Medical Errors Due to Patient Profiling
By Renate G Justin, MD

"In order to think about the world relatively accurately, one should avoid prejudices, or preconceptions about the traits of certain people and things--for example, that red-headed women are untrustworthy."¹

A few years ago, while attending a nonmedical conference, I was called to the aid of a somewhat overweight, middle-aged woman, who was sitting on the ladies' room floor. She was leaning against the wall, complaining of chest pain. I took a brief history, and I felt that the severity of her pain called for more thorough investigation; I recommended calling an ambulance. Another physician at the conference offered to help me. Her first comment was: "Oh, she's Jewish; she's probably just emotionally upset and hysterical, like most Jewish people get." Based on her preconceptions of Jewish people, this doctor did not think evaluation for coronary artery heart disease was needed. The patient might have suffered severe consequences, even death, because of the physician's generalizations about Jewish people. The doctor ignored the instructions in the literature: "Don't make assumptions about MI risk based on the patient's gender or race."²

Being Jewish, I was not only shocked at the danger to which the patient was exposed but also by the realization that it could have been me who had chest pain, which would have been attributed to my religiously related hysteria. This experience made me reexamine my own preconceived notions in an effort to be more acutely aware that my "profiling" might endanger my patients.

Racial Profiling is Humiliating

"Racial profiling," drawing conclusions about people by observing their appearance, is not only dangerous for law enforcement officers but also for physicians. While police officers may arrest the wrong person, doctors can make the wrong diagnosis by attributing certain characteristics to an individual on the basis of generalizations about his or her sex,³ color,⁴ dress, or religion.

When my African-American son was in medical school in Chicago, he went into a drugstore to buy toothpaste. While he was deciding which brand to buy, the Caucasian security guard approached him and accused him of stealing film. Before allowing him to leave, the guard inspected his briefcase and frisked him. This humiliating assault was precipitated by the law enforcement officer, not because he had any evidence of wrongdoing but because he expected certain behaviors from young African-American men. "The deep-seated problem(s) of racial profiling, especially for men of color"⁵ is "ingrained in history."⁶

Unexamined Opinions Lead to Unintended Errors

If a physician holds unexamined opinions about a group of people and then applies them to an individual patient, he or she can make unintended errors that can lead to serious illness. Even if the description of the group is based on statistics, it is not wise to apply these to an individual group member without further investigation. As Sherlock Holmes put it so aptly: "While the individual man is an insoluble puzzle, in the aggregate he becomes a mathematical certainty."⁷ It is difficult to estimate how often patient profiling causes errors in diagnosis and treatment. However, if patient safety is to improve, profiling should be considered along with all the other factors cited for causing mistakes, such as fatigue due to excess workload, inexperience, ignorance, and negligence as well as to lack of communication between various areas of the health care system. The doctor's unconscious characterization of a patient profoundly influences the data that lead to a diagnosis. This bias also determines the patient's access to health care and affects communication between doctor and patient.

In our country, women were and are often invested with the characteristics attributed to the Jewish woman with chest pain: emotional lability and dysphoria. Dysmenorrhea, including menstrual cramps, for a long time was thought to be caused by a woman's poor emotional adjustment expressed in exaggerated, hysterical reaction patterns to pain, by low pain threshold, and by sexual fears. When I was in medical school, my roommate endured severe pain every month and could not even make it to classes on the days of her menses. Her gynecologist sent her to a psychiatrist for her monthly attacks of "insanity." No severe emotional problems surfaced, and she underwent a presacral neurectomy. Her suffering was not relieved until hormones became generally available. Her life then took on a more predictable rhythm, and no one questioned her mental equilibrium any longer. Now that there is a medication that eliminates menstrual cramps, all these strange dysthymic females no longer need psychoanalysis. Physicians, because of ignorance and sexual profiling, inflicted treatments that were ineffective and erroneously attributed personality traits to half of the human race, which it did not own.

Conclusions Based on Economic Status

Health care workers often draw conclusions based on the economic status of patients, which causes a breakdown in communication. For example, we would be surprised to find an apparently destitute individual who holds a PhD in anthropology. Similarly, an apparently upper-income patient would be expected to have at least average intelligence. This kind of expectation, linking intelligence to economic status, can lead to poor clinician-patient communication. The clinician may "talk down" to one patient and fail to fully explain scientific terminology to another. This can arouse anger in patients, which will interfere with the doctor-patient relationship and the healing process.

Economic status can also engender preconceptions about emotional attachment. Years ago, I observed a Peace Corps nurse in Nicaragua severely scolding a disheveled woman for not bringing her baby, who had diarrhea and dehydration, to the hospital sooner. The nurse implied that this malnourished, poverty-stricken mother did not care about her baby. In spite of the language barrier, the mother burst into tears at the nurse's harangue and clutched her child, reluctant to give it to the health care personnel. What the nurse failed to appreciate, in her own frustration about the serious condition of the child, is that poor as well as rich people love their children and must be respected for what they are able to do for them. Neither the poverty nor wealth surrounding a child are indicators of the love that child receives. The nurse made a mistake by basing her judgment about the parent's emotional attachment to the child on the destitute condition of the mother.

