Preparing Medical Students to Address the Needs of Vulnerable Patient Populations: Implicit Bias Training in US Medical Schools

Abstract

Little is known about how medical students are trained to identify and reduce their own biases toward vulnerable patient groups. A survey was conducted among US medical schools to determine whether their curricula addressed physician implicit biases toward three vulnerable patient groups: lesbian, gay, bisexual, transgender, and questioning (LGBTQ) individuals, persons experiencing homelessness, and migrant farmworkers. Of 141 US medical schools, 71 (50%) responded. Survey respondents indicated that implicit bias is not routinely addressed in medical education, and training specific to vulnerable populations is infrequent. Recommendations for incorporating implicit bias training in medical school curricula are discussed.

Background

Patient populations are considered vulnerable if social barriers, such as discrimination, prevent them from achieving optimal functioning in social, educational, and occupational domains of their lives and prevent them from accessing services to fulfill basic needs, such as health care and housing [1]. Vulnerable populations, including those who are migrant farmworkers, experiencing homelessness, or identify as lesbian, gay, bisexual, transgender, or questioning (LGBTQ), frequently experience stigma from the general population based on group characteristics. Health care providers and students enrolled in health professions training programs often hold the same biases toward vulnerable populations as other members of society. Moreover, research shows a decline in student empathy toward patients during medical school [2]. Continuing emphasis on reducing biases toward vulnerable populations throughout medical education is likely critical for minimizing the negative impact of these biases on medical decision-making and patient interactions [3].

Many health professionals express discomfort providing care for vulnerable populations [4] and may exhibit biases toward vulnerable patient groups even in the absence of negative attitudes that are consciously-accessible and controlled (i.e., explicit bias) [4]. These attitudes are typically unconscious (i.e., outside of conscious awareness), automatic, and referred to as implicit bias [5, 6]. Implicit biases held by health care providers toward vulnerable populations can result in poor physician-patient communication, lack of trust, lower standard of care, restricted access to services, lower adherence to treatment, poorer personal health outcomes, and health disparities within the population [7–10].

National surveys indicate that medical schools devote little time to implicit biases that could negatively impact the health care needs of vulnerable populations [11, 12]. Research suggests that small group discussions are effective in increasing student and health care provider comfort levels working with vulnerable populations [13]. These types of training experiences provide opportunities for medical students to rehearse perspective-taking skills and to practice focusing on individual attributes rather than group membership. However, few US medical schools provide opportunities to practice bias awareness and reduction strategies such as the use of simulated patients or direct engagement with vulnerable patients through intergroup contact or hypothetical encounters. Despite evidence that health care provider implicit biases negatively impact vulnerable populations [7–10], there are many gaps in the research on whether training medical students to address implicit biases can improve physician-patient communication, physician behavior, or patient outcomes.

The implicit biases experienced by vulnerable populations in health care settings are well documented. LGBTQ individuals [14–25], persons experiencing homelessness [26–41], and migrant farmworkers [42–48] are at an elevated risk for negative health outcomes and face significant health disparities compared with the general population [14]. Therefore, focusing on these three vulnerable patient groups in medical education can help to overcome barriers to addressing implicit bias in several ways. First, the diversity in gender, race, class, and culture reflected within these groups will allow educators to maximize the scope of implicit bias training by focusing on intersecting vulnerabilities. This strategy strikes a balance between addressing biases encountered by all vulnerable groups, which would not be feasible to the given time constraints, and focusing on only one type of bias. Second, these three groups experience staggering health disparities across a wide variety of physical and mental health conditions. Hence, this strategy can help to overcome barriers related to the perceived relevance of implicit bias training in medical education. Third, these vulnerable patient groups that capture the rapidly changing sociocultural, economic, and political landscape of medicine in the US medical schools need to anticipate these shifts in patient demographics to effectively prepare the next generation of providers.

This article will focus on how medical schools are training students to identify and address implicit biases toward persons who identify as LGBTQ, persons experiencing homelessness, and migrant farmworkers. The first goal of this article is to illustrate the potential benefit for medical school students to focus on these three patient groups during their implicit bias reduction training. The second goal is to describe the extent to which medical students in the USA receive implicit bias reduction training, how training is structured, and if training focuses on the unique needs of these vulnerable population groups. Results of a survey sent to 141 allopathic medical schools in the USA will be discussed.

Activity

To better understand how medical schools in the USA are preparing students to address the needs of these vulnerable patient groups, online survey invitations were sent to one individual (i.e., associate dean, course director, or program coordinator) at each US accredited medical school who was identified as being knowledgeable about their curriculum. Of the 141 allopathic US medical schools sent with survey links, 71 (50%) responded. The survey was distributed using Research Electronic Data Capture (REDCap), a secure web-based platform. If respondents indicated that medical students received training on implicit bias relevant to a vulnerable patient group, then follow-up questions are given focusing on where in the curriculum it occurred (e.g., patient care, practice-based learning, and improvement), training methods used (e.g., lectures, simulation/standardized patients), names of training courses, and number of hours devoted to training in each year of medical school. All survey questions are listed in Table 1. The research protocol was approved by the institutional review board at Meharry Medical College.

Table 1.

