+1443 776-2705 panelessays@gmail.com


Module part

Module 1b: Introduction

Researchers at Harvard have developed over a dozen tests for measuring implicit bias related to race, sexuality, disability, religion, and other forms of prejudice as part of Project Implicit. Visit the projects Take a Test site (link below) and take TWO of the IATs (one must be the Race IAT) (you may also choose to take other tests). After you view your results, reflect on the test itself, your experience taking the test, and your interpretation of the results.


Answer the following 6 questions in the submission text box

  1. Summarize the module readings in one paragraph (don’t be vague)
  2. Define implicit bias in your own words
  3. What additional test did you take? What were your results for both tests?
  4. Have you ever thought about what biases you have and how they shape how you interact with others? why or why not? 
  5. What are your thoughts about this activity? (feelings, responses, etc)
  6. Was this activity eye-opening? why or why not?


The Journal of Infectious Diseases

S62 • jid 2019:220 (Suppl 2) • Marcelin et al

Correspondence: J. R. Marcelin, MD, Division of Infectious Diseases, University of Nebraska
Medical Center, 985400 Nebraska Medical Center, Omaha, NE 68198-5400 ([email protected]

The Journal of Infectious Diseases® 2019;220(S2):S62–73
© The Author(s) 2019. Published by Oxford University Press for the Infectious Diseases Society
of America. All rights reserved. For permissions, e-mail: [email protected]
DOI: 10.1093/infdis/jiz214

The Impact of Unconscious Bias in Healthcare: How to
Recognize and Mitigate It
Jasmine R Marcelin,1, Dawd S. Siraj,2 Robert Victor,3 Shaila Kotadia,3 and Yvonne A Maldonado3

1University of Nebraska Medical Center, Omaha; 2University of Wisconsin, Madison; and 3Stanford University School of Medicine, California

The increasing diversity in the US population is reflected in the patients who healthcare professionals treat. Unfortunately, this
diversity is not always represented by the demographic characteristics of healthcare professionals themselves. Patients from under-
represented groups in the United States can experience the effects of unintentional cognitive (unconscious) biases that derive from
cultural stereotypes in ways that perpetuate health inequities. Unconscious bias can also affect healthcare professionals in many
ways, including patient-clinician interactions, hiring and promotion, and their own interprofessional interactions. The strategies
described in this article can help us recognize and mitigate unconscious bias and can help create an equitable environment in health-
care, including the field of infectious diseases.

Keywords. Unconscious bias; diversity and inclusion; mitigating strategies.

There is compelling evidence that increasing diversity in the
healthcare workforce improves healthcare delivery, espe-
cially to underrepresented segments of the population [1, 2].
Although we are familiar with the term “underrepresented mi-
nority” (URM), the Association of American Medical Colleges,
has coined a similar term, which can be interchangeable:
“Underrepresented in medicine means those racial and ethnic
populations that are underrepresented in the medical profes-
sion relative to their numbers in the general population” [3].
However, this definition does not include other nonracial or
ethnic groups that may be underrepresented in medicine, such
as lesbian, gay, bisexual, transgender, or questioning/queer
(LGBTQ) individuals or persons with disabilities. US census
data estimate that the prevalence of African American and
Hispanic individuals in the US population is 13% and 18%, re-
spectively [4], while the prevalence of Americans identifying as
LGBT was estimated by Gallup in 2017 to be about 4.5% [5]. Yet
African American and Hispanic physicians account for a mere
6% and 5%, respectively, of medical school graduates, and ac-
count for 3% and 4%, respectively, of full-time medical school
faculty [6]. As for LGBTQ medical graduates, the Association
of American Medical Colleges does not report their preva-
lence [6]. Persons with disabilities are estimated to be 8.7% of
the general population [4], while the prevalence of physicians
with disabilities has been estimated to be a mere 2.7% [7].

Furthermore, although women currently outnumber men in
first-year medical school classes [8], gender disparities still exist
at higher ranks in women’s medical careers [9–11].

