Addressing Implicit or Unconscious Biases in Healthcare

Contributed by Maushumi Bhattacharjee 

Abstract: This article is a commentary on prevalence of implicit or unconscious biases in the healthcare systems in the US and Canada. The article analyses various empirical studies undertaken in North America to identify the presence of Implicit Biases in its various forms, including racism and gender insensitivity, and identifies possible areas for improvement.  Implicit Association Test (IAT) which is the most common tool used by researchers to identify implicit biases needs to be updated to recognizes several forms of biases which are beyond its scope at present. Medical practitioners must be trained to recognize and eliminate implicit or unconscious biases in healthcare as part of their clinical, ethical, and legal obligations, that is included within the ‘standard of care’ for patients.

Résumé: Cet article est un commentaire sur la prévalence des préjugés implicites ou inconscients dans les systèmes de santé aux États-Unis et au Canada. L’article analyse diverses études empiriques entreprises en Amérique du Nord pour identifier la présence de biais implicites sous ses diverses formes, y compris le racisme et l’insensibilité au genre, et identifie les domaines d’amélioration possibles. Le test d’association implicite (IAT), qui est l’outil le plus couramment utilisé par les chercheurs pour identifier les biais implicites, doit être mis à jour afin de reconnaître d’autres formes de biais, ce qui dépasse actuellement sa portée. Les médecins praticiens ont besoin d’être formés afin de reconnaître et éliminer les préjugés implicites ou inconscients en soins de santé selon  leurs obligations clinique, éthique et légale, qui sont incluses dans la “norme de soins” pour les patients.

Defining Implicit or Unconscious Biases in Healthcare and its Various Forms 

There have been several studies that document racial, ethnic or gender-based biases in healthcare across North America. These biases may often occur as blatant racism in many cases. However, they may also be more subtle, taking the form of unconscious behavior on the part of clinicians or healthcare service providers. Implicit biases result from unconscious attitudes and beliefs that are systemically ingrained in people who lack awareness and knowledge of how these biases affect their behavior towards certain people. These kinds of implicit, subtle, or unconscious biases in treatment and patients’ care are often unrecognized and therefore unaddressed. Implicit biases in healthcare could take several forms. For example, research on the use of stigmatizing language by physicians in patients’ medical records showed that most of the negative language used by physicians was implicit but had the potential to transmit bias which affected the quality of care to patients. The categories of such negative language used by physicians ranged from “questioning patient credibility”, “disapproval” to “stereotyping”. Another example could be of racial stereotyping wherein black patients being prescribed insufficient pain medication because of implicit stereotypical beliefs of clinicians that black people do not feel pain as much as white people, or common unfortunate misbeliefs such as black people have thicker skins. Underrepresented minorities, such as disabled people or people belonging to LGBTQ+ community are also frequently victims of such unconscious biases of medical practitioners while getting treatment. Racial, ethnic and gender minorities already suffer from disparity in access to healthcare, and such implicit biases adversely affects the quality of patient care that is provided to them. 

Implicit biases can manifest into repeated subtle, verbal, or non-verbal insults towards marginalized persons, which are termed as ‘microaggressions’. These are everyday intentional or unintentional insults directed towards vulnerable minorities based on their race, gender, ethnicity, culture, or religion. The fact that they are repeated, and commonplace makes them derogatory and hostile in nature. The major problem with implicit biases is that they are so often unrecognized by the person who is perpetrating them as well as those on the receiving end. Implicit biases can also be found in cases of determining capacity of patients, as in several cases the clinicians would determine that the patient does not have capacity just on the basis that the patient is unable to communicate their needs with them. However, there are several reasons that could cause patients to hesitate in communicating with their doctors. Most common reasons are different cultural backgrounds or linguistic barriers. For example, in South Asian or East Asian cultures, the patients do not make their medical decisions on their own, but with the help of their family members, which is why they may often be unable to communicate their needs. As is often the case, it is the cultural underpinning that is driving the patients’ presentation to their clinicians. Clinicians, on the other hand, are often guided by the bioethical principles of respecting patient autonomy, which leads them to ask several questions to determine patient capacity. This lack of communication from patients has got nothing to do with their capacity, but rather with where they come from, or their life experiences which formed their relationship with the clinician or the hospital. Lack of cultural competency among clinicians in United States and Canada is one of the major factors that results in unconsciously biased behavior towards patients.

