Return to Emily Friedman home page

Separate and Unequal

by Emily Friedman

First published in the September/October 2002 issue of Health Forum Journal

Although disparities in health status among ethnic groups are partially a societal problem, health workers can and must do more to address them.

When the U.S. Department of Health and Human Services issued its decennial set of goals for improvement of Americans' health in 1990, Healthy People 2000, the plan included this objective: "Healthy People 2000 sets specific targets to narrow the gap between the total population and those population groups that now experience above average incidences of death, disease, and disability." Examples of such targets were that although the goal for overall infant mortality was no more than 7 deaths per 1,000 live births, the goal for African Americans was 11 per 1,000 live births, for Native Americans 8.5 per 1,000 live births, and for Latinos 8 per 1,000 live births. For maternal mortality, the overall goal was 3.3 deaths per 100,000 live births, but HHS settled for a goal of 5 deaths for African-American mothers.

This document was among the first formal acknowledgments by a federal administration of disparities in health status among different racial and ethnic groups, although there were far earlier calls for action: the great African-American activist W. E. B. Du Bois wrote a book on the subject in 1906, and Booker T. Washington in 1914 called for a Health Improvement Week for African Americans.

However, at the same time, Healthy People 2000 could also be interpreted as accepting many disparities, because it set more modest goals for minorities than it did for whites. This raised hackles at the time, which in turn led to more strenuous efforts to bring this "hidden" health care issue to light. By the time Healthy People 2010 was being prepared, the Clinton administration--inspired in part by the unceasing efforts of then-Surgeon General David Satcher, M.D., to call attention to disparities--decided that no separate, lower (or higher, as the case might be) goals would be set for different populations unless clinical evidence warranted it. Indeed, Healthy People 2010 made the elimination of racial and ethnic disparities one of its key goals. Satcher saw disparities not just as a public health problem, but also as a moral wrong; he told an audience in December 1999 that "bioethical principles call for one standard of health for all Americans."

There have been other efforts as well. In November 2000, the federal Agency for Healthcare Research and Quality (AHRQ) funded several Excellence Centers to Eliminate Ethnic/Racial Disparities (EXCEED), whose charge is to identify disparities, determine their causes and circumstances, and find ways of ending them. (You can learn more at

Current HHS Secretary Tommy Thompson has named elimination of racial and ethnic disparities one of his top priorities and has launched an initiative, Closing the Health Gap ( HHS Deputy Secretary Claude Allen told the Leadership Summit on Eliminating Racial and Ethnic Disparities in Health in July that the department has targeted six areas for special attention: infant mortality, cancer screening and management, heart disease and stroke, diabetes, HIV/AIDS, and child and adult immunization.

Little Improvement

With all this attention over the past 15 years, the situation is much improved, right? Wrong. A study issued on July 12 by the Centers for Disease Control and Prevention found that although overall infant mortality was down to 6.9 deaths per 1,000 births in 2000 (thus fulfilling one of the goals of Healthy People 2000), African-American infant mortality was 14 deaths per 1,000 live births. Since 1980, the rate has declined 48 percent for whites, but only 37 percent for African Americans--and the disparity between the two is worse than it was 20 years ago. African-American maternal mortality is more than three times that for white women, a figure that has changed little in 20 years.

The Institute of Medicine (IOM), in a major report issued in March, found that, among many other things, minorities are less likely to be given appropriate medications or to undergo bypass surgery when suffering from cardiovascular disease; that although African Americans have strokes about a third more often than whites, they are less likely to receive the proper interventions; and that minorities are less likely to be placed on waiting lists for kidney transplants or to receive either dialysis or transplants. And it is not just an issue for African Americans; studies have found that diabetes is three times more common among Native Americans than among whites. There are Latino, Asian-American, and Pacific Islander examples as well.

And, as if the numbers needed repeating, although only 11 percent of nonelderly whites lack health insurance, 20 percent of African Americans, 34 percent of Latinos, and 21 percent of other ethnic groups are uninsured. The IOM report found, however, that minorities receive lower-quality care than whites even if they are insured. There is a large and unfortunate body of literature that reinforces this finding.

What is behind this persistent problem? Some observers blame politics, claiming that the health of minorities is not a matter of interest to conservatives. Whether that is true or not, there was little improvement in the gaps during the Clinton years, so although issues of funding and political priorities are undoubtedly important, they hardly fully explain the continuing crisis.

And, of course, it is always fashionable to blame the health care system, whether by asserting that the system itself is racist, citing the lack of minority physicians (and the declining numbers of underrepresented-minority medical students), or claiming that physicians and nurses deliberately treat minorities badly.