Descriptive statistics for medical education survey responses on implicit bias training for vulnerable populations

	Vulnerable population
	LGBTQ patients, N (%)	Patients experiencing homelessness, N (%)	Migrant farmworker patients, N (%)
Received training on implicit bias?
Yes	28 (39)	8 (11)	10 (14)
No	43 (61)	63 (89)	61 (86)
Domain of implicit bias training?
Patient care	21 (75)	5 (50)	1 (8)
Knowledge for practice	14 (50)	5 (50)	4 (33)
Practice-based learning and improvement	9 (32)	2 (20)	1 (8)
Interpersonal and communication skills	14 (50)	5 (50)	2 (17)
Professionalism	14 (50)	5 (50)	0 (0)
Systems-based practice	8 (29)	2 (20)	1 (8)
Interpersonal collaboration	9 (32)	2 (20)	1 (8)
Personal and professional development	6 (21)	2 (20)	0 (0)
Other	0 (0)	0 (0)	0 (0)
How is implicit bias training implemented?
Lectures	20 (71)	1 (10)	2 (17)
Conferences or workshops	9 (32)	0 (0)	0 (0)
Case- or problem-based learning	14 (50)	1 (10)	3 (25)
Small group discussion	13 (46)	4 (40)	2 (17)
Simulation/standardized patients	7 (25)	1 (10)	0 (0)
Patient care experiences	9 (32)	2 (20)	1 (8)
Other	4 (14)	0 (0)	0 (0)
When is implicit bias training implemented?
Year 1	13 (46)	3 (30)	3 (25)
Year 2	13 (46)	2 (20)	0 (0)
Year 3	14 (50)	4 (40)	4 (33)
Year 4	7 (25)	3 (30)	2 (17)

Frequencies are based on total number of responses

Results

Study findings indicate that fewer than half of US medical school respondents implemented implicit bias reduction training toward vulnerable populations. Of the 71 medical schools that responded, only 36 (51% of respondents and less than 25% of the medical schools indicated that their students received training on implicit bias (see Table 1). Implicit bias training occurred most frequently with LGBTQ patients (39% of respondents) followed by migrant farmworker patients (14%) and patients experiencing homelessness (11%).

How and when is implicit bias training implemented? Approximately half of medical schools that responded reported that students received training addressing implicit biases toward LGBTQ patients in the first 3 years and 25% of schools indicated that training continued across all 4 years of medical school. Implicit bias training was most commonly integrated into the medical education core competency domains of “knowledge for practice” (e.g., 33% of respondents that focused on migrant farmworker patients), “patient care” (e.g., 75% of respondents that focused on LGBTQ patients), and “interpersonal/communication skills” (e.g., 50% of respondents that focused on patients experiencing homelessness). The most common formats for training were lectures (71%), case- or problem-based learning (50%), and small group discussions (46%).

Implicit bias training focusing on patients experiencing homelessness was implemented across all 4 years in 20 to 40% of schools. Of the smaller number of schools that focused on implicit biases toward migrant farmworker patients, training was included in all years except for year 2. Medical students were most likely to receive education on implicit biases toward vulnerable patient populations in ethics courses (e.g., “Professionalism and Ethics,” “Medical Ethics”), introductory courses (e.g., “Introduction to Clinical Skills”), and in courses on patient care (e.g., “Patient Discrimination,” “Advanced Clinical Skills”). Importantly, training was often integrated across multiple courses and years.

Discussion

Incorporating bias reduction training into medical school curricula represents an important strategy toward minimizing the negative impact of provider biases on the health of vulnerable populations. Failure to address implicit bias training in medical education exacerbates health disparities experienced by vulnerable populations through at least two pathways. First, medical students who do not receive education about implicit biases and preparation for vulnerable populations are unlikely to be motivated to monitor and change the way they interact with—and make medical decisions for—these vulnerable patient groups. Second, failure to emphasize the importance of implicit bias reduction may decrease the likelihood that medical students and providers disclose their own sexual orientation and gender identity or past experiences of homelessness, immigration, or farm work to colleagues, thereby limiting opportunities for intergroup contact and maintaining the “hidden curriculum” that perpetuates implicit biases within institutions [49].

Prior research highlights, the importance of introducing implicit bias awareness activities in supportive and individualized learning environments, building motivation to manage and change implicit biases through education about their impact on health disparities, and teaching strategies to reduce and manage implicit biases through contact with vulnerable patient groups and perspective-taking [50]. These bias reduction techniques are likely to be effective for medical students learning to work with patients from a variety of vulnerable population groups who present with often intersecting vulnerabilities.

Limitations of this study provide directions for future research. Information regarding medical school training was obtained from surveys completed by volunteers and could therefore be influenced by nonresponse bias. Future studies should confirm the present findings through systematic searches of medical school training curricula, eliciting feedback from multiple respondents for each school, offering incentives for survey completion, and/or administering follow-up surveys.

Funding Information

This project is supported by the Health Resources and Services Administration (HRSA) of the US Department of Health and Human Services (HHS) under grant number UH1HP30348, entitled “Academic Units for Primary Care Training and Enhancement.” This information or content and conclusions are those of the authors and should not be construed as the official position or policy of nor should any endorsements be inferred by HRSA, HHS, or the US Government.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflicts of interest.

Ethical Approval

All procedures involving human participants were in accordance with the ethical standards of the Meharry Medical College Institutional Review Board and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.