group of physicians, who are more likely to practice in lowincome areas, provide most of the care to black patients. These providers are less likely than other physicians to be board certified and less able to provide highquality care and referrals to specialty care.32 Also, pharmacies in segregated neighborhoods are less likely to have adequate medication supplies, and hospitals in these neighborhoods are more likely to close.33 Disentangling the relative importance of the complex causal processes that lead to disparities in disease is challenging, but renewed efforts are needed to identify key points of intervention.

Where There Are No Disparities

Flu and pneumonia. Examining racial disparities over time reveals that success stories do exist. Flu and pneumonia is one such story. It is the seventh leading cause of death and is responsible for more than 65,000 deaths annually. However, both the absolute and the relative racial differences for deaths from flu and pneumonia were minimal in 2000 (Exhibit 2). In contrast, large racial differences existed in 1950, with black mortality being 70 percent higher than that of whites. Over time,
n

EXHIBIT 2 Age-Adjusted Death Rates For Blacks And Whites For Flu And Pneumonia And For Suicide, And Racial Disparities, 19502000
Cause
Flu and pneumonia White Black Difference Ratio Suicide White Black Difference Ratio

1950
44.8 76.7 31.9 1.7 13.9 4.5 9.4 0.3

1960
50.4 81.1 30.7 1.6 13.1 5.0 8.1 0.4

1970
39.8 57.2 17.4 1.4 13.8 6.2 7.6 0.5

1980
30.9 34.4 3.5 1.1 13.0 6.5 6.5 0.5

1990
36.4 39.4 3.0 1.1 13.4 7.1 6.3 0.5

2000
23.5 25.6 2.1 1.1 11.3 5.5 5.8 0.5

SOURCE: National Center for Health Statistics, Health, United States, 2003. NOTES: Deaths per 100,000 population. Difference is calculated as black death rates minus white death rates for each cause of death. Ratio refers to the ratio of black deaths to white deaths.

330

M a r c h /A p r i l 2 0 0 5

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

S o c i a l

S o urc e s

striking declines are evident for both races, with larger declines for blacks than for whites. Flu and pneumonia is an acute respiratory illness that can be prevented by vaccination and treated by antiviral medicines. It differs from the major chronic illnesses that typically have a large behavioral component, are long term in development, and have symptoms that are not always readily evident. The virtual elimination of this disparity suggests that the application of a widely diffused technology (facilitated by Medicare and Medicaid), in which social variations in motivation, knowledge, and resources play a small role, can eliminate a large disparity in health. n Suicide. Suicide is a success story of another sort. Suicide is the eleventh leading U.S. cause of death (30,000 deaths annually). Suicide rates for both racial groups have been fairly stable over time, with a slight decline for whites and a slight increase for blacks in recent years. However, black suicide death rates have been consistently lower than those of whites. The suicide data are consistent with national data, which indicate that the prevalence of major psychiatric disorders are lower for blacks than for whites.34 Suicide is an example of a health condition for which the socially disadvantaged group does not have elevated rates. This pattern highlights the importance of attending to protective resources that may improve health and protect vulnerable populations from at least some of the negative effects of environmental exposures. For example, high levels of self-esteem and religious involvement are potential contributors to blacks better suicide and mental health profile.

Policy Implications
Persisting disparities in health violate widely shared U.S. norms of equality of opportunity and the dignity of each person. Eliminating health disparities is also important for the overall well-being of the entire U.S. society. First, diseases that are initially more prevalent in disadvantaged geographic areas eventually diffuse and spread into adjacent affluent communities.35 Second, the illnesses and disabilities associated with racial disparities limit the productive capacities and output of adults in their prime working years. This can negatively affect productivity at the local and national levels and can lead to declines in tax revenues and increased costs of social services.36 Thus, effectively addressing racial disparities in health likely requires addressing distal social policies and arrangements that create the disparities in the first place.37 n Addressing segregation. Racial residential segregation is one of the primary causes of U.S. racial inequality, and although discrimination in the sale and rental of housing was made illegal in 1968, considerable evidence suggests that housing discrimination persists.38 Current public preferences and opportunities for the enforcement of equal opportunity statutes suggest that U.S. residential patterns are unlikely to change in the foreseeable future. Thus, the elimination of the negative effects of segregation on SES and health may require a major infusion of economic capital to improve the social, physical, and economic infrastructure of disadvantaged communities.39 Such investment could improve the economic circumstances and

H E A L T H A F F A I R S ~ Vo l u m e 2 4 , N u m b e r 2

331

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

O r i gi ns

O f

Dis pa r i t i e s

productivity of African American families and communities and have spillover benefits for health. n Narrowing the income gap. Over the past fifty years, changes in the blackwhite gap in income have been associated with parallel changes in the black-white gap in health. Between 1968 and 1978, in tandem with the narrowing of racial inequality attributable to the economic gains of the civil rights movement, black men and women experienced a larger decline in mortality than their white counterparts on both a percentage and absolute basis.40 However, as blacks median household income fell relative to that of whites from its 1978 level throughout the 1980s, the black-white gap in adult and infant mortality widened between 1980 and 1991.41 At the same time, although it is generally recognized that policies that disproportionately assist the disadvantaged are desirable, it is unclear whether those policies are best implemented at the federal, state, or local level and what optimal forms such policies should take.42 Greater attention needs to be given to rigorously evaluating the extent to which policies in multiple sectors of society have consequences for health and health disparities, so that we can have an improved understanding of the conditions under which specific policy initiatives are more or less likely to achieve desirable results. n Improving medical care. Improving access to medical care for vulnerable populations, especially for preventive services, can play a role in reducing racial disparities in health. According to a 2000 study, only half of physicians or fewer routinely counsel patients who smoke about smoking cessation, treat patients with elevated blood lipids for this condition, treat hypertensive patients for their high blood pressure, and routinely screen patients for diabetes.43 One way to improve medical care might be to provide physicians with incentives to ensure that they use evidence-based guidelines for treatment and follow national standards of care. Also, given that underrepresented minority providers are more likely than others to practice in underserved areas, increasing the numbers of blacks in the health professions is likely to be an effective strategy in improving access to care.44 n Rethinking health policy. There is a need to rethink what constitutes health policy. Given the broad social determinants of health, policies in societal domains far removed from traditional health policy can have decisive consequences for individual and population health. A recent federal report outlines an ambitious agenda to eliminate disparities in cancer.45 Recognizing that the determinants of cancer disparities transcend its scope, the U.S. Department of Health and Human Services (HHS) called for the creation of a Federal Leadership Council, led by HHS, that would leverage governmentwide resources to address disparities. This proposed council would include all federal departments that have policies that can affect health and health disparities, including the Departments of Labor, Education, Defense, Justice, Energy, and Transportation. Similar coordination is necessary at the regional and local levels. There are political, professional, and organizational barriers to such intersectoral collaboration, but multiple strategies to address them have

