The following is an excerpt from How We Do Harm: A Doctor Breaks Ranks About Being Sick in America, by Otis W. Brawley, MD, with Paul Goldberg, published in 2012.

Brawley is Bloomberg Distinguished Professor of Oncology and Epidemiology at Johns Hopkins University, and Goldberg is editor and publisher of The Cancer Letter. 

Brawley and Goldberg serve as co-editors of the Cancer History Project. 

Chapter 1: Chief Complaint

SHE WALKS THROUGH the emergency-room doors sometime in the early morning. In a plastic bag, she carries an object wrapped in a moist towel.

She is not bleeding. She is not in shock. Her vital signs are okay. There is no reason to think that she will collapse on the spot. Since she is not truly an emergency patient, she is triaged to the back of the line, and other folks, those in immediate distress, get in for treatment ahead of her. She waits on a gurney in a cavernous, green hallway.

The “chief complaint” on her chart at Grady Memorial Hospital, in downtown Atlanta, might have set off a wave of nausea at a hospital in a white suburb or almost any place in the civilized world. It reads, “My breast has fallen off. Can you reattach it?”

She waits for at least four hours—likely, five or six. The triage nurse doesn’t seek to determine the whereabouts of the breast.

Obviously, the breast is in the bag.

* * *

I am making rounds on the tenth floor when I get a page from Tammie Quest in the Emergency Department.

At Grady, we take care of patients who can’t pay, patients no one wants. They come to us with their bleeding wounds, their run-amok diabetes, their end-stage tumors, their drama. You deal with this wreckage for a while and you develop a coping mechanism. You detach. That’s why many doctors, nurses, and social workers here come off as if they have departed for a less turbulent planet.

Tammie is not like that. She emotes, and I like having her as the queen of ER—an experienced black woman who gives a shit. When Dr. Quest pages me, I know it isn’t because she needs a social interaction. It has to be something serious.

“We are wanted in the ER,” I tell my team.

The cancer team today consists of a fellow, a resident, two medical students, and yours truly, in a flowing white coat, as the attending physician. I lead the way down the hall. Having grown up Catholic, I can’t help thinking of the med students and young doctors as altar boys following a priest.

I am a medical oncologist, the kind of doctor who gives chemotherapy. My other interests are epidemiology and biostatistics. I am someone you might ask whether a drug works, whether you should get a cancer screening test, and whether a white man’s cancer differs from a black man’s cancer. You can also ask me if we are winning the “war” on the cluster of diseases we call cancer. As chief medical officer of the American Cancer Society—a position I have held since 2007—I often end up quoted in the newspapers, and I am on television a lot. In addition to my academic, journalistic, and public-policy roles, I have been taking care of cancer patients at Grady for nearly a decade, first as the founding director of the cancer center, and now as chief doctor at the ACS.

My retinue behind me, I keep up a fast pace, this side of a jog. Bill Bernstein, the fellow, is the most senior of the group. Bill is a Newton, Massachusetts, suburbanite, still boyish. He is having trouble adjusting to the South, to Atlanta, to its inner city. He is trying, but it’s hard to miss that black people and poor people perplex him. Contact with so much despair makes him awkward. But he has a good heart, a surfeit of common sense—and he is smart. Whatever we teach him at Grady will make him a better doctor wherever he ends up.

Grady suffers from what the administration here calls a “vertical transportation problem.” Our elevators are slow at best, broken at worst. We head for the stairs, rushing down to the first floor, then through long, green hallways into the ER.

Grady is a monument to racism. Racism is built into it, as is poverty, as is despair. Shaped like a capital letter H, Grady is essentially two hospitals with a hallway—a crossover—in the middle to keep things separate but equal for sixteen stories.

In the 1950s and ‘60s, white patients were wheeled into the front section, which faces the city. Blacks went to the back of the H. This structure—built in 1953—was actually an improvement over the previous incarnation. The Big H—the current Grady—replaced two separate buildings—the whites got a brick building, the blacks a run-down wood-frame structure. Older Atlantans continue to refer to the place in a chilling plural, the Gradys.