There has been an eruption of outrage across the world in response to recent events highlighting racial inequality and discrimination in North America, whether it is the brutal murder of George Floyd or the uncovering of bodies of indigenous children in residential schools in Canada. On these events, Canadian Prime Minister, Justin Trudeau used the terms “unconscious biases” and “systemic racism” to address and acknowledge the problem of racism in Canada. Racism as an unconscious or implicit bias is widely prevalent in medical practice across North America. It means that clinicians may unconsciously project racially insensitive behavior towards people of color or minorities. In these cases, racism is exhibited through far more subtle and non-violent biases that vastly go unrecognized. However, when unconscious biases among people leads to support of a collective bias, it gets transformed in systemic racism. One such example of systemic racism is of Brian Sinclair, an indigenous homeless person who was “ignored to death” in the waiting room of a hospital in Manitoba by the medical staff who made a biased assumption that he was in a drunken state instead of treating him for bladder infection. The hospital staff tried to justify their actions by stating that they treated Mr. Sinclair as they would have treated everyone else, which only shows that they were blinded by their own preconceived notions and existing stereotypes against indigenous people. 

Measuring Implicit Biases and their Impact in Healthcare

Evidence-based research is considered as the most appropriate means for formulation of better health policies that also ultimately determine the standard of patient care. An important example of positive outcomes of evidence-based medical practice and policy is the use of harm-reduction methods for dealing with patients of substance-abuse. In the case of measuring evidence of implicit or unconscious biases in healthcare delivery, surveys and questionnaires won’t be ideal because most people are unaware of these biases even though they might be unconscious perpetrators or victims of such biases in healthcare. So, to measure unconscious bias two professors of clinical and social psychology developed a scientific research tool that measures peoples’ hidden biases and prejudices. It is called the Implicit Association Test (IAT)and has been widely used in fields of medicine and law. The Race IAT has been the most popular one and extensively used between 2002 and 2017 in the United States. The results of Race IAT studies have shown prevalence of automatic white preference, that is “most people associating white people with goodness and black people with badness”. IAT is said to measure biases and implicit associations in people even when they exhibit and portray themselves as unbiased people favoring racial equality. 

Despite IAT being a unique research tool and the first of its kind to measure unconscious biases, it has several limitations which can make its results unreliable. While there is no doubt about the existence of implicit biases in healthcare, especially racial and ethnic biases, a lot more systematic research and data collection must be done to have definitive results on the impact of such biases in healthcare. At present, there is a lack of uniform research methods for measuring implicit biases in healthcare. Results from some IAT studies show that implicit bias more frequently affects the relationship between patient and clinician or health service provider. However, these studies do not consider several relevant factors such as background of clinicians or provider, or intersectionality within participants. 

Since most research in implicit bias has focused on race, there hasn’t been enough investigation on biases based on other characteristic identities, such as, gender, sexual orientation, disability, religion, and culture. There hasn’t been any IAT studies conducted in countries outside of global north, where there may exist other forms of implicit biases within the medical community. For example, the prevalence of caste discrimination in India, growing hostilities against certain religious minorities, or lack of representation from LGBTQ community could be translated into explicit or implicit biases in behavior of clinicians towards their patients. Casteism is also prevalent in North America among South Asian diaspora communities, and could turn up as biased behavior from clinicians belonging to such communities. Research on implicit biases is crucial for improving patient welfare and reducing systemic racism and other biases within healthcare community. In order to obtain comprehensive and accurate data on implicit biases, the research methodologies need to be improved and current methodological limitations must be addressed. Some recommendations include asking more nuanced questions to participants, more cross-sectional studies for estimating national estimates on implicit bases, more representation within the sample, and consideration of factors other than just race. 