332

M a r c h /A p r i l 2 0 0 5

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

S o c i a l

S o urc e s

been identified, including the need to establish a permanent locus for intersectoral activity regarding health.46 Although much is yet to be learned about the specific pathways by which the social environment creates disease, much progress can be made toward eliminating disparities by acting on current knowledge.
Research for this paper was supported by the John D. and Catherine T. MacArthur Foundation Research Network on Socioeconomic Status and Health and the Robert Wood Johnson Foundation. The authors thank Car Nosel, Trisha Matelski, and Natalie Moran for assistance with preparation of the manuscript. NOTES
1. R.S. Levine et al., Black-White Inequalities in Mortality and Life Expectancy, 19331999: Implications for Healthy People 2010, Public Health Reports 116, no. 5 (2001): 474483; and National Center for Health Statistics, Health, United States, 2003 (Hyattsville, Md.: U.S. Government Printing Office, 2003). T.A. Piffath et al., Ethnic Differences in Cancer Mortality Trends in the U.S., 19501992, Ethnicity and Health 6, no. 2 (2001): 105119. M. Marmot, The Influence of Income on Health: Views of an Epidemiologist, Health Affairs 21, no. 2 (2002): 3146; and N.E. Adler and K. Newman, Socioeconomic Disparities in Health: Pathways and Policies, Health Affairs 21, no. 2 (2002): 6076. J.S. House et al., The Social Stratification of Aging and Health, Journal of Health and Social Behavior 35, no. 3 (1994): 213234. E. Pamuk et al., Health, United States, 1998, with Socioeconomic Status and Health Chartbook (Hyattsville, Md.: NCHS, 1998). Ibid. R. Cooper et al., Trends and Disparities in Coronary Heart Disease, Stroke, and Other Cardiovascular Diseases in the United States: Findings of the National Conference on Cardiovascular Disease Prevention, Circulation 102, no. 25 (2000): 31373147. B.G. Link and J. Phelan, Social Conditions as Fundamental Causes of Disease, Journal of Health and Social Behavior, Extra Issue (1995): 8094. B.S. McEwen, Protective and Damaging Effects of Stress Mediators, New England Journal of Medicine 338, no. 3 (1998): 171179. D.R. Williams, H. Neighbors, and J.S. Jackson, Racial/Ethnic Discrimination and Health: Findings from Community Studies, American Journal of Public Health 93, no. 2 (2003): 200208. D.R. Williams and C. Collins, Racial Residential Segregation: A Fundamental Cause of Racial Disparities in Health, Public Health Reports 116, no. 5 (2001): 404416. D.S. Massey, Segregation and Stratification: A Biosocial Perspective, Du Bois Review 1, no. 1 (2004): 725. Ibid. D.S. Massey and N. Denton, American Apartheid: Segregation and the Making of the Underclass (Cambridge, Mass.: Harvard University Press, 1993). D.M. Cutler, E.L. Glaeser, and J.L. Vigdor, Are Ghettos Good or Bad? Quarterly Journal of Economics 112, no. 3 (1997): 827872. R.J. Sampson and W. Wilson, Toward a Theory of Race, Crime, and Urban Inequality, in Crime and Inequality, ed. J. Hagan and R.D. Peterson (Stanford, Calif.: Stanford University Press, 1995), 3754. M. Testa et al., Employment and Marriage among Inner-City Fathers, in The Ghetto Underclass, ed. W.J. Wilson (Newbury Park, Calif.: Sage, 1993), 96108. R.J. Sampson, Urban Black Violence: The Effect of Male Joblessness and Family Disruption, American Journal of Sociology 93, no. 2 (1987): 348382. Ibid. Sampson and Wilson, Toward a Theory of Race, 41. K.E. Pickett and M. Pearl, Multilevel Analyses of Neighborhood Socioeconomic Context and Health Outcomes: A Critical Review, Journal of Epidemiology and Community Health 55, no. 2 (2001): 111122. Neighborhood Safety and the Prevalence of Physical InactivitySelected States, 1996, Morbidity and Mor-

2. 3.

4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

H E A L T H A F F A I R S ~ Vo l u m e 2 4 , N u m b e r 2

333

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

O r i gi ns

O f

Dis pa r i t i e s

tality Weekly Report 48, no. 7 (1999): 143146. 23. A. Cheadle et al., Community-Level Comparisons between the Grocery Store Environment and Individual Dietary Practices, Preventive Medicine 20, no. 2 (1991): 250261. 24. D.J. Moore, J.D. Williams, and W.J. Qualls, Target Marketing of Tobacco and Alcohol-related Products to Ethnic Minority Groups in the United States, Ethnicity and Disease 6, nos. 12 (1996): 8398. 25. R.M. Weinick, S.H. Zuvekas, and J.W. Cohen, Racial and Ethnic Differences in Access to and Use of Health Care Services, 1977 to 1996, Medical Care Research and Review 57, Supp. 1 (2000): 3654. 26. Office of the President, The Annual Report of the Council of Economic Advisers (Washington: Office of the President, 1998). 27. A.R. Harris et al., Murder and Medicine: The Lethality of Criminal Assault, 19601999, Homicide Studies 6, no. 2 (2002): 128166. 28. P.J. Hanke and J.H. Gundlach, Damned on Arrival: A Preliminary Study of the Relationship between Homicide, Emergency Medical Care, and Race, Journal of Criminal Justice 23, no. 4 (1995): 313323. 29. B.D. Smedley, A.Y. Stith, and A.R Nelson, eds., Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Washington: National Academies Press, 2002). 30. Department of Health and Human Services, Trans-HHS Cancer Health Disparities Progress Review Group, Making Cancer Health Disparities History (Washington: DHHS, 2004). 31. K. Baicker et al., Who You Are and Where You Live: How Race and Geography Affect the Treatment of Medicare Beneficiaries, Health Affairs, 7 October 2004, content.healthaffairs.org/cgi/content/abstract/ hlthaff.var.33 (15 December 2004). 32. P.B. Bach et al., Primary Care Physicians Who Treat Blacks and Whites, New England Journal of Medicine 351, no. 6 (2004): 575584. 33. Williams and Collins, Racial Residential Segregation; and T.C. Buchmueller, M. Jacobson, and C. Wold, How Far to the Hospital? The Effect of Hospital Closures on Access to Care, NBER Working Paper no. w10700 (Cambridge, Mass.: National Bureau of Economic Research, 2004). 34. R.C. Kessler et al., Lifetime and Twelve-Month Prevalence of DSM-III-R Psychiatric Disorders in the United States: Results from the National Comorbidity Survey, Archives of General Psychiatry 51, no. 1 (1994): 819. 35. R. Wallace, D. Wallace, and R.G. Wallace, Coronary Heart Disease, Chronic Inflammation, and Pathogenic Social Hierarchy: A Biological Limit to Possible Reductions in Morbidity and Mortality, Journal of the National Medical Association 96, no. 5 (2004): 609619. 36. J. Bound et al., The Labor Market Consequences of Race Differences in Health, Milbank Quarterly 81, no. 3 (2003): 441473. 37. Link and Phelan, Social Conditions. 38. M. Fix and R.J. Struyk, Clear and Convincing Evidence: Measurement of Discrimination in America (Washington: Urban Institute Press, 1993). 39. D.R. Williams and C. Collins, Reparations: A Viable Strategy to Address the Enigma of African American Health, American Behavioral Scientist 47, no. 7 (2004): 9771000. 40. R.S. Cooper et al., Improved Mortality among U.S. Blacks, 19681978: The Role of Antiracist Struggle, International Journal of Health Services 11, no. 4 (1981): 511522. 41. Office of the President, The Annual Report of the Council of Economic Advisers; and Williams and Collins, Racial Residential Segregation. 42. A. Deaton, Policy Implications of the Gradient of Health and Wealth, Health Affairs 21, no. 2 (2002): 1330; and D. Mechanic, Disadvantage, Inequality, and Social Policy, Health Affairs 21, no. 2 (2002): 4859. 43. Cooper et al., Trends and Disparities in Coronary Heart Disease. 44. M. Komaromy et al., The Role of Black and Hispanic Physicians in Providing Health Care for Underserved Populations, New England Journal of Medicine 334, no. 20 (1996): 13051310. 45. DHHS, Making Cancer Health Disparities History. 46. S.L. Syme, B. Lefkowitz, and B.K. Krimgold, Incorporating Socioeconomic Factors into U.S. Health Policy: Addressing the Barriers, Health Affairs 21, no. 2 (2002): 113118.