That health care providers expect certain behaviors to be related to the economic status of a patient is confirmed in an article in which the authors show that the majority of patients screened by their physicians for 'partner violence' during pregnancy are those on Medicaid and attending free clinics. Among other explanations for this observation they write: "This may reflect a perception among providers that poorer women, even outside the public provider setting, are more likely to experience partner violence."⁸ They then add that family violence can occur "among women in all sociodemographic groups."⁸

Assumptions About Older People

We also make assumptions about older people that may be statistically valid but not individually relevant. One of my patients, facing a mastectomy, said: "I am 82 and naively assumed that I wouldn't care if I had a mastectomy. I am really surprised at how sad and angry I am." Before her surgery she told me: "I really like my breasts, they are part of me; John likes them. I feel lopsided already." I thought, as my elderly patient did, that when her breast was removed, she would not experience the same feelings of loss and mourning that younger women who are subjected to mastectomy experience. I made an age-related assumption that turned out to be false for this particular patient.

Gypsies

Gypsies, because of their reputation for stealing, lying, and not paying their bills, are frequently denied access to office medical care. They used to camp in the community where I practiced and found their way to my office. They did not conform to the picture both the medical and lay community painted of them. They paid their bills slowly but they paid as well as other economically strapped patients. There was never anything missing after their visits, and they were as truthful as the rest of us. However, their reputation made it difficult for them to get appointments at doctors' offices for both their children and for sick adults.

Prostitutes

Prostitutes have an image of being hardened, noncaring people, whose trade is the lowest on the rank of professions. After the Second World War, I became acquainted with several girls who survived the war years by prostituting. They were little girls who played, laughed, and cried as their peers did but who had desensitized a part of their body and the emotions associated with it in order to stay alive. They were caring youngsters, who did not fit the picture painted of prostitutes in our society. It was my responsibility to take them to a physician's office. We were treated with disrespect and discourtesy. The girls were not asked how they happened to acquire their trade; they were automatically put in a category of sluts and harlots who deserve to be rejected and to live outside the human community. In this instance, the particular doctor we saw did not take time to look at each girl as an individual. Doctors have the same prejudices as others in society.⁹

Medicine, as law enforcement, must examine how conclusions about individuals are reached. Are facts or preconceived notions used in the reasoning process? "... [T]he drive to improve patient safety is requiring us to unlearn a set of assumptions and behaviors"¹⁰ and increase attentiveness to the individual's needs and characteristics. We are fallible as human beings, and all carry group pictures in our minds; but when we peruse the group, we must analyze the individual carefully and not attribute traits to one person that we think the group as a whole may possess; "prediction, however reliable in the aggregate, is notoriously uncertain at the individual level."¹¹ Further, we must be careful that the group picture is not based on rumor, hearsay, and fear. A physician especially must be willing to become acquainted with each patient, listen to the entire story that unravels, and write notes and impressions on a blank page. This is difficult because doctors, as law enforcement officers, bring with them family and learned cultural concepts about any individual who can be labeled as part of a stereotyped group. To reexamine these concepts is hard and needs to be a lifelong process. It involves combating the daily exposure to political and societal stereotypes in the media and elsewhere. Usually these turn out to be exaggerated or false, based on fear and ignorance, but are difficult to erase from our thinking once established. To avoid errors, a physician needs to exercise great caution and not fall into the trap of basing medical diagnosis on prejudicial judgments and patient profiling.

References

1. Milosz C. Milosz's ABC's. Translated by MG Levin. Creative Nonfiction;16:18.
2. Pinkowish MD. Evaluating the pateint with chest pain. Patient Care 2000 Mar 15. Available on the World Wide Web (accessed October 11, 2001): http://pc.pdr.net/pc/public.htm?path=content/journals/p/data/2001/0315 /03a01chestpain.html.
3. Diamant AL, Wold C, Spritzer K, Gelberg L. Health behaviors, health status, and access to and use of health care: a population-based study of lesbian, bisexual, and heterosexual women. Arch Fam Med 2000 Nov-Dec;9(10):1043-51.
4. Byrd WM, Clayton LA. An American health dilemma: a medical history of African Americans and the problem of race. New York: Routledge; 2000.
5. All Rights Reserved. American Civil Liberties Union/ACLLU of CO. 3-2001:1.
6. Purdy M. Ignoring, and then embracing , the truth about racial profiling. The New York Times New York Report 3-11-2001:23.
7. Doyle Sir AC. The Complete Sherlock Holmes Vol 1: Doubleday & Co, Inc, Garden City NJ 1927;137.
8. Clark KA, Martin SL, Petersen R, et al. Who gets screened during pregnancy for partner violence? Arch Fam Med 2000 Nov-Dec;9(10):1093-9.
9. Albert T. Accept no barrier: new physician finds some see only blindness. Am Med News 2001 Jan 15. Available on the World Wide Web (accessed October 11, 2001): http://www.ama-assn.org/sci-pubs/amnews/pick_01/prsa0115.htm.
10. Marcus LJ, Dorn BC. Facing change? Get everyone to do some "unlearning." Am Med News 2001 Jan 15. Available on the World Wide Web (accessed October 11, 2001): http://www.ama-assn.org/sci-pubs/amnews/pick_00/prca1225.htm.
11. Hunter KM. A science of individuals. Medicine and uncertainty. Hosp Pract (Off Ed) 1992 May 15;27(5):183-6, 194-200, 205-8 passim.

To Soul of the Healer Index >>