Unconscious or implicit bias describes associations or
attitudes that reflexively alter our perceptions, thereby af-
fecting behavior, interactions, and decision-making [12–14].
The Institute of Medicine (now the National Academy of
Medicine) notes that bias, stereotyping, and prejudice may play
an important role in persisting healthcare disparities and that
addressing these issues should include recruiting more med-
ical professionals from underrepresented communities [1].
Bias may unconsciously influence the way information about
an individual is processed, leading to unintended disparities
that have real consequences in medical school admissions,
patient care, faculty hiring, promotion, and opportunities for
growth (Figure 1). Compared with heterosexual peers, LGBT
populations experience disparities in physical and mental health
outcomes [15, 16]. Stigma and bias (both conscious and uncon-
scious) projected by medical professionals toward the LGBTQ
population play a major role in perpetuating these disparities
[17]. Interventions on how to mitigate this bias that draw roots
from race/ethnicity or gender bias literature can also be applied
to bias toward gender/sexual minorities and other underrepre-
sented groups in medicine.

The specialty of infectious diseases is not free from
disparities. Of >11 000 members of the Infectious Diseases
Society of America (IDSA), 41% identify as women, 4% identify
as African American, 8% identify as Hispanic, and <1% identify
as Native American or Pacific Islander (personal communica-
tion, Chris Busky, IDSA chief executive officer, 2019). However,
IDSA data on members who identify as LGBTQ and members
with disabilities are not available.



nloaded from



62/5552356 by guest on 20 July 2021

Unconscious Bias: Challenges and Solutions • jid 2019:220 (Suppl 2) • S63

The 2017 IDSA annual compensation survey reports that
women earn a lower income than men [18], and a review of
the full report demonstrates similar disparities among URM
physicians, compared with their white peers [19]. While it
may not be feasible to assign a direct causal relationship be-
tween unconscious bias and disparities within the infectious
diseases specialty, it is reasonable and ethical to attempt to ad-
dress any potential relationship between the two. In this article,

we define unconscious bias and describe its effect on health-
care professionals. We also provide strategies to identify and
mitigate unconscious bias at an organizational and individual
level, which can be applied in both academic and nonacademic


Even in 2019, overt racism, misogyny, and transphobia/homo-
phobia continue to influence current events. However, in the
decades since the healthcare community has moved toward
becoming more egalitarian, overt discrimination in medi-
cine based on gender, race, ethnicity, or other factors have
become less conspicuous. Nevertheless, unconscious bias still
influences all human interactions [13]. The ability to rapidly
categorize every person or thing we encounter is thought to be
an evolutionary development to ensure survival; early ancestors
needed to decide quickly whether a person, animal, or situation
they encountered was likely to be friendly or dangerous [20].
Centuries later, these innate tendencies to categorize everything
we encounter is a shortcut that our brains still use.

Stereotypes also inadvertently play a significant role in med-
ical education (Figure 1). Presentation of patients and clinical
vignettes often begin with a patient’s age, presumed gender,
and presumed racial identity. Automatic associations and mne-
monics help medical students remember that, on examination,
a black child with bone pain may have sickle-cell disease or a
white child with recurrent respiratory infections may have
cystic fibrosis. These learning associations may be based on true
prevalence rates but may not apply to individual patients. Using
stereotypes in this fashion may lead to premature closure and
missed diagnoses, when clinicians fail to see their patients as
more than their perceived demographic characteristics. In the
beginning of the human immunodeficiency virus (HIV) epi-
demic, the high prevalence of HIV among gay men led to ini-
tial beliefs that the disease could not be transmitted beyond the
gay community. This association hampered the recognition of
the disease in women, children, heterosexual men, and blood
donor recipients. Furthermore, the fact that white gay men
were overrepresented in early reported prevalence data likely
led to lack of recognition of the epidemic in communities of
color, a fact that is crucial to the demographic characteristics
of today’s epidemic. Today, there is still no clear solution to
learning about the epidemiology of diseases without these im-
precise associations, which can impact the rapidity of accurate
diagnosis and therapy.