334

M a r c h /A p r i l 2 0 0 5

Downloaded from content.healthaffairs.org by Health Affairs on October 26, 2011 by guest

undermines our quality of life: the heightened risk of heart disease, hypertension, diabetes, cancer, arthritis, gallbladder disease, trauma, gout, blindness, birth defects, and other aches, pains, and physical indignities too numerous to mention. What we are in danger of losing in the epidemic of obesity is not merely our health but our memory of health. Those Indian towns high in the Sierra Madre should remind the people of Sacaton--and all the rest of us as well--that it is still possible, even for a Pima, to be fit.!

together for a very long time," Bennett recalls. "My reaction as a stranger going in was: Gee, I think these people are really very friendly, very coperative. They seem to be interested in what we want to do, and they are willing to stick their arms out and let us take blood samples." He laughed. "Which is always a good sign." The little town in the Sierra Madre is home to the Mexican Pima, the southern remnants of a tribe that once stretched from present-day Arizona down to central Mexico. Like the Pima of the Gila River reservation, they are farmers, living in small clusters of wood-and-adobe rancheras among the pine trees, cultivating beans, corn, and potatoes in the valleys. On that first trip, the N.I.H. team examined no more than a few dozen Pima. Since then, the team has been back five or six times, staying for as many as ten days at a time. Two hundred and fifty of the mountain Pima have now been studied. They have been measured and weighed, their blood sugar has been checked, and their kidneys and eyes have been examined for signs of damage. Genetic samples have been taken and their metabolism has been monitored. The Mexican Pima, it turns out, eat a diet consisting almost entirely ofbeans, potatoes, and corn tortillas, with chicken perhaps once a month. They take in twenty-two hundred calories a day, which is slightly more than the Pima of Arizona do. But on the average each of them puts in twenty-three hours a week of moderate to hard physical labor, whereas the average Arizona Pima puts in two hours. The Mexican Pima's rates of diabetes are normal. They are slightly shorter than their American counterparts. In weight, there is no comparison: "I would say they are thin," Bennett says. "Thin. Certainly by American standards." There are, of course, a hundred reasons not to draw any great lessons from this. Subsistence farming is no way to make a living in America today, nor are twentythree hours ofhard physical labor feasible in a society where most people sit at a desk from nine to five. And even if the Arizona Pima wanted to return to the land, they couldn't. It has been more than a hundred years since the Gila River, which used to provide the tribe with fresh fish and with water for growing beans and squash, was diverted upstream for commercial farming. Yet there is value in the example of the Mexican Pima. People who work with the Pima of Arizona say that the biggest problem they have in trying to fight diabetes and obesity is fatalism--a sense among the tribe that nothing can be done, that the way things are is the way things have to be. It is possible to see in the attitudes of Americans toward weight loss the same creeping resignation. As the world grows fatter, and as one best-selling diet scheme after another inevitably fails, the idea that being slender is an attainable--or even an advisable--condition is slowly receding. Last month, when The New England Journal of Medicine published a study suggesting that the mortality costs of obesity had been overstated, the news was greeted with resounding relief, as if we were all somehow off the hook, as if the issue with obesity were only mortality and not the thousand ways in which being fat

weakness. The elegant solution is to find the place in the chain where we are losing water." The steps that take place in the brain when it receives the leptin message are known as the Ob pathway, and any number of these steps may lend themselves to pharmaceutical intervention. Using the Ob pathway to fight obesity represents a quantum leap beyond the kinds of diet drugs that have been available so far. Fenphen, the popular medication removed from the market last year because of serious side effects, was, by comparison, a relatively crude product, which worked indirectly to suppress appetite. Hoffmann-La Roche is working now on a drug called Xenical, a compound that blocks the absorption of dietary fat by the intestine. You can eat fat; you just don't keep as much of it in your system. The drug is safe and has shown real, if modest, success in helping chronically obese patients lose weight. It will probably be the next big diet drug. But no one is pretending that it has anywhere near the potential of, say, a drug that would resensitize your leptin receptors. Campfield talks about the next wave of drug therapy as the third leg of a threelegged stool--as the additional element that could finally make diet and exercise an easy and reliable way to lose weight. Wadden speaks of the new drugs as restoring sanity:"What I think will happen is that people on these medications will report that they are less responsive to their environment. They'll say that they are not as turned on by Wendy's or McDonald's. Food in America has become a recreational activity. It is divorced from nutritional need and hunger. We eat to kill time, to stimulate ourselves, to alter our mood. What these drugs may mean is that we're going to become less susceptible to these messages." In the past thirty years, the natural relationship between our bodies and our environment--a relation that was developed over thousands of years--has fallen out ofbalance. For people who cannot restore that natural balance themselves--who lack the discipline, the wherewithal, or, like the Pima, the genes--drugs could be a way of restoring it for them.

5.
Seven years ago, Peter Bennett, the epidemiologist who first stumbled on the Gila River Pima twenty-eight years earlier, led an N.I.H. expedition to Mexico's Sierra Madre Mountains. Their destination was a a tiny Indian community on the border of Sonora and Chihuahua, seven thousand feet above the desert. "I had known about their existence for at least fifteen years before that," Bennett says. "The problem was that I could never find anyone who knew much about them. In 1991, it just happened that we linked up with an investigator down in Mexico." The journey was a difficult one, but the Mexican government had just built a road linking Sonora and Chihuahua, so the team didn't have to make the final fifty- or sixty-mile trek on horseback. "They were clearly a group who have got along

This doesn't mean that diets can't work. In those instances in which dieters have the discipline and the will power to restrict their calories permanently, to get regular and vigorous exercise, and to fight the attempt by their own bodies to maintain their current weight, pounds can be lost. (There is also some evidence that if you can keep weight off for an extensive period--three years, say--a lower setpoint can be established.) Most people, though, don't have that kind of discipline, and even if they do have it the amount of weight that most dieters can expect to lose on a permanent basis may be limited by their setpoint range. The N.I.H. has a national six-year diabetes-prevention study going on right now, in which it is using a program of intensive, one-on-one counselling, dietary modification, and two and a half hours of exercise weekly to see if it can get overweight volunteers to lose seven per cent of their body weight. If that sounds like a modest goal, it should. "A lot of studies look at ten-per-cent weight loss," said Mary Hoskin, who is cordinating the section of the N.I.H. study involving the Pima. "But if you look at long-term weight loss nobody can maintain ten per cent. That's why we did seven." On the other hand, now that we're coming to understand the biology of weight gain, it is possible to conceive of diet drugs that would actually work. If your body sabotages your diet by lowering leptin levels as you lose weight, why not give extra leptin to people on diets? That's what a number of drug companies, including Amgen and Eli Lilly, are working on now. They are trying to develop a leptin or leptin-analogue pill that dieters could take to fool their bodies into thinking they're getting fatter when they're actually getting thinner. "It is very easy to lose weight," Jos Caro told me. "The difficult thing is to maintain your weight loss. The thinking is that people fail because their leptin goes down. Here is where replacement therapy with leptin or an Ob-protein analogue might prevent the relapse. It is a subtle and important concept. What it tells you is that leptin is not going to be a magic bullet that allows you to eat whatever you want. You have to initiate the weight loss. Then leptin comes in." Another idea, which the Hoffmann-La Roche company is exploring, is to focus on the problems obese people have with leptin. Just as Type II diabetics can become resistant to insulin, many overweight people may become resistant to leptin. So why not try to resensitize them? The idea is to find the leptin receptor in the brain and tinker with it to make it work as well in a fat person as it does in a thin person. (Drug companies have actually been pursuing the same strategy with the insulin receptors of diabetics.) Arthur Campfield, who heads the leptin project for Roche, likens the process by which leptin passes the signal about fat to the brain to a firemen's bucket brigade, where water is passed from hand to hand. "If you have all tall people, you can pass the bucket and it's very efficient,"he said. "But if two of the people in the chain are small children, then you're going to spill a lot of water and slow everything down. We want to take a tablet or a capsule that goes into your brain and puts a muscular person in the chain and overcomes that