You end up at Grady for four main reasons. It could happen because you have no insurance and are denied care at a private hospital, or because you are unconscious when you arrive by ambulance. When your lights are out, you are in no position to ask to be taken to a cleaner, better-lit, suburban palace of medicine. A third, small contingent are older black folks with insurance, who could go anywhere but have retained a dim memory of Grady as the only Atlanta hospital that accepted us. The fourth category, injured cops and firemen, know that we see a lot of shock and trauma and are good at it. We are their ER of choice.

Today, our 950-bed behemoth stands for another form of segregation: poor versus rich, separate but with no pretense of equality. Grady is Atlanta’s safety-net hospital. It is also the largest hospital in the United States.

The ER, arguably the principal entry point to Grady, was built in the center of the hospital, filling in some of the H on the first floor. To build it, Grady administrators got some federal funds in time for the 1996 Summer Olympics. This fueled financial machinations, which led to criminal charges, which led to prison terms. (In retrospect, the bulk of the money was put to good use. Many of the victims of the Olympic Park bombing came through our ER.)

The hallways here are incredibly crowded, even by the standards of inner-city hospitals. Patients are triaged into three color-coded lines—surgery, internal medicine, obstetrics—and placed on gurneys two-deep, leaving almost no room for staff to squeeze through.

You might see a homeless woman drifting in and out of consciousness next to a Georgia Tech student bloodied from being pistol-whipped in an armed robbery, next to a fifty-seven-year-old suburban secretary terrified by a sudden loss of vision, next to a twenty-eight-year-old hooker writhing in pain that shoots up from her lower abdomen, next to a conventioneer who blacked out briefly in a cylindrical tower of a downtown hotel, next to a fourteen-year-old slum dweller who struggles for breath as his asthma attack subsides.

When I first arrived in Atlanta and all of this was new to me, I took my wife, Yolanda, through the Grady ER on a Friday night.

“Oh, the humanity,” she said.

Yolanda, a lawyer with the U.S. Securities and Exchange Commission, feels happier above the Mason-Dixon Line.

* * *

TAMMIE Quest—I use her real name—is cute, has a broad, infectious smile, and comes from privilege. She grew up in Southern California and frequently refers to herself as a “black Valley girl.”

Though she identifies with the West Coast, a lot of Atlanta has rubbed off on her in the Grady ER.

No two ERs are alike. Ours tells the story of Atlanta more clearly, more poignantly than its skyline. Patients everywhere are scared of their wounds or diseases that rage inside them. Here, in the middle of this big, hot, loud, violent city, they have an added fear: they are terrified of each other, often with good cause.

Elsewhere, patients might trust us doctors, admire us, even bow to our robes, our honorifics, and the all-caps abbreviations that follow our names. Here, not so much.

A place called Tuskegee is about two hours away from here. It’s where government doctors staged a medical experiment in the thirties: they watched black men die of syphilis, withholding treatment even after effective drugs were invented.

Tuskegee is not an abstraction in these parts. It’s a physical place, as palpable as a big, deep wound, and eighty-plus years don’t mean a thing. Tuskegee is a huge, flashing CAUTION sign in the consciousness of Southern black folks. It explains why they don’t trust doctors much and why good docs such as Tammie have to fight so hard to earn their elementary trust.

Like me, Tammie is a member of the medical-school faculty at Emory University, and, like me, she has several academic interests. One of these interests is end-of-life care for cancer patients: controlling the symptoms when someone with advanced cancer shows up in your ER.

Seeing us approach, she walks toward us and hands me a wooden clipboard with the Grady forms. I look at her face, gauging the mixture of sadness, moral outrage, and fatigue.

She says something like “This patient needs someone who cares,” and disappears.

I glance at the chief complaint.

“Holy shit,” I say to Bill Bernstein and, more so, to myself.

I introduce myself to a trim, middle-aged, black woman, not unattractive, wearing a blue examination gown conspicuously stamped GRADY. (At Grady, things such as gowns, infusion pumps, and money tend to vanish.)

