Socially Constructed Risk Factors in Medicine
Medical schools need to stop teaching their students that race, sex or gender are risk factors for disease. Stopping our inquiry at the label of black, male, or woman, denies us the possibility of discovering the real reasons behind group differences in disease risk or outcome. It also leaves out the large contingent of our population who are mixed race, transgender or intersex, and don’t fit neatly into the boxes we create. At the root of this problem is the variables used in medical research, but medical schools are also at fault for failing to demand better of the research they are using to teach our next generation of doctors. Studies routinely identify participants by race and ethnicity and draw conclusions based on those identifiers without ever defining what either term means. Sex and gender are also rarely defined, often used interchangeably and stand as surrogates for body size, fat mass, and hormone levels.
I am not arguing that there are no differences in medical outcomes between blacks and whites, or men and women. It has been well established that black Americans routinely have worse health outcomes than their white peers [1], but claiming that the difference is a result of their skin color lets us conveniently ignore factors such as the stress of racism or the worse air quality in majority black neighborhoods. In assessing differences across sex or gender, our lack of correct terminology combined with our lack of further questioning leaves me as a medical student woefully unprepared to provide adequate care to the non-binary patients I will see. Do my female patients need different drug dosing because of their presumed smaller size? Because of their estrogen levels? Our half-hearted effort to create individualized medicine can actually result in harming patients and giving the wrong care.
One prominent example of race being unreasonably inserted into diagnostic evaluation is the estimated glomerular filtration rate to evaluate kidney function[2]. The standard equation used in hospitals across the country is the MDRD equation, which has just 4 variables. Creatinine is a measured level of a protein in the blood that builds up when kidneys are working less efficiently. The other three variables are age, sex, and black race. Without going into significant detail, the absurdity of this equation is still easily shown with the simple question -how black does my patient need to be to use the black race variable?
Arguments in favor of using race or gender as factors in research often hinge on the assumption that people of a given race or gender have commonalities that are not held by those outside that group. But many of those commonalities are not in fact exclusive to the group in question, and members of the group might not have that trait. Uterine cancer for example is certainly more common in women than men, but it is not actually the patient’s identity as a woman that puts them at risk, it is the fact that they have a uterus. I will have future patients who are women without uteruses (whether because they are trans or because they have had a hysterectomy), and I will have patients who are trans men with uteruses. Similarly sickle cell disease is often associated with black skin, but in fact is the result of a single gene mutation that can be present in a patient of any skin color. In our increasingly international world, even diseases that have historically been linked to ancestral geographic regions that may be tightly correlated with skin color are increasingly prevalent in patients across the continents regardless of their perceived race. Doctors need to be taught to associate diseases with their actual root causes, not to use faulty shortcuts like race or gender.
My vision for medical research and education is one in which every socially constructed factor is followed up on until the true explanation(s) can be uncovered. If there is a difference by sex, is it hormones, body size, having a uterus, sexism? If a difference is found by race, how was race defined? Would country of origin (often a factor for infectious disease risk) have been a better factor? Did we account for socioeconomic status, housing, and access to preventive care? And if those were found to explain the racial difference, are we taking steps to reduce those disparities? If a research project is unable to follow the trail to the bottom, it should offer up at least three plausible explanations for why the differences might have been seen. This will facilitate future research and also make apparent to all readers that although the difference was found on, for example, the lines of self-described race, the authors are not claiming that race itself was the causal factor.
Another solution that might be offered is to simply erase factors like race, gender, or sex from our medical teaching and research at all. This amounts to a colorblind solution that pretends that ignoring the reality of race and gender that we as a society have built can make it disappear. As long as racial disparities exist in medicine, race is something we will need to continue acknowledging, but we need to acknowledge it thoughtfully and constructively. We need to ask ourselves what the mechanisms could be that would cause a given race to have worse outcomes and when the answer points to issues at the societal level, we need to make societal changes to address those disparities. Research is supposed to be about unearthing truths, not obfuscating them with socially constructed variable and medicine is supposed to be about finding the root causes of illnesses so that we can heal our patients, not blaming their skin color or their gender for their disease. Let’s start doing our jobs.
[1] Levine, R. S., Foster, J. E., Fullilove, R. E., Briggs, N. C., Hull, P. C., Husaini, B. A., et al. (2001). Black-white inequalities in mortality and life expectancy, 1933–1999: Implications for Healthy People 2010. Public Health Reports, 116, 474–483
[2] Levey, A. S., Bosch, J., Lewis, J., Greene, T., Rogers, N., & Roth, D. (1999). A More Accurate Method To Estimate Glomerular Filtration Rate from Serum Creatinine: A New Prediction Equation. Annals of Internal Medicine, 130(6), 461. doi:10.7326/0003–4819–130–6–199903160–00002
Image from: https://inkbotdesign.com/healthcare-branding/