If the situation were that simple, laws and regulations could be passed, if not on the federal level, then by the states (although there are quite a few on the books as it is). Overtly discriminatory policies could be identified and the perpetrators hauled off to jail.

However, as a growing literature demonstrates, most discrimination within the health care system is unintentional. Few physicians or nurses seek to avoid treating minority patients; besides, it is illegal to refuse to treat someone on the basis of his or her race (although it does happen). The sad truth is that there are so many factors involved--cultural differences, language problems, patient preferences, provider misconceptions, insurance status, bad previous experience--that just yelling "racism" explains little.

In saying this, I do not deny that overt prejudice and stereotyping occur in the health care system. One study cited by the IOM report found that physicians believed that African-American patients were "less intelligent, less educated, more likely to abuse drugs and alcohol, less apt to comply with medical advice, more likely to lack social support, and less likely to participate in cardiac rehabilitation than white patients, even after their income, education, and personality characteristics were taken into account." That ought to make for a fun patient encounter.

Systemic Problems

But we must keep in mind that any health care system reflects the society that created it, and only an extremist or the utterly na´ve would claim that there are not significant problems of racial and ethnic (and religious, gender, age, sexual orientation, and so on) bigotry in this society--and in every other society on earth. Some of that is bound to leak into health care.

It is also probable that being subjected to prejudice over and over again in the course of one's lifetime will sooner or later affect a person's health. Professor Michael Lu, M.D., of the UCLA Schools of Medicine and Public Health, has done exhaustive surveys of racial disparities in maternal and infant mortality, and has tried to consider every possible variable. His sad conclusion is that women who have often encountered racism are far more likely to have poorer birth outcomes. In his words, "Racial and ethnic disparities in birth outcomes are the consequences of differential exposures to damaging physical and social environments accumulated over the life course on a woman's reproductive health."

There's a tragic logic to this line of thinking. If children who have been abused or abandoned tend to have a harder time in many fields of endeavor, how could it not affect their health? If you have been called names or ostracized because of some personal characteristic, isn't that likely to lead to anything from stress-related disease to unhealthy and self-destructive habits? A physician told me recently that the rate of cigarette smoking among gay men is the highest of any U.S. subgroup, probably because of, in his words, "a pervasive sense of fatalism in that community."

All these factors, and more--especially insurance status--are undoubtedly involved in the stubborn riddle of disparities. It is thus unfair and inaccurate (not to mention ineffective) to point the finger solely at the health care system. But it is equally unfair and inaccurate to take the health care system off the hook and say that it's all just a big social disaster and there isn't anything we can do about it.

Former Surgeon General Satcher is right. This isn't just a public health problem or a demographic inevitability; this is an ethical issue for health care. It also challenges our claims that we are providing the best-quality care we can. Not when someone might not have access to transplantation or certain medications or a regular physician or appropriate prenatal care because of the color of her skin.

There are actions the health care system can take. A good framework for those actions was provided by the late John Eisenberg, M.D., director of AHRQ, who died far too young in March of this year. He saw persistent disparities in health status, insurance, and care as problems of quality. He also saw them as problems of systems, not of individuals. In a 2000 speech at Duke University, he said, "Whatever the cause, the disparity in the treatment given specific populations is a system error."

Some Solutions

So if we can keep our paranoia down to a dull roar, here are a few things that the health care system can do to reduce disparities in health status.

Identify and eliminate obvious prejudice. We have all heard remarks about racial and ethnic groups made by physicians, nurses, administrators, trustees, and other health care leaders. Yet a recent Kaiser Family Foundation study found that a majority of physicians believe that the health care system "rarely or never" discriminates against patients on the basis of personal characteristics; the sole exception (surprise) was insurance status. Minority physicians are far more likely to believe that discrimination takes place on a regular basis.

If inappropriate comments are made, if there is a pattern of lower-quality care, if certain patients are treated differently, it is incumbent upon the organization and the people who work in it to put a stop to the behaviors and the patterns, and, if necessary, the policy behind them--however delicately it might be framed.

Understand cultural factors. It is obvious that culture is a huge factor in how health care is designed, provided, and received. Nowhere is this more true than in the area of health disparities. Cultural attitudes concerning risk, death, technology, power relationships, duty to family, and many other perceptions, beliefs, and even superstitions affect patient and provider behavior alike. We need to learn more about these beliefs and how we can deal with them.