the same mechanism may work in human beings. If you start to overeat, your fat cells will produce more leptin, so your body will do everything it can to get back to the setpoint. That's why after gaining a few pounds over the holiday season most of us soon return to our normal weight. But if you eat too little or exercise too much, the theory goes, the opposite happens: leptin levels fall. "This is probably the reason that virtually every weight-loss program known to man fails," Jos F. Caro, vice-president of endocrine research and clinical investigation at Eli Lilly & Company, told me. "You go to Weight Watchers. You start losing weight. You feel good. But then your fat cells stop producing leptin. Remember, leptin is the hormone that decreases appetite and increases energy expenditure, so just as you are trying to lose weight you lose the hormone that helps you lose weight." Obviously, our body's fat thermostat doesn't keep us at one weight all our adult lives. "There isn't a single setpoint for a human being or an animal," Thomas Wadden, the director of the Weight and Eating Disorders Clinic at the University of Pennsylvania, told me. "The body will regulate a stable weight but at very different levels, depending on food intake--quality of the diet, high fat versus low fat, high sweet versus low sweet--and depending on the amount of physical activity." It also seems to be a great deal easier to move the setpoint up than to move it down--which, if you think about the Pima, makes perfect sense. In their long history in the desert, those Pima who survived were the ones who were very good at gaining weight during times of plenty--very good, in other words, at overriding the leptin system at the high end. But there would have been no advantage for the ones who were good at losing weight in hard times. The same is probably true for the rest of us, albeit in a less dramatic form. In our evolutionary history, there was advantage in being able to store away whatever calorific windfalls came our way. To understand this interplay between genes and environment, imagine two women, both five feet five. The first might have a setpoint range of a hundred and ten to a hundred and fifty pounds; the second a range of a hundred and twenty-five to a hundred and eighty. The difference in the ranges of the two women is determined by their genes. Where they are in that range is determined by their life styles. Not long after leptin was discovered, researchers began testing obese people for the hormone, to see whether a fat person was fat because his body didn't produce enough leptin. They found the opposite: fat people had lots of leptin. Some of the researchers thought this meant that the leptin theory was wrong--that leptin didn't do what it was supposed to do. But some other scientists now think that as people get fatter and fatter, their bodies simply get less and less sensitive to leptin. The body still pumps out messages to the brain calling for the metabolism to speed up and the appetite to shrink, but the brain just doesn't respond to those messages with as much sensitivity as it did. This is probably why it is so much easier to gain weight than it is to lose it. The fatter you get, the less effective your own natural weight-control system becomes.

pounds. Clearly, the men who gained just nine pounds were the ones whose genes had given them the fastest possible metabolism--the ones who burn the most calories in daily living and are the least efficient at storing fat. These are people who have the easiest time staying thin. The men at the other end of the scale are closer to the Pima in physiology. Their obesity genes thriftily stored away as much of the thousand extra calories a day as possible. One of the key roles for genes appears to be in determining what obesity researchers refer to as setpoints. In the classic experiment in the field, researchers took a group of rats and made a series of lesions in the base of each rat's brain. As a result, the rats began overeating and ended up much more obese than normal rats. The first conclusion is plain: there is a kind of thermostat in the brain that governs appetite and weight, and if you change the setting on that thermostat appetite and weight will change accordingly. With that finding in mind, the researchers took a second step. They took those same brain-damaged rats and put them on a diet, severely limiting the amount of food they could eat. What happened? The rats didn't lose weight. In fact, after some initial fluctuations, they ended up at exactly the same weight as before. Only, this time, being unable to attain their new thermostat setting by eating, they reached it by becoming less active--by burning less energy. Two years ago, a group at Rockefeller University in New York published a landmark study essentially duplicating in human beings what had been done years ago in rats. They found that if you lose weight your body responds by starting to conserve energy: your metabolism slows down; your muscles seem to work more efficiently, burning fewer calories to do the same work. "Let's say you have two people, side by side, and these people have exactly the same body composition," Jules Hirsch, a member of the Rockefeller team, says. "They both weigh a hundred and thirty pounds. But there is one difference--the first person maintains his weight effortlessly, while the second person, who used to weigh two hundred pounds, is trying to maintain a lower weight. The second will need fifteen per cent fewer calories per day to do his work. He needs less oxygen and will burn less energy." The body of the second person is backpedalling furiously in response to all that lost weight. It is doing everything it can to gain it back. In response to weight gain, by contrast, the Rockefeller team found that the body speeds up metabolism and burns more calories during exercise. It tries to lose that extra weight. Human beings, like rats, seem to have a predetermined setpoint, a weight that their body will go to great lengths to maintain. One key player in this regulatory system may be a chemical called leptin--or, as it is sometimes known, Ob protein--whose discovery four years ago, by Jeff Friedman, of the Howard Hughes Medical Institute at Rockefeller University, prompted a flurry of headlines. In lab animals, leptin tells the brain to cut back on appetite, to speed up metabolism, and to burn stored fat. The theory is that

When I first arrived at Atkins' headquarters, two members of his staff took me on a quick tour of the facility, a vast medical center, where Atkins administers concoctions of his own creation to people suffering from a variety of disorders. Starting from the fifth floor, we went down to the third, and then from the third to the second, taking the elevator each time. It's a small point, but it did strike me as odd that I should be in the headquarters of the world's most popular weightloss expert and be taking the elevator one floor at a time. After watching Atkins' show, I was escorted out by his public-relations assistant. We were on the second floor. He pressed the elevator button, down. "Why don't we take the stairs?" I asked. It was just a suggestion. He looked at me and then at the series of closed doors along the corridor. Tentatively, he opened the second. "I think this is it," he said, and we headed down, first one flight and then another. At the base of the steps was a door. The P.R. man, a slender fellow in a beautiful Italian suit, peered through it: for the moment, he was utterly lost. We were in the basement. It seemed as if nobody had gone down those stairs in a long time.

4.
Why are the Pima so fat? The answer that diet books would give is that the Pima don't eat as well as they used to. But that's what is ultimately wrong with diet books. They talk as if food were the only cause of obesity and its only solution, and we know, from just looking at the Pima, that things are not that simple. The diet of the Pima is bad, but no worse than anyone else's diet. Exercise is also clearly part of the explanation for why obesity has become epidemic in recent years. Half as many Americans walk to work today as did twenty years ago. Over the same period, the number of calories burned by the average American every day has dropped by about two hundred and fifty. But this doesn't explain why obesity has hit the Pima so hard, either, since they don't seem to be any less active than the rest of us. The answer, of course, is that there is something beyond diet and exercise that influences obesity--that can make the consequences of a bad diet or of a lack of exercise much worse than they otherwise would be--and this is genetic inheritance. Claude Bouchard, a professor of social and preventive medicine at Laval University, in Quebec City, and one of the world's leading obesity specialists, estimates that we human beings probably carry several dozen genes that are directly related to our weight. "Some affect appetite, some affect satiety. Some affect metabolic rate, some affect the partitioning of excess energy in fat or lean tissue," he told me. "There are also reasons to believe that there are genes affecting physical-activity level." Bouchard did a study not long ago in which he took a group of men of similar height, weight, and life style and overfed them by a thousand calories a day, six days a week, for a hundred days. The average weight gain in the group was eighteen pounds. But the range was from nine to twenty-six