From the moment Tammie paged me, I knew that the situation had to be more than a run-of-the-mill emergency. This patient clearly is not about to die on the examination table. She doesn’t need emergency treatment. Before anything, she needs somebody to talk to. She needs attention, both medical and human.

The patient, Edna Riggs, is fifty-three. She works for the phone company and lives on the southeast side of Atlanta.

Sitting on an exam table, she looks placid. When she extends her hand, it feels limp. She makes fleeting eye contact. This is depression, maybe. Shame does the same thing, as does a sense of doom. Fatalism is the word doctors have repurposed to describe this last form of alienation.

In medicine, we speak a language of our own, and Edna’s physical problem has a name in doctorese: automastectomy. It’s a fancy way of saying that the patient’s breast has fallen off by itself.

An automastectomy can occur when a tumor grows so big and so deep that it cuts off the blood supply from the chest to the breast. Denied oxygen, breast tissue dies and the breast starts to detach from the chest wall. At places such as Grady, automastectomies are seen a couple of times a year, often enough to be taken in stride.

This case is different from others I have seen only because Edna Riggs has wrapped her detached breast in a moist, light-blue towel and brought it with her for reattachment. I can’t help wondering why the towel is moist. Some deliberateness has gone into the breast’s care. I cringe at the thought that Edna has kept that package next to her on the gurney in the ER for hours.

In the exam gown, Edna’s chest looks surprisingly normal. I ask how long she has had a “breast problem.”

She first felt something in her breast when her son was in second grade, she replies. It has grown over the years. She speaks correct English, not the language of the streets. She sounds like someone who has had schooling, a person who reads. Her hair is clean and combed, she is dressed neatly.

What grade is her son in now?


I don’t react, not visibly. She has known she had a problem for nine years—why did she do nothing?

I ask Edna’s permission to examine her. She nods. I ask her to lie down, my entourage gathering around.

I help her remove her right arm from the gown, trying to respect her modesty and preserve as much dignity as possible. I undrape the right breast, or the place where the right breast had been. The chest wall is now rugged. I see yellowish, fibrous tissue and dry blood. There is the unforgettable smell of anaerobic bacteria. The wound is infected.

I reach for examination gloves. I palpate her chest wall and feel under her armpit, looking for evidence of enlarged nodes. After examining the breast wall, I look in the towel. Her amputated breast could fit on her chest as if it were a puzzle piece.

I am not looking forward to Edna’s repeating her request to reattach the breast. If she asks directly, I will have to say that this is not possible and explain why not. My preference is to move slowly, to let her adjust, to make her comfortable with me, with receiving medical care for her condition. I fear that she will get up, leave, and never return. Fortunately, Edna doesn’t repeat her request. Perhaps the magnitude of the problem confronting her is starting to sink in.

Edna’s breast cancer has been growing for at least nine years. It’s unheard-of that cancer such as this would be anything but metastatic. The disease has to have disseminated to her bones, lungs, brain, liver. I feel a wave of frustration and anger.

Another day at Grady Memorial Hospital. Here I sit, talking with a patient whom we would probably have cured nine years earlier, and today I will have to tell her that she has a terminal disease.

The rest is logistics. I arrange for the pathology and radiology to get confirmation. We always get pathologic confirmation of cancer, even when we are almost certain that it is cancer. An old medical saying goes: “When you hear hoofbeats, think horses, not zebras.” This saying has an important corollary: “You don’t want to be bit on the ass by a zebra.” There is a remote possibility that Edna’s automastectomy was caused by leprosy or some unusual infectious disease. It’s cheap and easy to get verification that it’s cancer.

I ask Bill Bernstein to talk with Edna, to take a full history, to perform a full examination. The objective is to rule out neurologic problems from spread of the disease to the brain or spine, to look for other evidence of problems caused by the disease.

