Politically incorrect medicine
Article by Ilana Yurkiewicz When should medicine talk about race? published in Scientific American Blogging Network.
Translated by Evgenia Ryabtseva
The concept of race constantly pops up when solving various medical issues. The characteristics of diseases currently, as a rule, contain information about which populations are most and least susceptible to them, and drug descriptions – about which patients of which ethnicity respond better or worse to treatment.
Such information is so common that it is very easy to accept it as having unambiguous evidence. However, the issue of race in medical practice remains the subject of heated discussions. With the appearance of each new study, the results of which are presented depending on race, groups of its supporters and critics inevitably form.
The question under discussion is: is there a place for race in medicine?
Recently, there has been a growing number of scientists who believe that we should completely abandon this way of thinking. Many argue that the concept of race is closer to a social structure than to a biological category. This gives rise to a justified fear that talking about differences between races forms an idea of the biological nature of these differences. Even if there are certain differences between populations, it is obvious that the race indicated by the patient is, at best, a rough approximation to reality. Moreover, research very often does not take into account variables other than genetic origin, such as socio-economic status, cultural characteristics and discrimination. This means that when demonstrating differences, there is an involuntary tendency to believe that biological features may overlap important differences in environmental conditions that deserve close attention. There are also social issues generated by the fact that historically, when studying this issue, the concept of ethnicity was often distorted by pseudo-scientists with racist views, which led to a demonstration of the superiority of "whites" or "true Aryans" over "lower races". In light of this, the need for extreme sensitivity in using race as a medical variable is obvious and justified.
Part of the problem is due to the fact that some scientists completely ignore this question. They classify any data they receive concerning any health-related issue, regardless of race, as others have followed this principle before. However, when processing data, the analysis must be carried out in full. There is no such thing as "just facts", since there is no such thing as predefined datasets that we either provide or hide. In relation to each question, we make a choice. Collecting, sorting, and presenting data presupposes a choice. When comparing groups, we can draw dividing lines where we see fit. With regard to this issue, it should be mentioned that in many studies that consider race, only "whites" and "blacks" are compared, while other ethnic groups, including patients with mixed race, are completely ignored.
If a researcher is faced with a choice, he is obliged to approach the issue with full responsibility. It is absolutely not enough to explain the benefits of carrying out the project with any of the versions of the following wording: "this complements the literary data, showing us something that we did not know before." We don't know a huge number of things. Research should have value. When making any decision, first of all it is necessary to determine what is the essence of the question? Do the differences I am trying to demonstrate have any significance? Is the information obtained applicable for prevention, diagnosis, control or treatment?
Sometimes the answer is unequivocally positive. There are cases when considering race, even in a rough approximation, has benefited patients. Knowing that sickle cell anemia is most common among representatives of African peoples inhabiting the territories south of the Sahara Desert can make it easier for a doctor to diagnose the disease and increase the effectiveness of its therapy. Since Tay-Sachs disease is a genetic disease that is quite common among Ashkenazi Jews, Jewish communities have long and willingly resorted to genetic testing of future parents, which has significantly reduced the incidence of the disease. People of Asian origin are relatively often carriers of certain single nucleotide polymorphisms that cause a slow metabolism of drugs. This means that in order to achieve the desired effect and prevent the toxic effect of the drug, they need reduced doses compared to the traditionally recommended ones. There are still quite a large number of similar examples. Therefore, in this case, it is not superfluous to repeat that, while racial and even more so ethnicity is only a rough approximation of genetics, the medical value of knowledge of such trends is beyond doubt.
Medicine is an area in which heuristic approaches or rules of thumb are often used to help focus on the best guess in the absence of the possibility of conducting a detailed search. The frequent use of these approaches is due to the fact that medicine is simultaneously characterized by information overload and lack of time. The best assumptions in medicine are probabilistic; to select the most and least likely options, doctors collect information from various sources. Every test, every new piece of information helps to make this choice. Therefore, some experts are of the opinion that, since doctors base their best assumptions on the complex of symptoms and the results of the examination, race can also be used as an approximate guideline. Since the application of the principles of heuristics is fraught with errors, the way out in this case is not to completely abandon them, but to increase the awareness of doctors about the possibility of such errors to eliminate them as much as possible and increase the effectiveness of the application of empirical rules.