"I've stopped taking junk foods," George says. "I don't eat eggs. I don't eat bacon." "Then that's-- See there." Atkins' voice rose in exasperation. "What do you have for breakfast?" "I have skim milk, cereal, with banana." "That's three carbs!" Atkins couldn't believe that in this day and age people were still consuming fruit and skim milk. "That's how you are getting into trouble!... What you need to do, George, seriously, is get ahold of'New Diet Revolution' and just read what it says." Atkins took another call. This time, it was from Robert, forty-one years old, three hundred pounds, and possessed of a formidable Brooklyn accent. He was desperate to lose weight--up on a ledge and wanting Atkins to talk him down. "I really don't know anything about dieting," he said. "I'm getting a little discouraged." "It's really very easy," Atkins told him, switching artfully to the Socratic method. "Do you like meat?" "Yes." "Could you eat a steak?" "Yes." "All by itself, without any French fries?" "Yes." "And let's say we threw in a salad, but you couldn't have any bread or anything else." "Yeah, I could do that." "Well, if you could go through life like that.... Do you like eggs in the morning? Or a cheese omelette?" "Yes,"Robert said, his voice almost giddy with relief. He called expecting a life sentence of rice cakes. Now he was being sent forth to eat cheeseburgers. "Yes, I do!" "If you just eat that way," Atkins told him, "you'll have eighty pounds off in six months."

quantity, the A.M.A. pointed out, "that could not possibly account for the dramatic results claimed for such diets." In "Dr. Atkins' New Diet Revolution," not surprisingly, he's become rather vague on the subject, mysteriously invoking something he calls Fat Mobilizing Substance. Last year, when I interviewed him, he offered a new hypothesis: that ketosis takes more energy than conventional food metabolism does, and that it is "a much less efficient pathway to burn up your calories via stored fat than it is via glucose." But he didn't want to be pinned down. "Nobody has really been able to work out that mechanism as well as I would have liked,"he conceded. Atkins is a big, white-haired man in his late sixties, well over six feet, with a barrel chest and a gruff, hard-edged voice. On the day we met, he was wearing a high-lapelled, four-button black suit. Given a holster and a six-shooter, he could have passed for the sheriff in a spaghetti western. He is an intimidating figure, his manner brusque and impatient. He gives the impression that he doesn't like having to explain his theories, that he finds the details tedious and unnecessary. Given the Photocopier Effect, of course, he is quite right. The appearance of an explanation is more important than the explanation itself. But Atkins seems to take this principle farther than anyone else. For example, in an attempt to convince his readers that eating pork chops, steaks, duck, and rack of lamb in abundance is good for them, Atkins points out that primitive Eskimo cultures had virtually no heart disease, despite a high-fat diet of fish and seal meat. But one obvious explanation for the Eskimo paradox is that cold-water fish and seal meat are rich in n-3 fatty acids--the "good" kind of fat. Red meat, on the other hand, is rich in saturated fat--the "bad" kind of fat. That dietary fats come in different forms, some of which are particularly bad for you and some of which are not, is the kind of basic fact that seventh graders are taught in Introduction to Nutrition. Atkins has a whole chapter on dietary fat in "New Diet Revolution" and doesn't make the distinction once. All diet-book authors profit from the Photocopier Effect. Atkins lives it. I watched Atkins recently as he conducted his daily one- hour radio show on New York's WEVD. We were in a Manhattan town house in the East Fifties, where he has his headquarters, in a sleek, modernist office filled with leather furniture and soapstone sculpture. He sat behind his desk--John Wayne in headphones--as his producer perched in front of him. It was a bravura performance. He spoke quickly and easily, glancing at his notes only briefly, and then deftly gave counsel to listeners around the region. The first call came from George, on his car phone. George told Atkins his ratio of triglycerides to cholesterol. It wasn't good. George was a very unhealthy man. "You're in big trouble," Atkins said. "You have to change your diet. What do you generally eat? What's your breakfast?"

Eat the steak, hold the French fries. Here is the list of ingredients for one of his breakfast "weight loss" recommendations: scrambled eggs for six. Keep in mind that Atkins is probably the most influential diet doctor in the world. 12 link sausages (be sure they contain no sugar) 1 3-ounce package cream cheese 1 tablespoon butter 3/4 cup cream 1/4 cup water 1 teaspoon seasoned salt 2 teaspoons parsley 8 eggs, beaten Atkins' Patent Claim centers on the magical weight-loss properties of something called "ketosis." When you eat carbohydrates, your body converts them into glycogen and stores them for ready use. If you are deprived of carbohydrates, however, your body has to turn to its own stores of fat and muscle for energy. Among the intermediate metabolic products of this fat breakdown are ketones, and when you produce lots of ketones, you're in ketosis. Since an accumulation of these chemicals swiftly becomes toxic, your body works very hard to get rid of them, either through the kidneys, as urine, or through the lungs, by exhaling, so people in ketosis commonly spend a lot of time in the bathroom and have breath that smells like rotten apples. Ketosis can also raise the risk of bone fracture and cardiac arrhythmia and can result in light-headedness, nausea, and the loss of nutrients like potassium and sodium. There is no doubt that you can lose weight while you're in ketosis. Between all that protein and those trips to the bathroom, you'll quickly become dehydrated and drop several pounds just through water loss. The nausea will probably curb your appetite. And if you do what Atkins says, and suddenly cut out virtually all carbohydrates, it will take a little while for your body to compensate for all those lost calories by demanding extra protein and fat. The weight loss isn't permanent, though. After a few weeks your body adjusts, and the weight--and your appetite--comes back. For Atkins, however, ketosis is as "delightful as sex and sunshine," which is why he wants dieters to cut out carbohydrates almost entirely. (To avoid bad breath he recommends carrying chlorophyll tablets and purse-size aerosol breath fresheners at all times; to avoid other complications, he recommends regular blood tests.) Somehow, he has convinced himself that his kind of ketosis is different from the bad kind of ketosis, and that his ketosis can actually lead to permanent weight loss. Why he thinks this, however, is a little unclear. In "Dr. Atkins' Diet Revolution" he thought that the key was in the many trips to the bathroom:"Hundreds of calories are sneaked out of your body every day in the form of ketones and a host of other incompletely broken down molecules of fat. You are disposing of these calories not by work or violent exercise--but just by breathing and allowing your kidneys to function. All this is achieved merely by cutting out your carbohydrates." Unfortunately, the year after that original edition of Atkins' book came out, the American Medical Association published a devastating critique of this theory, pointing out, among other things, that ketone losses in the urine and the breath rarely exceed a hundred calories a day--a

step aside. But here's where the study gets interesting. Langer then did the experiment a third time, in this case replacing the specific reason with a statement of the obvious: "Excuse me, I have five pages. May I use the Xerox machine, because I have to make some copies?" The percentage who let her do so this time was almost exactly the same as the one in the previous round--ninetythree per cent. The key to getting people to say yes, in other words, wasn't the explanation "because I'm in a rush" but merely the use of the word "because." What mattered wasn't the substance of the explanation but merely the rhetorical form--the conjunctional footprint--of an explanation. Isn't this how diet books work? Consider the following paragraph, taken at random from "The Zone": In paracrine hormonal responses, the hormone travels only a very short distance from a secreting cell to a target cell. Because of the short distance between the secreting cell and the target cell, paracrine responses don't need the long-distance capabilities of the bloodstream. Instead, they use the body's version of a regional system: the paracrine system. Finally, there are the autocrine hormone systems, analogous to the cord that links the handset of the phone to the phone itself. Here the secreting cells release a hormone that comes immediately back to affect the secreting cell itself. Don't worry if you can't follow what Sears is talking about here--following isn't really the point. It is enough that he is using the word "because."