If you take me aside and ask why I’m withdrawing from the scene, I will say that I am trying to awaken Bill’s compassion. But it is something else as well, something about me. I am afraid of growing callous. I acknowledge this readily, as a means of staving it off. I am trying to avoid accepting the unnecessary loss of yet another life. In the case of Edna Riggs, the abstract, scholarly term health disparities acquires a very real smell of a rotting breast.

I take my leave and, with the resident, start arranging tests to confirm the diagnosis and get Edna ready for treatment. We will fight, even though we are going to lose. Metastatic breast cancer always wins. We have drugs to decrease pain and even make most people live longer, but we can beat breast cancer only when it’s caught early.

* * *

WE admit Edna Riggs into the hospital, to get the tests done and to start antibiotic treatment of the infected wound. We could have done the workup without admitting her, but I fear that she will leave the system as abruptly as she entered. Psychological and emotional support are legitimate reasons for admittance, though most insurance companies and Medicaid would disagree.

As she starts to trust me, Edna tells me how frightened she was when she found a lump in her breast. Right away, she knew it was breast cancer, and in her experience, everyone who got breast cancer died quickly, painfully. Insurance problems kept her away from the doctor, as did the fear of dying. She knew she would die after going to the doctor. Several of her friends had.

Early on, Edna had some insurance, which didn’t do her any good. Her employer wouldn’t let her take just two or three hours of sick leave to go to the doctor. If she needed to take sick leave, she had to take it in increments of one day. This guaranteed that an employee would exhaust all the leave quickly. If Edna had been fired for taking time off after exhausting her sick leave, her three kids, too, would have lost support and insurance.

Acknowledging the physical problem and facing the consequences became increasingly difficult. Edna tells me that she feared the disease, but she also feared the system. Would the doctors scold her? Would they experiment on her? Would they give her drugs that caused nausea, vomiting, hair loss? Would the hospital kill her?

Edna’s decision to stay out of the medical system was about fear: fear of breast cancer, fear of the medical profession, fear of losing the roof over her kids’ heads. Fear intensified after her employer started to require copayments from workers who wanted to be insured. This extra $3,000 a year made health insurance too expensive to keep.

Payment for medical services and sick-leave policies determine the quality of care we receive. Several years ago, my research team at the American Cancer Society published data showing that people diagnosed with cancer who had no insurance or were insured through Medicaid were 1.6 times more likely to die in five years as those with private insurance.

In breast cancer, patients with private insurance were more likely to be diagnosed with Stage I breast cancer than those who had no insurance or were receiving Medicaid.

In colon cancer, too, the chances of catching the disease at an earlier, treatable stage were lower in the uninsured and Medicaid populations.

Even when the disease was found early, an uninsured patient did worse than one with insurance. For example, an insured patient with Stage II colon cancer had better odds of being alive five years after diagnosis than an uninsured patient with what should be highly curable Stage I cancer.

Another study focused on emergency surgery to treat bowel perforation, peritonitis, or obstruction in colon-cancer patients under age sixty-five. This surgical emergency was 2.6 times more frequent in uninsured patients than in those with private insurance. Among patients receiving Medicaid, the odds of needing this surgery was 2.1 times higher than in those with private coverage.

ACS epidemiologists estimate that the lack of insurance annually costs eight thousand Americans their lives due to inability to receive cancer treatment. Even controlling cancer pain is no small challenge if you are poor. Uninsured patients cannot afford pain medicines. The social programs that give them medication heavily ration pain meds.

Even if you have insurance that will pay for your treatment, you may still not be able to afford to receive it.

I have seen poor breast-cancer patients choose mastectomy (surgical removal of the entire breast) over a lumpectomy (removal of the tumor) because of employer sick-leave policies. A woman who chooses a lumpectomy must also receive radiation, which has to be given daily, Monday through Friday, for six to eight weeks. The treatment requires fifteen minutes in the clinic, but it’s done only during business hours. Unfortunately, this less disfiguring treatment is hardly an option for a woman who knows that longer postoperative treatment will cause her to lose her job.