Unconscious bias describes associations or attitudes that un-
knowingly alter one’s perceptions and therefore often go unrec-
ognized by the individual, whereas conscious bias is an explicit
form of bias that is based on one’s discriminatory beliefs and

Figure 1. Glossary of key terms.

Glossary of key terms used in discussion of uncon-
scious bias

Active bystander—A person who witnesses a situation,
acknowledges the potential problem, and speaks up about
it [59]

Bias—Tendency to favor one group over another; biases
can be favorable or unfavorable and can be unconscious
(implicit or unintentional) or conscious (explicit or inten-
tional) [14]

Cultural humility—Defined by its ongoing self-reflection:
a lifelong commitment to continuously evaluate one’s own
behaviors, beliefs, and identities and determine how poten-
tial biases and assumptions may surface when collaborating
with an individual of a different background [72]

Intent vs impact—Concept that the focus of behavioral
change should consider the impact on the recipient regard-
less of the intent of the offending behavior (ie, whether a
result of unconscious or conscious bias) [59]

Microaggression—“Brief and commonplace daily verbal/
nonverbal behavioral, and environmental indignities
whether intentional or unintentional that communicate
hostile, derogatory or negative racial/ethnic, gender, sexual
orientation, and religious slights and insults” [73], (p. 271);
these can occur wherever people are perceived as “other”;
some groups have a lifetime burden of microaggressions
that can contribute to physical or psychological illness

Prejudice—Outward expressions of negative attitudes
towards different social groups [20]

Stereotype—An oversimplified, fixed, and widely held be-
lief about an entire group of people; stereotypes may not
always be accurate, especially when they lead to judgments
applied to individuals within that group [14]

Unconscious bias—Attitudes or stereotypes that un-
consciously alter our perceptions or understanding of our
experiences, thereby affecting behavior, interactions, and
decision-making [12–14]

Underrepresented minority—Understood to mean
either underrepresented minorities or underrepresented in


nloaded from



62/5552356 by guest on 20 July 2021

S64 • jid 2019:220 (Suppl 2) • Marcelin et al

values and can be targeted in nature [14]. While neither form
of bias belongs in the healthcare profession, conscious bias
actively goes against the very ethos of medical professionals
to serve all human beings regardless of identity. Conscious
bias has manifested itself in severe forms of abuse within the
medical profession. One notable historical example being the
Tuskegee syphilis study, in which black men were targeted to
determine the effects of untreated, latent syphilis. The Tuskegee
study demonstrated how conscious bias, in this case manifested
in the form of racism, led to the unethical treatment of black
men that continues to have long-lasting effects on health equity
and justice in today’s society [21]. Given the intentional nature
of conscious bias, a different set of tools and a greater length of
time are likely required to change one’s attitudes and actions.
Tackling unconscious bias involves willingness to alter one’s
behaviors regardless of intent, when the impact of one’s biases
are uncovered and addressed [22]

There is still debate, however, about the degree to which
unconscious bias affects clinician decision-making. In one
systematic review on the impact of unconscious bias on health-
care delivery, there was strong evidence demonstrating the
prevalence of unconscious bias (encompassing race/ethnicity,
gender, socioeconomic status, age, weight, persons living with
HIV, disability, and persons who inject drugs) affecting clinical
judgment and the behavior of physicians and nurses toward
patients [12]. However, another systematic review found only
moderate-quality evidence that unconscious racial bias affects
clinical decision-making [23]. A detailed discussion of the im-
pact of unconscious bias on healthcare delivery is out of the
scope of this article, which is focused on the impact of uncon-
scious bias as it relates to healthcare professionals themselves.
Nevertheless, strategies to mitigate the effects of unconscious
bias (discussed later) can be applied to healthcare delivery and
patient interactions.