Acknowledge the effect of history. The health care system has not always been the most welcoming environment for people who are "different." After all, Charles Drew, M.D., the African-American physician whose work made modern blood transfusions possible, bled to death in 1950 because he was not allowed to receive "white" blood. A recent Pittsburgh Post-Gazette story noted that if potential patients perceive that the health care system is "uncaring and racially and culturally insensitive, they will stay away." Wilford Payne, director of a clinic that treats many low-income African Americans, says, "Black people don't go to the doctor until it hurts. When you grow up poor, you only go to the doctor because of a sickness your mom can't cure." Avoiding the system except in a dire emergency can thus "get rooted in the culture," he says.

Take the social context into account. People in this society are bombarded by negative (and positive) stereotypes of racial and ethnic groups. More people grow up today in racially homogeneous communities where contact with those who are different is formal, controlled--and often nonexistent. Even the schools, which were once seen as the hope for a more tolerant future, have become resegregated with a vengeance.

Health care does not operate in a vacuum, and its patients do not live in one. It can be extremely helpful to acknowledge this and ask patients (and for patients to ask providers) if they have been hurt, helped, or influenced by what's going on in the larger culture. Arab Americans, for example, have had a lot of recent experience with this.

Obtain better data on specific groups' health status. One of the main goals of the AHRQ initiatives is to obtain objective, scientific knowledge about the health status of different groups. Deputy HHS Secretary Allen has noted the dreadful--and only recently recognized--fact that Vietnamese women contract cervical cancer at a rate five times that of whites. Among the Pima tribe of Arizona, 80 percent of the elders have diabetes. African-American men are more likely to be hypertensive; white and Asian-American women are more likely to get osteoporosis. We need to find out why, rather than assuming that it's insurance or personal behavior or some social factor. It's hard to fight an enemy you can't identify.

Get together. The chances of one health care organization, one community clinic, or one crusading physician or nurse or administrator solving this problem alone are nil. In Rochester, New York, a broad coalition of groups has been formed to improve access to care for Latinos. Nationwide, Community Care Networks are trying to address disparities and access issues. This challenge requires the participation of providers, policymakers, insurers, police, churches, community groups, and anyone else who is interested. It is a broad-based crisis; it requires broad-based solutions.

Get real. There has been enough window dressing around this issue to improve every window in America. One more general study or cosmetic initiative accomplishes little. Even if actions are well intentioned, they may not get to the heart of the matter. Earlier this year, HHS Secretary Thompson, as a part of his Closing the Gap program, declared September 24th as Take a Loved One to the Doctor Day for African Americans. The implication is if they would just go get preventive care, we could clean up these disparities in no time.

This is unlikely. There is a long and wretched history of racial discrimination in health care, however much it has lessened in recent times. One in every five African Americans is uninsured. Studies show that many physicians and hospitals--for whatever reason--provide African Americans with a lower standard of care. I'm afraid some African Americans might try to take a loved one to the doctor only to find that no doctors are around; or if they are, they're "all booked up"; or if they aren't all booked up, the patient can't pay because he's uninsured. The Commonwealth Fund reported earlier this year that 28 percent of African Americans have no regular physician (as is true of 19 percent of whites, 31 percent of Asian Americans, and 41 percent of Latinos). Maybe we should declare Take a Loved One to the Emergency Department Day instead.

Monitor, feed back, and correct. One of John Eisenberg's core beliefs was that once you implement a program, you should track its outcomes. We have launched a thousand ships against disparities in recent years; most of them sank. Whatever you decide to do, find out if it works. If it doesn't, try something else.

Think and act for the long term. Ours is a last-minute society, and thinking for the long term is rarely practiced in policy, in business, or in health care. But Michael Lu points out that if you try to counter a lifetime of bias and suffering with nine months of prenatal care, you will fail. "It's like cramming for a final," he says. "It just won't work." We in health care need to think and act in a much longer time frame, working with children, with adolescents, and with teens, and trying to prevent the damage--social, psychological, and clinical--before it takes its toll. Plan on the basis of a generation or more. That is real preventive medicine, and it is probably our best way out of this great national shame.

Those who are at risk have a right to expect that we will not accept the status quo, no matter how convenient that might be. As Dr. Satcher has said, "If we are going to close the gap, we have to ask ourselves tough questions about what it will really take to address it." And we have to provide answers, however tough they, too, may be.

This article is dedicated to the memory of John Eisenberg, M.D., who throughout his professional life pursued improvement in the quality of care and the elimination of disparities in health status.

This article first appeared in the September/October 2002 issue of Health Forum Journal

Return to Emily Friedman home page