From the point of view of medicine, race is a very sensitive issue that requires an appropriate approach. It should be noted that, in contrast to the shameful periods of history during which race was viewed from unethical points of view, the overwhelming majority of modern research is based solely on good intentions, the purpose of which is to adapt medical care to the needs of a diverse population of patients. However, in order to avoid abuse, every researcher who wants to talk about race should take responsibility for a clear explanation of what benefits the data they receive will bring to humanity.
This raises the question: what should be the next step? Suppose a researcher feels that classifying results based on origin can improve the quality of medical care for patients. What then? How should he properly describe the role of race in medicine?
The answer to this question is contained in an article published in 2003 by Judith B. Kaplan and Trude Bennett in the journal JAMA. The authors formulated seven meaningful recommendations concerning the attitude to race in biomedical research. The authors' conclusions are given below.
1. If race/ethnicity is used as a variable during the study, the reason for choosing this particular variable should be indicated.
The authors explain this by the fact that the danger of using race without explanation is that such an approach can be considered as classifying the study participants "by default".
2. When quoting information about race/ethnicity from any source, the method by which individuals were divided into racial/ethnic categories should be described. If the race/ethnicity was determined by the participants themselves, the authors should indicate whether the participants answered the question without the proposed set of possible answers or chose from a limited number of categories.
The way the questions are formulated affects the result. For example, the authors of the article note that when choosing from a limited number of categories, participants may not identify themselves with any of the options. If such individuals mark something only because there is no more suitable option, it can distort the results. Moreover, the authors explain that participants in different studies are offered a choice of different sets of answers, which makes it difficult to compare the results. Transparency of the methodology can help in solving this problem.
3. Race/ethnicity should be used as a substitute for genetic variation. Claims of genetic differences must be supported by evidence obtained during genetic research. Genetic hypotheses should be firmly supported by existing data, clearly formulated and thoroughly tested.
Differences between races can be due to various reasons: mixed origin, socio-economic status, cultural differences and discrimination are far from an exhaustive list. Considering that when demonstrating the differences between groups, society still tends to an explanation that sounds something like "Aha! It's all biology!", researchers should clearly indicate that racial differences as such do not say anything about genetic differences. The only tool that can somehow explain the genetic differences is, not surprisingly, genetics.
4. When formulating a hypothesis and describing the results of the study, the authors should distinguish between race/ethnicity as a risk factor and race/ethnicity as a risk marker.
While epidemiologists have not come to a consensus on clear definitions of these terms, one of the options is the following: a risk marker indicates the existence of a correlation, whereas a risk factor implies the existence of a causal relationship. Thus, if race is statistically significantly associated with the outcome, but is not involved in the causal mechanism, it should be considered as a marker.
5. When interpreting racial/ethnic differences, all factors potentially relevant to the issue should be taken into account, including racism and discrimination, socio-economic status, social class, personal or family well-being, environmental factors, type of health insurance, age, diet, attitude to health status and relevant features behavior, level of education, language of communication, religious views, belonging to a particular tribe, country of birth, country of birth of parents, duration of residence in the country of permanent residence and place of residence.
This and the following recommendations are interrelated with recommendation 3. There are many reasons why it is possible to identify differences between racial and ethnic groups. This recommendation is critical, since, if we do not consider these variables, when reading the article, we can conclude that the detected differences are due to biological characteristics. This conclusion is not only wrong, but may also distract from differences in environmental conditions worthy of our attention.
6. Since the absence of corrections for socio-economic status or belonging to a particular social class is the main reason for the appearance of errors in the study of racial/ethnic differences, when comparing racial/ethnic groups, researchers should make maximum efforts when making amendments to the conceptually relevant features of socio-economic status or belonging to that or another social class. The results obtained without amendments should be clearly marked and, for comparison, presented in combination with the corrected data.
Making amendments to all the variables relevant to the issue is a very difficult task. However, researchers are obliged to make maximum efforts, especially with regard to factors (such as socio-economic status) that are a stable source of errors.
7. When describing racial/ethnic groups, researchers should not use "stigmatizing" terminology that reflects unscientific classification systems and implies that race/ethnicity is a specific unchangeable characteristic of a person.
The authors give an example of using the term "native of Asia" instead of "Oriental" or "Asian" origin. In addition to condemning the use of stigmatizing terminology, they also condemn the use of overly extensive definitions.
***
What is the outcome? It is necessary to consider the issue as carefully as possible; be extremely accurate; treat the study participants with respect, since the ultimate goal of medical research is to help patients.
09.01.2013