3.
If there is any book that defines the diet genre, however, it is "Dr. Atkins' New Diet Revolution." Here is the conversion narrative at its finest. Dr. Atkins, a humble corporate physician, is fat. ("I had three chins.") He begins searching for answers. ("One evening I read about the work that Dr. Garfield Duncan had done in nutrition at the University of Pennsylvania. Fasting patients, he reported, lose all sense of hunger after forty-eight hours without food. That stunned me. . . . That defied logic.") He tests his unorthodox views on himself. As if by magic, he loses weight. He tests his unorthodox views on a group of executives at A.T. & T. As if by magic, they lose weight. Incredibly, he has come up with a diet that "produces steady weight loss" while setting "no limit on the amount of food you can eat." In 1972, inspired by his vision, he puts pen to paper. The result is "Dr. Atkins' Diet Revolution," one of the fifty best-selling books of all time. In the early nineties, he publishes "Dr. Atkins' New Diet Revolution," which sells more than three million copies and is on the Times best-seller list for almost all of 1997. More than two decades of scientific research into health and nutrition have elapsed in the interim, but Atkins' message has remained the same. Carbohydrates are bad. Everything else is good. Eat the hamburger, hold the bun.

deal of the kind of unrealistic self- sacrifice that causes many people to fall off the diet wagon. . . . In fact, I can even show you how to stay within these dietary guidelines while eating at fast-food restaurants." It is the very discipline of the Zone system that allows its adherent to lose weight without discipline. Or consider this from Adele Puhn's recent runaway best- seller, "The 5-Day Miracle Diet." America's No. 1 diet myth, she writes, is that "you have to deprive yourself to lose weight": Even though countless diet programs have said you can have your cake and eat it, too, in your heart of hearts, you have that "nibbling" doubt: For a diet to really work, you have to sacrifice. I know. I bought into this myth for a long time myself. And the fact is that on every other diet, deprivation is involved. Motivation can only take you so far. Eventually you're going to grab for that extra piece of cake, that box of cookies, that cheeseburger and fries. But not the 5-Day Miracle Diet. Let us pause and savor the five-hundred-and-forty-degree rhetorical triple gainer taken in those few sentences: (1) the idea that diet involves sacrifice is a myth; (2) all diets, to be sure, say that on their diets dieting without sacrifice is not a myth; (3) but you believe that dieting without sacrifice is a myth; (4) and I, too, believed that dieting without sacrifice is a myth; (5) because in fact on all diets dieting without sacrifice is a myth; (6) except on my diet, where dieting without sacrifice is not a myth. The expository sequence that these books follow--last one picked, moment of enlightenment, assertion of the one true way--finally amounts to nothing less than a conversion narrative. In conception and execution, diet books are selfconsciously theological. (Whom did Harvey Diamond meet after his impulsive, desperate mission to Santa Barbara? A man he will only identify, pseudonymously and mysteriously, as Mr. Jensen, an ethereal figure with "clear eyes, radiant skin, serene demeanor and well-proportioned body.") It is the appropriation of this religious narrative that permits the suspension of disbelief. There is a more general explanation for all this in the psychological literature--a phenomenon that might be called the Photocopier Effect, after the experiments of the Harvard social scientist Ellen Langer. Langer examined the apparently common-sense idea that if you are trying to persuade someone to do something for you, you are always better off if you provide a reason. She went up to a group of people waiting in line to use a library copying machine and said, "Excuse me, I have five pages. May I use the Xerox machine?" Sixty per cent said yes. Then she repeated the experiment on another group, except that she changed her request to "Excuse me, I have five pages. May I use the Xerox machine, because I'm in a rush?" Ninety-four per cent said yes. This much sounds like common sense: if you say, "because I'm in a rush"--if you explain your need--people are willing to

and I knew all the pain that only a fat kid can know.... I was always the last one reluctantly chosen for the teams." Martin Katahn, in his best-seller "The Rotation Diet," writes, "I was one of those fat kids who had no memory of ever being thin. Instead, I have memories such as not being able to run fast enough to keep up with my playmates, being chosen last for all games that required physical movement." Out of that darkness comes light: the Eureka Moment, when the author explains how he stumbled on the radical truth that inpired his diet. Sears found himself in the library of the Boston University School of Medicine, reading everything he could on the subject: "I had no preconceptions, no base of knowledge to work from, so I read everything. I eventually came across an obscure report..." Rachael Heller, who was a co-author of the best-selling "The Carbohydrate Addict's Diet" (and, incidentally, so fat growing up that she was "always the last one picked for the team"), was at home in bed when her doctor called, postponing her appointment and thereby setting in motion an extraordinary chain of events that involved veal parmigiana, a Greek salad, and two French crullers: "I will always be grateful for that particular arrangement of circumstances.... Sometimes we are fortunate enough to recognize and take advantage of them, sometimes not. This time I did. I believe it saved my life." Harvey Diamond, the co-author of the three-million-copy-selling "Fit for Life," was at a music festival two thousand miles from home, when he happened to overhear two people in front of him discussing the theories of a friend in Santa Barbara: "'Excuse me,' I interrupted, 'who is this fellow you are discussing?' In less than twenty-four hours I was on my way to Santa Barbara. Little did I know that I was on the brink of one of the most remarkable discoveries of my life." The Eureka Moment is followed, typically within a few pages, by the Patent Claim--the point at which the author shows why his Eureka Moment, which explains how weight can be lost without sacrifice, is different from the Eureka Moment of all those other diet books explaining how weight can be lost without sacrifice. This is harder than it appears. Dieters are actually attracted to the idea of discipline, because they attribute their condition to a failure of discipline. It's just that they know themselves well enough to realize that if a diet requires discipline they won't be able to follow it. At the same time, of course, even as the dieter realizes that what he is looking for--discipline without the discipline--has never been possible, he still clings to the hope that someday it might be. The Patent Claim must negotiate both paradoxes. Here is Sears, in his deft sixparagraph Patent Claim: "These are not unique claims. The proponents of every new diet that comes along say essentially the same thing. But if you're reading this book, you probably know that these diets don't really work."Why don't they work? Because they "violate the basic biochemical laws required to enter the Zone."Other diets don't have discipline. The Zone does. Yet, he adds, "The beauty of the dietary system presented in this book is that . . . it doesn't call for a great

about insulin, all he has done is come up with another low-calorie diet. He doesn't do the math for his readers, but some nutritionists have calculated that if you follow Sears's prescriptions religiously you'll take in at most seventeen hundred calories a day, and at seventeen hundred calories a day virtually anyone can lose weight. The problem with low-calorie diets, of course, is that no one can stay on them for very long. Just ask Sears. "Diets based on choice restriction and calorie limits usually fail," he writes in the second chapter of"The Zone," just as he is about to present his own choice-restricted and calorie-limited diet. "People on restrictive diets get tired of feeling hungry and deprived. They go off their diets, put the weight back on (primarily, as increased body fat) and then feel bad about themselves for not having enough will power, discipline, or motivation." These are not, however, the kinds of contradiction that seem to bother Sears. His first book's dust jacket claims that in the Zone you can "reset your genetic code" and "burn more fat watching TV than by exercising." By the time he's finished, Sears has held up his diet as the answer to virtually every medical ill facing Western society, from heart disease to cancer and on to alcoholism and PMS. He writes, "Dr. Paul Kahl, the same physician with whom I did the aids pilot study"-yes, Sears's diet is just the thing for aids, too--"told me the story of one of his patients, a fifty-year-old woman with MS." Paul put her on a Zone-favorable diet, and after a few months on the program she came in for a checkup. Paul asked the basic question: "How are you feeling?" Her answer was "Great!" Noticing that she was still using a cane for stability, Paul asked her, "If you're feeling so great, why are you still using the cane?" Her only response was that since developing MS she always had. Paul took the cane away and told her to walk to the end of the hallway and back. After a few tentative steps, she made the round trip quickly. When Paul asked her if she wanted her cane back, she just smiled and told him to keep it for someone who really needed it. Put down your carbohydrates and walk! It is hard, while reading this kind of thing, to escape the conclusion that what is said in a diet book somehow matters less than how it's said. Sears, after all, isn't the only diet specialist who seems to be making things up. They all seem to be making things up. But if you read a large number of popular diet books in succession, what is striking is that they all seem to be making things up in precisely the same way. It is as if the diet-book genre had an unspoken set of narrative rules and conventions, and all that matters is how skillfully those rules and conventions are adhered to. Sears, for example, begins fearful and despondent, his father dead of a heart attack at fifty-three, a "sword of Damocles" over his head. Judy Moscovitz, author of "The Rice Diet Report" (three months on the Times best-seller list), tells us, "I was always the fattest kid in the class,