While we know that unconscious bias is ubiquitous, it can
be difficult to know how much it affects a person’s daily
interactions. In many cases, an individual’s unconscious
beliefs may differ from their explicit actions. For example,
healthcare professionals, if asked, might say they try to treat
all patients equally and may not believe they hold negative
attitudes about patients. However, by definition, they may
lack awareness of their own potential unconscious biases, and
their actions may unknowingly suggest that these biases are

To measure unconscious bias, Drs Mahzarin Banaji and
Anthony Greenwald developed the IAT in 1998 [24]. Many
versions of the IAT are accessible online (available at: https://
implicit.harvard.edu/implicit/), but one of the most studied
is the Race IAT. The IAT has been extensively studied as an

inexpensive tool that provides feedback on an individual biases
for self-reflection. The IAT calculates how quickly people asso-
ciate different terms with each other. To determine unconscious
race bias, the race IAT asks the subject to sort pictures (of white
and black people) and words (good or bad) into pairs. For ex-
ample, in one part of the Race IAT, participants must associate
good words with white people and bad words with black people.
In another part of the Race IAT, they must associate good words
with black people and bad words with white people. Based on
the reaction times needed to perform these tasks, the software
calculates a bias score [20, 24]. Category pairs that are uncon-
sciously preferred are easier to sort (and therefore take less time)
than those that are not [24]. These unconscious associations can
be identified even in individuals who outwardly express egali-
tarian beliefs [20, 24]. According to Project Implicit, the Race
IAT has been taken >4 million times between 2002 and 2017,
and 75% of test takers demonstrate an automatic white pref-
erence, meaning that most people (including a small group
of black people) automatically associate white people with
goodness and black people with badness [20]. Proponents of
the IAT state that automatic preference for one group over an-
other can signal potential discriminatory behavior even when
the individuals with the automatic preference outwardly ex-
press egalitarian beliefs [20]. These preferences do not neces-
sarily mean that an individual is prejudiced, which is associated
with outward expressions of negative attitudes toward different
social groups [20].

Many of the studies of unconscious bias described in this ar-
ticle use the IAT as the primary tool for measuring the phe-
nomenon. Nevertheless, the degree to which the IAT predicts
behavior is as of yet unclear, and it is important to recognize
the limitations and criticisms of the IAT, as this is pertinent to
its potential application in mitigating unconscious bias. Blanton
et  al reanalyzed data from 2 studies supporting the validity of
the IAT, claiming that there is no evidence predicting individual
behavior, with concerns for interjudge reliability and inclusion
of outliers affecting results [25]. Response to this criticism by
McConnell et al describes extensive training of test judges and
evidence that the reanalysis was not a perfect replication of
methods [26]. Blanton et  al argue further in a different article
that attempting to explain behavior on the basis of results of
the IAT is problematic because the test relies on an arbitrary
metric, leading to identified preferences when individuals are
“behaviorally neutral” [27]. Notwithstanding the limitations of
the IAT, none of its critics refute the existence of unconscious
bias and that it can influence life experiences. The following
sections review how unconscious bias affects different groups in
the healthcare workforce.

Racial Bias

Medical school admissions committees serve as an important
gatekeeper to address the significant disparities between racial


nloaded from



62/5552356 by guest on 20 July 2021

Unconscious Bias: Challenges and Solutions • jid 2019:220 (Suppl 2) • S65

and ethnic minorities in healthcare as compared to the general
population. Yet one study demonstrated that members of a
medical school admissions committee displayed significant un-
conscious white preference (especially among men and faculty
members) despite acknowledging almost zero explicit white
preference [28]. An earlier study of unconscious racial and
social bias in medical students found unconscious white and
upper-class preference on the IAT but no obvious unconscious
preferences in students’ response to vignette-based patient
assessments [29]. Unconscious bias affects the lived experiences
of trainees, can potentially influence decisions to pursue cer-
tain specialties, and may lead to isolation. A  recent study
by Osseo-Asare et  al described African American residents’
experiences of being only “one of a few” minority physicians;
some major themes included discrimination, the presence of
daily microaggressions, and the burden of being tasked as race/
ethnic “ambassadors,” expected to speak on behalf of their dem-
ographic group [30].