Most states have laws that require doctors to tell the patients that they have a choice of full removal of the breast or removal of the cancerous lump with follow-up radiation. I do better than that. I tell my patients that I believe that lumpectomy with radiation is a better option, as it is less deforming and likely to lead to less long-term swelling.

* * *

PATIENTS most likely to have the worst outcomes are defined in a couple of ways. Poverty is the biggest driver, followed by race. (Race is complicated. For now, let’s think of it as just another snippet of data collected from cancer patients, and delve into its significance later.)

Much of the problem is that poor people don’t get care that would be likely to help them. The reasons for this are complex. Perhaps they can’t get care, or don’t know where care is available, or they haven’t been offered insurance or steady access to care by their jobs or social services.

Here is the problem: Poor Americans consume too little health care, especially preventive health care. Other Americans—often rich Americans—consume too much health care, often unwisely, and sometimes to their detriment. The American health-care system combines famine with gluttony.

We could improve dismal health outcomes on both ends of the socioeconomic spectrum if we were simply faithful to science, if we provided and practiced care that we know to be effective.

* * *

EARLY on, Edna ignored her tumor. She accomplished this easily during her busy days, but not when she was alone at night.

The disease progressed relentlessly. The lump grew. Then the tumor broke through the skin, causing a gaping wound, which became infected. The odor caused problems at work. Edna tried to conceal it with body powder and cologne, which worked at first. Her kids started trying to get her to come in and get help several months earlier, after a powerful, relentless stench finally set in.

Since Edna couldn’t pay for private insurance and have enough money left over to provide for her family, she had to come to Grady. Officially, Grady treats any resident of the two counties that support it: Fulton and DeKalb. When I arrived in Atlanta in 2001, the hospital was lax in enforcing the residency requirement. It ended up being the hospital for poor people in many surrounding counties, even though only Fulton and DeKalb taxpayers paid. As costs grew, Grady was forced to require proof of residency.

Our doctors are good, but free care comes at the cost of time lost waiting for appointments, waiting for tests. You can spend an entire day waiting for a service that a private doctor’s office provides in fifteen minutes or less. People like Edna, who need every day’s earnings and who can easily be jettisoned from their jobs, can afford time away from work even less than professionals, who may have some savings and job security. So people like Edna wait until it’s impossible to wait any longer; they come to see us when it’s too late.

* * *

WHY do black women end up with more aggressive breast cancer? Is this due to some biological characteristic that correlates with race, perhaps even determined by it? Can there be such a thing as white breast cancer and black breast cancer? Could these be different diseases?

You have to synthesize a pile of statistical data and medical literature to get insight into these problems, but it’s worth the effort: You end up with extraordinarily valuable insights into the epidemiology and biology of cancer. More than that, you gain insight into economic structures in our society and, ultimately, something very big: the meaning of race.

At a glance, breast cancer in a black woman like Edna appears to differ from breast cancer in an average white woman. If you plot breast cancer on a spectrum from the worst prognosis to the best, a higher proportion of black women would wind up on the worst end.

One of the most ominous varieties of breast cancer is called triple-negative, because it is immune to three commonly used treatments. The surface of the cancer cell in that form of the disease lacks receptors to the hormones estrogen and progesterone and is similarly devoid of receptors to the HER-2/neu protein.

We have drugs that target breast cancer through these three channels. However, in triple-negative breast cancer, these drugs have nothing to latch onto, and all we can do is resort to desperate measures: harsher chemotherapies, which we know are frequently of little or no use.

About 30 percent of breast cancer in black women is triple-negative disease, compared to 18 percent in white women.

This disparity could appear to suggest a biological difference, but in fact it’s rooted in cultural, historical, and societal divides. To understand this, we have to look at the potential causes of breast cancer in white and black women.

To start with, let’s consider the incidence of better-prognosis cancer among white women. Instead of asking why black women are more likely to get more virulent breast cancer, let’s ask why white women are more likely to develop the disease that has a better prognosis.

The answer can be gleaned in part from the incidence statistics.