fish, and very little red meat. Good nutrition, though, isn't really the point. Sears's argument is that being in the Zone can induce permanent weight loss-that by controlling carbohydrates and the production of insulin you can break your obsession with food and fundamentally alter the way your body works. "Weight loss . . . can be an ongoing and usually frustrating struggle for most people," he writes. "In the Zone it is painless, almost automatic." Does the Zone exist? Yes and no. Certainly, if people start eating a more healthful diet they'll feel better about themselves. But the idea that there is something magical about keeping insulin within a specific range is a little strange. Insulin is simply a hormone that regulates the storage of energy. Precisely how much insulin you need to store carbohydrates is dependent on all kinds of things, including how fit you are and whether, like many diabetics, you have a genetic predisposition toward insulin resistance. Generally speaking, the heavier and more out of shape you are, the more insulin your body needs to do its job. The Pima have a problem with obesity and that makes their problem with diabetes worse--not the other way around. High levels of insulin are the result of obesity. They aren't the cause of obesity. When I read the insulin section of "The Zone" to Gerald Reaven, an emeritus professor of medicine at Stanford University, who is acknowledged to be the country's leading insulin expert, I could hear him grinding his teeth. "I had the experience ofbeing on a panel discussion with Sears, and I couldn't believe the stuff that comes out of this guy's mouth," he said. "I think he's full of it." What Sears would have us believe is that when it comes to weight loss your body treats some kinds of calories differently from others--that the combination of the food we eat is more critical than the amount. To this end, he cites what he calls an "amazing" and "landmark" study published in 1956 in the British medical journal Lancet. (It should be a tipoff that the best corroborating research he can come up with here is more than forty years old.) In the study, a couple of researchers compared the effects of two different thousand-calorie diets--the first high in fat and protein and low in carbohydrates, and the second low in fat and protein and high in carbohydrates--on two groups of obese men. After eight to ten days, the men on the low-carbohydrate diet had lost more weight than the men on the high-carbohydrate diet. Sears concludes from the study that if you want to lose weight you should eat protein and shun carbohydrates. Actually, it shows nothing of the sort. Carbohydrates promote water retention; protein acts like a diuretic. Over a week or so, someone on a high-protein diet will always look better than someone on a high-carbohydrate diet, simply because of dehydration. When a similar study was conducted several years later, researchers found that after about three weeks--when the effects of dehydration had evened out--the weight loss on the two diets was virtually identical. The key isn't how you eat, in other words; it's how much you eat. Calories, not carbohydrates, are still what matters. The dirty little secret of the Zone system is that, despite Sears's expostulations

and fried food clogs arteries, that fresh vegetables and fruits help to ward off cancer, that fibre is good and sugar is bad and whole-wheat bread is better than white bread. That few of us are able to actually follow this advice is either our fault or the fault of the advice. Medical orthodoxy, naturally, tends toward the former position. Diet books tend toward the latter. Given how often the medical orthodoxy has been wrong in the past, that position is not, on its face, irrational. It's worth finding out whether it is true. Arguably the most popular diet of the moment, for example, is one invented by the biotechnology entrepreneur Barry Sears. Sears's first book, "The Zone," written with Bill Lawren, sold a million and a half copies and has been translated into fourteen languages. His second book, "Mastering the Zone," was on the bestseller lists for eleven weeks. Madonna is rumored to be on the Zone diet, and so are Howard Stern and President Clinton, and if you walk into almost any major bookstore in the country right now Sears's two best-sellers--plus a new book, "Zone Perfect Meals in Minutes"--will quite likely be featured on a display table near the front. They are ambitious books, filled with technical discussions of food chemistry, metabolism, evolutionary theory, and obscure scientific studies, all apparently serving as proof of the idea that through careful management of"the most powerful and ubiquitous drug we have: food" we can enter a kind of highefficiency, optimal metabolic state--the Zone. The key to entering the Zone, according to Sears, is limiting your carbohydrates. When you eat carbohydrates, he writes, you stimulate the production of insulin, and insulin is a hormone that evolved to put aside excess carbohydrate calories in the form of fat in case of future famine. So the insulin that's stimulated by excess carbohydrates aggressively promotes the accumulation of body fat. In other words, when we eat too much carbohydrate, we're essentially sending a hormonal message, via insulin, to the body (actually to the adipose cells). The message: "Store fat." His solution is a diet in which carbohydrates make up no more than forty per cent of all calories consumed (as opposed to the fifty per cent or more consumed by most Americans), with fat and protein coming to thirty per cent each. Maintaining that precise four-to-three ratio between carbohydrates and protein is, in Sears's opinion, critical for keeping insulin in check. "The Zone" includes all kinds of complicated instructions to help readers figure out how to do things like calculate their precise protein requirements in restaurants. ("Start with the protein, using the palm of your hand as a guide. The amount of protein that can fit into your palm is usually four protein blocks. That's about one chicken breast or 4 ounces sliced turkey.") It should be said that the kind of diet Sears suggests is perfectly nutritious. Following the Zone diet, you'll eat lots of fibre, fresh fruit, fresh vegetables, and

For one reason or another, we cannot stop eating. "Since many people cannot lose much weight no matter how hard they try, and promptly regain whatever they do lose," the editors of The New England Journal of Medicine wearily concluded last month, "the vast amount of money spent on diet clubs, special foods and over-the-counter remedies, estimated to be on the order of $30 billion to $50 billion yearly, is wasted." Who could argue? If the Pima--who are surrounded by the immediate and tangible consequences of obesity, who have every conceivable motivation--can't stop themselves from eating their way to illness, what hope is there for the rest of us? In the scientific literature, there is something called Gourmand Syndrome--a neurological condition caused by anterior brain lesions and characterized by an unusual passion for eating. The syndrome was described in a recent issue of the journal Neurology, and the irrational, seemingly uncontrollable obsession with food evinced by its victims seems a perfect metaphor for the irrational, apparently uncontrollable obsession with food which seems to have overtaken American society as a whole. Here is a diary entry from a Gourmand Syndrome patient, a fifty-five-year-old stroke victim who had previously displayed no more than a perfunctory interest in food. After I could stand on my feet again, I dreamt to go downtown and sit down in this well-known restaurant. There I would get a beer, sausage, and potatoes. Slowly my diet improved again and thus did quality of life. The day after discharge, my first trip brought me to this restaurant, and here I order potato salad, sausage, and a beer. I feel wonderful. My spouse anxiously registers everything I eat and nibble. It irritates me. A few steps down the street, we enter a coffee-house. My hand is reaching for a pastry, my wife's hand reaches between. Through the window I see my bank. If I choose, I could buy all the pastry I wanted, including the whole store. The creamy pastry slips from the foil like a mermaid. I take a bite.

