Prejudice in Perceptions of Physicians?: The Influence of Race
and Gender on Evaluations of Medical Errors
N. Derek Brown, B.A.
, Larry R. Martinez, Ph.D.
,andMichelleBMikki^ R. Hebl, Ph.D.
Department of Psychology, Portland State University, Portland, OR, USA;
Rice University, Houston, TX, USA.
KEY WORDS: Physician race; Physician gender; Medical errors; Patient
J Gen Intern Med 33(6):807–8
© Society of General Internal Medicine 2018
Little is known about prejudicial attitudes patients have
toward physicians. Past research shows that physicians
commonly face discrimination
, and although Asian
physicians constitute the largest ethnic minority popula-
tion of practicing physicians,
, there is no research that
examines their experiences with discrimination explicitly.
Asian professionals are subject to both positive and neg-
ative stereotypes, such that they are categorized as high-
achieving (e.g., intelligent, successful), yet also perpetu-
ally foreign and never accepted into the American mi-
This research examined how patient biases can
influence evaluations of physicians as a function of phy-
sician race, gender, and severity of a medical error.
We recruited 482 adults (59% female) using Amazon’s
Participants were randomly assigned to view
one website and one incident report that we altered to
manipulate physician ethnicity (White vs. Indian), gender
(male vs. female), and medical error severity (major vs.
minor error). We assessed participant evaluations of the
physician with questions about patient care characteristics
(α = .86) and the appropriateness of various consequences
that could be Btaken against the physician^ (α =.89). We
compared mean differences based on participant impres-
sions as a function of error severity and physician eth-
nicity and gender (MANOVA) and assessed for potential
relationships between physician demographic characteris-
tic and suggested consequences using regression analysis
(Hayes’ PROCESS Macro Model 4). An aprioripower
analysis (with α = .05 and power = .95) revealed an ade-
quate sample size to detect differences in effect sizes as
small as .025.
The mean age of participants was 37.09 (SD = 12.54) and
77% of participants were White/Caucasian. Table 1 pro-
vides means, standard deviations, reliabilities, and correla-
tions of the variables included in the analyses. There were
no differences in care quality perception or suggested con-
sequences for severe errors between race and gender
groups (Table 2). For all groups, patient care perceptions
were more negative and consequences for major errors
were more severe compared to minor ones (F(2,473) =
38.97, p <.001).
For minor errors, there was no difference between pa-
tient care perception or desired consequences for White or
Indian men or Indian women. However, White women
were viewed more positively in terms of their provision
of patient care (F(1, 474) = 4.77, p < .05, η
= .01) com-
pared to other physicians, evaluated as a group (M = 4.24,
SD = 1.08). Similarly, White women were treated more
leniently in terms of suggested consequences (F(1,
474) = 4.52, p < .05, η
= .01) compared to the other physi-
cians (M =2.94, SD = 1.32). For major errors, there were
no differences based on race or gender.
In our multivariable model, we found that the leniency for
White female physicians could be explained by the perception
that White women have greater ability to provide care follow-
ing a minor error, b = − 0.19 (SE = 0.09), 95% CI [− 0.40, −
Our results suggest that evaluations of physicians are not
only linked to the severity of the error committed, but also
to physician gender and race. Unexpectedly, we found that
Indian physicians were rated no differently than White
male physicians, despite having sufficient power to detect
an effect. Results also revealed differential evaluations for
White female physicians in comparison to White male,
Indian male, and Indian female physicians, though only
for minor errors. Specifically, White female physicians
were rated as better caretakers and elicited the least severe
sanctions (despite equivalent performance).
Our data suggest that physician race and gender may be
instrumental in an individual’s evaluation toward
Published online March 12, 2018