For the past three decades—or for as long as we have had a national registry—the incidence of breast cancer has been higher in white women than in black women.

In 2000, the National Cancer Institute’s Surveillance, Epidemiology, and End Results registry reported that during the previous year, blacks had an age-adjusted incidence rate of 125 per 100,000 women. In the past twenty years, the black incidence rate has bounced between the low of 105 per 100,000 in 1989 to the high point of 126 in 2008.

In 2000, white women had an incidence rate of 143 per 100,000. The breast cancer incidence rate in whites had risen from the 1970s, peaked at 147 per 100,000 in 1999, and has fallen to 129 per 100,000 in 2008.

The incidence rates were substantially apart over the past couple of decades, but have now nearly evened out. Was this occurring because white women were using mammography more and were therefore more likely to get diagnosed?

Not quite. The proportion of women getting mammography screening is roughly the same among whites and blacks. (I suspect that the proportion getting high-quality mammography is greater among whites than blacks, but this difference has not been adequately studied.)

The delay of pregnancy and childbirth is a more plausible explanation. White women tend to have children later in life than black women. Professional women, regardless of their race, go to college, establish their careers, and then have kids. Delaying childbirth past the age of thirty clearly increases the risk of breast cancer. To be specific, it increases the risk of estrogen-receptor-positive breast cancer, which has a better prognosis.

Also, white women have been more likely to use postmenopausal hormone-replacement therapy (HRT). Doctors prescribed HRT because it made sense logically. Without definitive data on the therapy’s biological effect, doctors were, in effect, staging a decades-long societal experiment.

By 2003, 35 percent of postmenopausal white American women had taken this therapy at some time. For cultural and socioeconomic reasons, black women tended not to take HRT. Fewer than 5 percent of postmenopausal black women took HRT. This is important, because HRT is associated with better prognosis breast cancer.

In 2003, an analysis from the well-designed study called the Women’s Health Initiative showed that HRT was correlated with an increased risk of breast cancer. It was actually correlated with an increased risk of estrogen-receptor-positive, better prognosis breast cancer. The societal experiment was over.

The analysis led to a drop in the use of HRT, which likely accounts for the drop in breast cancer in white women from 147 per 100,000 in 1999 to 129 per 100,000 in 2008.

* * *

A focus on some geographic areas offers insight into what drives breast cancer in educated white women.

Consider Long Island. The area has been known to have a higher incidence of breast cancer than among the general United States population.

In the early 1990s, breast cancer advocates petitioned the U.S. Congress to force the National Cancer Institute to study the “high rates of breast cancer in Nassau and Suffolk Counties on Long Island.” This led to Public Law 103-43, which prompted a series of studies, called the Long Island Breast Cancer Study Project.

At the beginning of the project, the incidence of breast cancer among Long Island’s white women was 138.7 per 100,000 in Nassau County and 142.7 per 100,000 in Suffolk, compared to 127.8 per 100,000 in the United States as a whole.

Mortality from breast cancer on Long Island wasn’t especially elevated. Indeed, several areas of New York State and many areas of the United States had higher death rates. A similarly high incidence was found among white women in the area north of San Francisco Bay.

Could the elevated incidence have been due to electromagnetic fields, hazardous wastes, or some other environmental cause?

After spending at least $20 million, the Long Island study project did not identify any pollutant that could be responsible for the elevated incidence of breast cancer.

However, those who believed that an environmental factor was at play were right. At least two such factors were driving the disparity between Long Island and the rest of the country: a higher level of education among area women and their choice to delay childbirth.

* * *

LET’S return to the disparity in triple-negative breast cancer by race: 30 percent in black breast cancer patients, and 18 percent in white patients.

There is no difference in the proportion of black and white women with progesterone-positive or HER2-positive disease. So if we are to focus on the 12 percent disparity, we must look exclusively at the racial difference in the prevalence of the estrogen receptor.

Does this suggest that skin color stands for some biological difference?

Not really.