2.
Is there an easy way out of this problem? Every year, millions of Americans buy books outlining new approaches to nutrition and diet, nearly all of which are based on the idea that overcoming our obsession with food is really just a matter of technique: that the right foods eaten in the right combination can succeed where more traditional approaches to nutrition have failed. A cynic would say, of course, that the seemingly endless supply of these books proves their lack of efficacy, since if one of these diets actually worked there would be no need for another. But that's not quite fair. After all, the medical establishment, too, has been giving Americans nutritional advice without visible effect. We have been told that we must not take in more calories than we burn, that we cannot lose weight if we don't exercise consistently, that an excess of eggs, red meat, cheese,

The Pima have built a new wellness center in downtown Sacaton, with a weight room and a gymnasium. They now have an education program on nutrition aimed at preschoolers and first graders, and at all tribal functions signs identify healthful food choices--a tray of vegetables or of fruit, say. They are doing, in other words, what public-health professionals are supposed to be doing. But results are hard to see. "We've had kids who were diabetic, whose mothers had diabetes and were on dialysis and had died of kidney failure," one of the tribe's nutritionists told me. "You'd think that that would make a difference--that it would motivate them to keep their diet under control. It doesn't." She got up from her desk, walked to a bookshelf, and pulled out two bottles of Coca-Cola. One was an old glass bottle. The other was a modern plastic bottle, which towered over it. "The original Coke bottle, in the nineteen-thirties, was six and a half ounces." She held up the plastic bottle. "Now they are marketing one litre as a single serving. That's five times the original serving size. The McDonald's regular hamburger is two hundred and sixty calories, but now you've got the double cheeseburger, which is four hundred and forty-five calories. Portion sizes are getting way out of whack. Eating is not about hunger anymore. The fact that people are hungry is way down on the list of why they eat." I told her that I had come to Sacaton, the front lines of the weight battle, in order to find out what really works in fighting obesity. She looked at me and shrugged. "We're the last people who could tell you that," she said. In the early nineteen-sixties, at about the time the N.I.H. team stumbled on the Pima, seventeen per cent of middle-aged Americans met the clinical definition of obesity. Today, that figure is 32.3 per cent. Between the early nineteen-seventies and the early nineteen-nineties, the percentage of preschool girls who were overweight went from 5.8 per cent to ten per cent. The number of Americans who fall into what epidemiologists call Class Three Obesity--that is, people too grossly overweight, say, to fit into an airline seat--has risen three hundred and fifty per cent in the past thirty years. "We've looked at trends by educational level, race, and ethnic group, we've compared smokers and non-smokers, and it's very hard to say that there is any group that is not experiencing this kind of weight gain," Katherine Flegal, a senior research epidemiologist at the National Center for Health Statistics, says. "It's all over the world. In China, the prevalence of obesity is vanishingly low, yet they are showing an increase. In Western Samoa, it is very high, and they are showing an increase." In the same period, science has unlocked many of obesity's secrets, the American public has been given a thorough education in the principles of good nutrition, health clubs have sprung up from one end of the country to another, dieting has become a religion, and health food a marketing phenomenon. None of it has mattered. It is the Pima paradox: in the fight against obesity all the things that worked in curbing behaviors like drunk driving and smoking and in encouraging things like safe sex and the use of seat belts--education, awareness, motivation--don't seem to work.

When I visited the town, on a monotonously bright desert day not long ago, I watched a group of children on a playing field behind the middle school moving at what seemed to be half speed, their generous shirts and baggy jeans barely concealing their bulk. At the hospital, one of the tribe's public-health workers told me that when she began an education program on nutrition several years ago she wanted to start with second graders, to catch the children before it was too late. "We were under the delusion that kids didn't gain weight until the second grade," she said, shaking her head. "But then we realized we'd have to go younger. Those kids couldn't run around the block." From the beginning, the N.I.H. researchers have hoped that if they can understand why the Pima are so obese they can better understand obesity in the rest of us; the assumption is that obesity in the Pima is different only in degree, not in kind. One hypothesis for the Pima's plight, favored by Eric Ravussin, of the N.I.H.'s Phoenix team, is that after generations of living in the desert the only Pima who survived famine and drought were those highly adept at storing fat in times of plenty. Under normal circumstances, this disposition was kept in check by the Pima's traditional diet: cholla-cactus buds, honey mesquite, povertyweed, and prickly pears from the desert floor; mule deer, white-winged dove, and blacktailed jackrabbit; squawfish from the Gila River; and wheat, squash, and beans grown in irrigated desert fields. By the end of the Second World War, however, the Pima had almost entirely left the land, and they began to eat like other Americans. Their traditional diet had been fifteen to twenty per cent fat. Their new diet was closer to forty per cent fat. Famine, which had long been a recurrent condition, gave way to permanent plenty, and so the Pima's "thrifty" genes, once an advantage, were now a liability. N.I.H. researchers are trying to find these genes, on the theory that they may be the same genes that contribute to obesity in the rest of us. Their studies at Sacaton have also uncovered valuable clues to how diabetes works, how obesity in pregnant women affects their children, and how human metabolism is altered by weight gain. All told, the collaboration between the N.I.H. and the Pima is one of the most fruitful relationships in modern medical science--with one fateful exception. After thirty-five years, no one has had any success helping the Pima lose weight. For all the prodding and poking, the hundreds of research papers describing their bodily processes, and the determined efforts of health workers, year after year the tribe grows fatter. "I used to be a nurse, I used to work in the clinic, I used to be all gung ho about going out and teaching people about diabetics and obesity," Teresa Wall, who heads the tribe's public-health department, told me. "I thought that was all people needed--information. But they weren't interested. They had other issues." Wall is a Pima, short and stocky, who has long, straight black hair, worn halfway down her back. She spoke softly. "There's something missing. It's one thing to say to people, 'This is what you should do.' It's another to actually get them to take it in."

Sacton lies in the center of Arizona, just off interstate 10, on the Gila River reservation of the Pima Indian tribe. It is a small town, dusty and unremarkable, which looks as if it had been blown there by a gust of desert wind. Shacks and plywood bungalows are scattered along a dirt-and-asphalt grid. Dogs crisscross the streets. Back yards are filled with rusted trucks and junk. The desert in these parts is scruffy and barren, drained of water by the rapid growth of Phoenix, just half an hour's drive to the north. The nearby Gila River is dry, and the fields of wheat and cushaw squash and tepary beans which the Pima used to cultivate are long gone. The only prepossessing building in Sacaton is a gleaming low-slung modern structure on the outskirts of town--the Hu Hu Kam Memorial Hospital. There is nothing bigger or more impressive for miles, and that is appropriate, since medicine is what has brought Sacaton any wisp of renown it has. Thirty-five years ago, a team of National Institutes of Health researchers arrived in Sacaton to study rheumatoid arthritis. They wanted to see whether the Pima had higher or lower rates of the disease than the Blackfoot of Montana. A third of the way through their survey, however, they realized that they had stumbled on something altogether strange--a population in the grip of a plague. Two years later, the N.I.H. returned to the Gila River Indian Reservation in force. An exhaustive epidemiological expedition was launched, in which thousands of Pima were examined every two years by government scientists, their weight and height and blood pressure checked, their blood sugar monitored, and their eyes and kidneys scrutinized. In Phoenix, a modern medical center devoted to Native Americans was built; on its top floor, the N.I.H. installed a state-of-the-art research lab, including the first metabolic chamber in North America--a sealed room in which to measure the precise energy intake and expenditure of Pima research subjects. Genetic samples were taken; family histories were mapped; patterns of illness and death were traced from relative to relative and generation to generation. Today, the original study group has grown from four thousand people to seven thousand five hundred, and so many new studies have been added to the old that the total number of research papers arising from the Gila River reservation takes up almost forty feet of shelf space in the N.I.H. library in Phoenix. The Pima are famous now--famous for being fatter than any other group in the world, with the exception only of the Nauru islanders of the West Pacific. Among those over thirty- five on the reservation, the rate of diabetes, the disease most closely associated with obesity, is fifty per cent, eight times the national average and a figure unmatched in medical history. It is not unheard of in Sacaton for adults to weigh five hundred pounds, for teen-agers to be suffering from diabetes, or for relatively young men and women to be already disabled by the disease--to be blind, to have lost a limb, to be confined to a wheelchair, or to be dependent on kidney dialysis.