Gender Bias

Gender bias in medical education and leadership develop-
ment has been well documented [11, 31]. Medical student
evaluations vary depending on the gender of the student and
even the evaluator [31]. Similar studies have demonstrated
gender bias in qualitative evaluations of residents and letters of
recommendations, with a more positive tone and use of agentic
descriptors in evaluations of male residents as compared to fe-
male residents [11]. Studies evaluating inclusion of women
as speakers have also demonstrated gender bias, with fewer
women invited to speak at grand rounds [9] and differences
in the formal introductions of female speakers as compared to
male speakers [32, 33], with men more likely referred to by their
official titles than women.

Sexual and Gender Minority Bias

Sexual and gender minority groups are underrepresented in
medicine and experience bias and microaggressions similar to
those experience by racial and ethnic minorities. Experiences
with or perceptions of bias lead to junior physicians not
disclosing their sexual identity on the personal statement
part of their residency applications for fear of application re-
jection or not disclosing that they are gay to colleagues and
supervisors for fear of rejection or poor evaluations [34]. In
one study, some physician survey respondents indicated some
level of discomfort about people who are gay, transgender,
or living with HIV being admitted to medical school. These
respondents were less likely to refer patients to physician
colleagues who were gay, transgender, or living with HIV [35].
These explicit biases were significantly reduced, compared
with those revealed in prior surveys done in 1982 and 1999;
opposition to gay medical school applicants went from 30% in
1982 to 0.4% in 2017, and discomfort with referring patients

to gay physicians went from 46% in 1982 to 2% in 2017 [35].
The 2017 survey did not measure levels of unconscious bias,
which is likely to still be pervasive despite decreased explicit
bias. As with other types of bias, these data reveal that ex-
plicit bias against gay physicians has decreased over time; the
degree of unconscious bias, however, likely persists. While
this is encouraging to some degree, unconscious bias may be
much more challenging to confront than explicit bias. Thus,
members of underrepresented groups may be left wondering
about the intentions of others and being labeled as “too

Studies including the perspectives of LGBTQ healthcare
professionals demonstrate that major challenges to their aca-
demic careers persist to this day. These include lack of LGBTQ
mentorship, poor recognition of scholarship opportunities,
and noninclusive or even hostile institutional climates [36].
Phelan et  al studied changes in biased attitudes toward sexual
and gender minorities during medical school and found
that reduced unconscious and explicit bias was associated
with more-frequent and favorable interactions with LGBTQ
students, faculty, residents, and patients [37].

Disability Bias

Physicians with disabilities constitute another minority group
that may experience bias in medicine, and the degree to which
they experience this may vary, depending on whether disabilities
may be visible or invisible. One study estimated the prevalence
of self-disclosed disability in US medical students to be 2.7% [7].
Medical schools are charged with complying with the Americans
With Disabilities Act, but only a minority of schools support the
full spectrum of accommodations for students with disabilities
[38]. Many schools do not include a specific curriculum for
disability awareness [39]. Physicians with disabilities have felt
compelled to work twice as hard as their able-bodied peers for
acceptance, struggled with stigma and microaggressions, and
encountered institutional climates where they generally felt like
they did not belong [40]. These are themes that are shared by
individuals from racial and ethnic minorities.


