White patients’ physical responses to healthcare treatments are influenced by provider race and gender

The healthcare workforce in the United States is becoming increasingly diverse, gradually shifting society away from the historical overrepresentation of White men among physicians. However, given the long-standing underrepresentation of people of color and women in the medical field, patients may still associate the concept of doctors with White men and may be physiologically less responsive to treatment administered by providers from other backgrounds. To investigate this, we varied the race and gender of the provider from which White patients received identical treatment for allergic reactions and measured patients’ improvement in response to this treatment, thus isolating how a provider’s demographic characteristics shape physical responses to healthcare. A total of 187 White patients experiencing a laboratory-induced allergic reaction interacted with a healthcare provider who applied a treatment cream and told them it would relieve their allergic reaction. Unbeknownst to the patients, the cream was inert (an unscented lotion) and interactions were completely standardized except for the provider’s race and gender. Patients were randomly assigned to interact with a provider who was a man or a woman and Asian, Black, or White. A fully blinded research assistant measured the change in the size of patients’ allergic reaction after cream administration. Results indicated that White patients showed a weaker response to the standardized treatment over time when it was administered by women or Black providers. We explore several potential explanations for these varied physiological treatment responses and discuss the implications of problematic race and gender dynamics that can endure “under the skin,” even for those who aim to be bias free.

The face of medicine is changing. Women and people of color make up an increasing percentage of health care providers (1–3). In 2017, for the first time in history, women were the majority of accepted medical school applicants in the United States and the number of non-White accepted applicants rose to above 50%. Here, we ask whether this recent demographic shift in the race and gender of doctors is also shifting long-held, societally pervasive notions of what a doctor “looks like.”Despite the increasing diversity of the medical field, for most people in most contexts, the association between “doctor” and “White man” is still likely strong and pervasive. This is hardly surprising. For most of medical history in North America, the majority of physicians fit this profile (see Fig. 1 A and B), and even now the majority of practicing physicians are still men and nearly half are White (see Fig. 1 C and D). Consequently, the emerging links between “Doctor and Woman” and between “Doctor and Black person,” for example, are likely weak. Moreover, to the extent that those associations exist, they are likely to have to compete for attention with an array of strong, frequent, and negative associations that undermine the links between women and competence and African Americans and competence (4–6).Open in a separate windowFig. 1.The change in the representation of women (A) and people color (B) in the number of accepted applicants to US medical schools, as well as the current representation of professionally active women physicians (C) and physicians of color (D). (A and B) From the Association of American Medical Colleges (AAMC). (C) From the Henry J. Kaiser Family Foundation. (D) From 2013 from the Association of American Medical Colleges (AAMC). AAMC data on race/ethnicity were not available for 2013 or 2014, hence explaining the gaps in the graph around these years in B.In patient–provider interactions, as in every social encounter, people bring with them a set of learned associations about social groups that have been formed by their various life experiences (e.g., personal interactions, media exposure) (6–12). Mirroring the historical representation of doctors in actual medical practice, representations of doctors in popular media have overwhelmingly been as White men (13–15). Patients who have learned this societally pervasive “Doctor = White man” association through their actual encounters with physicians as well as through movies, television, books, and advertising may respond less positively to care from Black and women providers. These associations may exist at an implicit level even in the context of positive explicit attitudes toward Black doctors and women doctors (16, 17), and they are potentially powerful, influencing the course of medical care. Also, while it is clear from past research that being a target of bias can be harmful to health (e.g., people who face race-based discrimination face adverse physical and mental health consequences) (18), it is unclear whether viewing another social group in light of societally pervasive associations (e.g., about doctors on the basis of gender and race) can be harmful to the health of the perceiver.Here, we focus on how the race and gender of doctors may impact patients’ responses to the expectations doctors set about medical treatment. Previous research shows that a provider’s expressed expectations for a medical treatment (i.e., that it will benefit patients) can improve patient engagement, adherence, and physiological responses to treatment (19–25). Based on these findings, we anticipate that patients who interact with a doctor whose personal characteristics (e.g., race, gender) do not conform to dominant societal representations of what a doctor looks like may be less responsive to such expectations. We hypothesize that patients may be less responsive to the exact same medical treatment when the doctor who sets expectations that this treatment will be beneficial is not a White man.This hypothesis draws on a large and growing body of research suggesting that the total effect of a healthcare treatment depends on the social context in which that treatment takes place (25–29). The realization that the social context can influence treatment and medical outcomes is bolstered by a large body of research on the placebo effect (26). Although people may sometimes assume that actual pharmaceutical properties of a medication or treatment are solely responsible for its total benefit, placebo paradigms show that the total effect of treatment is in fact a combined product of the drug and their medical properties (e.g., acetaminophen, antihistamines), the body’s natural healing abilities (e.g., endogenous opioids and antihistamines), and the psychological and social context (e.g., what a patient believes about treatment and the qualities of the person who administers the treatment) (SI Appendix, Fig. S1). For example, past research suggests that a physician’s characteristics, such as their projected warmth and competence, influences how much a patient improves in response to treatment. In one recent study (22), the researchers independently manipulated whether a provider acted more or less warm, and more or less competent, toward a patient during an allergy skin prick test that induced a mild allergic reaction. The provider set positive expectations about a placebo cream (i.e., unscented hand lotion) placed on the reaction, informing patients that this cream was an antihistamine that would reduce the reaction. When the provider was both warm and competent, patients showed a stronger physiological response to the placebo treatment over time; their allergic reaction decreased the most rapidly in size, in response to the positive expectations that the provider had set. Thus, aspects of social interactions with providers can influence the degree to which the positive expectations that a provider sets about treatment ultimately influence physiological treatment response.As in most social interactions in the United States, race and gender are likely salient aspects of the social context in patient–provider interactions (30, 31, 32). Previous research has found, for example, that patient race can influence the quality of care received from doctors in myriad ways (33–36). Here, we focus on provider race and provider gender as features of the social context that can influence patients’ response to treatment. Specifically, we ask the following: will White patients exhibit a weaker physiological response to the expectations set about treatment by doctors who are not White and men?