Because of dietary differences that are caused by culture and socioeconomic status, a black girl in the United States accumulates weight much faster than a white girl. In the 1960s, the Centers for Disease Control and Prevention compared the start of menstruation by age. The study showed that the average age of menarche for white American girls was 12.8 years. For black American girls, it was 12.4 years. This is a bigger difference than it might seem. It means that 53 percent of black girls have started menstruating by their thirteenth birthday, compared to 43 percent of white girls.

Body mass index, a calculation based on weight and height, correlates with early nutrition status, which has a lot to do with age at first menstruation. Poor Americans have diets higher in calories and reach the weight of one hundred pounds faster.

Just the simple number of uninterrupted menstrual cycles increases the risk of breast cancer later in life.

The reason for this rapid weight gain in black girls has nothing to do with race, but reflects a high caloric intake and a diet rich in carbohydrates, a socioeconomic determinant of health. It’s not about race. It’s at least in part about the sort of food that is available in poor areas of inner cities.

The area of Detroit where I grew up and the areas of Atlanta where my patients come from are known as produce deserts. Grocery stores there carry all the chips, sodas, and mentholated cigarettes you may desire, but if you want a head of lettuce, you are out of luck.

You observe the same problems among poor whites, yet you don’t see them among wealthy, well-educated blacks.

I cite the CDC data from the 1960s because they measure the racial differences that are driving breast cancers we are diagnosing today. This disparity has since widened, and if we trace it, we can project the differences in breast cancer rates and prevalence of triple-negative disease for decades into the future.

This extrapolation produces a deeply disturbing picture: the black-white gap in the onset of menstruation and body weight has dramatically widened, which means that the disease disparities will widen also.

* * *

FOR the sake of argument, let’s set aside everything we know about Grady, Atlanta, and our race-obsessed society. Perhaps the best way to learn about breast cancer is to look at Scotland.

Scotland, which is virtually all-white, collects data at its every-ten-year census using a unique tool called the deprivation index. The index measures socioeconomic factors beyond household income. It asks about indoor plumbing, electricity, even how many servants one might employ. This index can discern that a college philosophy professor earning $55,000 per year is in a different socioeconomic stratum from a garbage collector earning $70,000 per year.

Using this index, a group of researchers found evidence pointing to a correlation between social deprivation and incidence of breast cancer that lacks estrogen receptors, a characteristic which makes the disease harder to treat. The deprived or poor who developed breast cancer were more likely to develop this kind of breast cancer. I find it ironic that one of the most important studies in minority health was an all-white study.

My friend Samuel Broder, when he was the director of the National Cancer Institute, used to say that poverty is a carcinogen. Skin color can be a marker of some sort, but you have to be careful not to rely on it too heavily. Wealth is a marker, too, as is education. Area of geographic origin and family history are also important, and all these factors must be considered.

* * *

EDNA has Stage IV breast cancer. Disease has spread all over her body. Had she come to see me early in the course of her disease, it would have cost about $30,000 to cure her. She could have remained a taxpayer. Her kids could have had a mother. Now, the cure is not an option. Still, we’ll fight. We will give her breast-cancer chemotherapy that will cost more than $150,000, even though the chances are she will still die in less than two years. If you are a caring doctor, you realize she is just fifty-three, with kids and folks who love her, and your motivation is akin to a philosophy of Wayne Gretzky: “You miss every shot you don’t take.”

Every time I start chemo for metastatic disease I think of a patient named Sandra, a lively, young black woman whom I have treated for six years. She had brain metastases when I first met her. She has had active disease ever since, and even the doctor who sent her to me reminds me every time he sees me that he is amazed that she is alive, functional, and enjoying life.

Yes, sometimes cancer drugs give us “long-term survival,” in the dispassionate language of those of us who study outcomes. But for every Sandra, we get fifty patients with metastatic disease who “don’t do well.” They live a median eighteen months, which means that half are living and half are dead a year and a half after diagnosis.

We try three treatments and contain Edna’s disease for a while. She dies at age fifty-five, about twenty months after walking into the ER.