A strategy to counter unconscious bias requires an intentional
multidimensional approach and usually operates in tandem with
strategies to increase diversity, inclusion, and equity [41, 42].
This is becoming increasingly important in training programs
in the various specialties, including infectious diseases. The
Accreditation Council for Graduate Medical Education recently
updated their common program requirements for fellowship
programs and has stipulated that, effective July 2019, “[t]he
program’s annual evaluation must include an assessment of
the program’s efforts to recruit and retain a diverse workforce”
[43]. The implication of this requirement is that recognition


nloaded from



62/5552356 by guest on 20 July 2021

S66 • jid 2019:220 (Suppl 2) • Marcelin et al

and mitigation of potential biases that may influence retention
of a diverse workforce will ultimately be evaluated (directly or

Mitigating unconscious bias and improving inclusivity is a
long-term goal requiring constant attention and repetition and
a combination of general strategies that can have a positive in-
fluence across all groups of people affected by bias [44]. These
strategies can be implemented at organizational and individual
levels and, in some cases, can overlap between the 2 domains
(Figure 2). In this section, we review how infectious diseases
clinicians and organizations like IDSA and hospitals can use
some of these strategies to address and mitigate implicit bias in
our specialty.

Organizational Strategies
Commitment to a Culture of Inclusion: More Than Just Diversity
Training or Cultural Competency
Creating change requires more than just a climate survey,
a vision statement, or creation of a diversity committee [45].
Organizations must commit to a culture shift by building insti-
tutional capacity for change [41, 46]. This involves reaffirming
the need not only for the recruitment of a critical mass of un-
derrepresented individuals, but equally importantly, the re-
cruitment of critical actor leaders who take the role of change
agents and have the power to create equitable environments
[41, 47–49]. These change agents need not themselves be un-
derrepresented; indeed, the success of culture change requires
the involvement of allies within the majority group (eg, men,
white people, and cis-gender heterosexual individuals). IDSA
has demonstrated a commitment to this type of culture change
with recent changes in leadership structure and with intentional

recruitment of individuals invested in diversity and inclusion;
however, there is always room for reevaluation of other areas
where diversity is desired.

Committing to a culture of inclusion at the academic-
institution level involves creating a deliberate strategy for
medical trainee admission and evaluation and faculty hiring,
promotion, and retention. Capers et  al describe strategies for
achieving diversity through medical school admissions, many
of which can also be applied to faculty hiring and promotion
[49]. Notable strategies they suggest include having admissions
(or hiring) committee members take the IAT and reflect on
their own potential biases before they review applications or
interview candidates [49]. They also recommend appointing
women, minorities, and junior medical professionals (students
or junior faculty) to admissions committees, emphasizing the
importance of different perspectives and backgrounds [49].
Organizations can also survey employee perception of inclu-
sivity. These assessments include questions on the degree to
which an individual feels a sense of belonging within an in-
stitution, alongside questions pertaining to experiences of
bias on the grounds of cultural or demographic factors [50].
Conducting regular assessments and analysis of survey results,
particularly on how individuals of diverse backgrounds feel
they can exist within the organization and their culture simul-
taneously, allows organizations to ensure that their trainings
on unconscious bias and promotion of cultural humility lead
to long-term positive change. Furthermore, realizing that dif-
ferent demographic groups may feel less respected than others
provides information on areas of focus for consequent refresher
seminars on combating unconscious bias in conjunction with
cultural humility.










reflection on



to culture


Strategies to





Figure 2. Organization-level and personal-level strategies to mitigate unconscious bias. Orange circles indicate organization-specific strategies, green circles indicate
individual-level strategies, and blue circles represent strategies that can be emphasized on both organizational and individual levels to mitigate implicit bias.


nloaded from



62/5552356 by guest on 20 July 2021

Unconscious Bias: Challenges and Solutions • jid 2019:220 (Suppl 2) • S67

Meaning ful Diversity Training and the Usefulness of the IAT
Notwithstanding potential criticisms of the IAT with respect
to prediction of discriminatory behavior, this can be a useful
tool within a comprehensive organizational training seminar
directed toward understanding and addressing individual un-
conscious bias. In the study by Capers et  al, over two thirds
of admissions committee members who took the IAT and
responded to the post-IAT survey felt positive about the poten-
tial value of this tool in reducing their unconscious bias [2