Recognizing and Eliminating Implicit Biases as Part of Standard of Care

The medical community states that the standard of care which a patient must get depends on evidence-based medical and health policy research, The standard of patient care is determined after considering the medical knowledge of clinicians and nurses. In a bioethical sense from the perspective of a clinical ethicist, the standard of care must be patient centered care, which entails that patient must determine the standard for their care. Aside from respecting patient autonomy, minimizing harm to patients and ensuring justice are important bioethical principles that inform the ethical standards of patient care that all clinicians must judiciously follow. It is therefore important to recognize implicit and unconscious biases in clinicians’ behavior which may lead to harm and result in racial injustices. While IAT studies, medical research and surveys continue to collect evidence, showcasing the prevalence of implicit biases and its impact on patients’ care, it is crucial to recognize such biases and eliminate them as an ethical obligation on part of clinicians or health service providers. 

The legal definition of standard of care has been developed historically through several cases in common law and civil law jurisdictions. In US and Canada, the standard of care warrants a reasonable degree of care towards patients on part of the medical practitioner. The reasonability criterion has been explained in a Canadian judicial decision on medical liability and standard of reasonable care:

 “Every medical practitioner must bring to his task a reasonable degree of skill and knowledge and must exercise a reasonable degree of care. He is bound to exercise that degree of care and skill which could reasonably be expected of a normal, prudent practitioner of the same experience and standing, and if he holds himself out as a specialist, a higher degree of skill is required of him than of one who does not profess to be so qualified by special training and ability.”

A deviation or departure from such standard of care by a medical practitioner towards patient would be termed as medical malpractice. Another legal case in US from 1969 involved a patient who had progressed from an operation but took a fall, after which his condition deteriorated. The patient was left unattended when he fell. The court explained that the question whether it was the hospital’s neglect that caused the patient’s condition to worsen was irrelevant. The court held that it was “hospital’s duty to exercise reasonable care for the safety and well-being of the patient”, which was the standard of care that the patient should have received. The neglect not only caused harm to the patient but to the relationship and trust between the patient and clinician. Similarly, patients lose trust in their clinicians because of microaggressions and implicit biases, which are woven into the way care is being provided, and so result in a departure from the standard of care.

In Schoendorff v Society of New York Hospital, the clinicians did not follow explicit instructions of the patient which was to only look at what was in the uterus without performing any surgery before discussing with her. But when they found fibroids in her uterus, they decided to take them out. While gender insensitivity and bias were in play here, the case was about recognizing patients’ basic right to consent to treatment before they are subjected to it. The court has recognized respect for autonomy of patients in their care as a standard of care that is legally mandated upon medical practitioners. The physician has a duty of reasonable care to warn of dangers lurking in proposed treatment and to impart information which patient has every right to expect. A reasonable explanation about the treatment is a right of every patient. It means generally informing the patient in non-technical terms as to what is at stake, i.e., the therapy alternatives and risks which may ensue from treatment or no treatment. The court said that this standard of what ought to be disclosed and shared as conversation between the patient and the clinicians is not measured by what the clinicians think is reasonable, but rather, is measured by what the patient thinks is reasonable. Respect for patient’s right of self-determination in therapy demands an objective legal standard for physicians rather than one which they may or may not choose to impose upon themselves. Unlike the standard of what a physician is technically capable of accomplishing, this aspect of their duty, showing respect for the patient’s rights, should be determined by a community standard and not by standards created by clinicians. 

The medical staff at the Manitoba hospital where Brian Sinclair, an indigenous homeless man who was ignored to death, claimed that they had given the same standard of care to Brian as they would have to non-indigenous patients. However, this was far from the truth. Brian Sinclair did not die due to long waiting periods, but due to systemic racism as he was denied a simple procedure and some antibiotics to treat his bladder infection. This can easily fall within the legal definition of standard reasonable care expected from health service providers. Brian Sinclair was a victim of implicit biases that resulted in his delayed treatment and thereafter in his death. Recognizing and eliminating such implicit biases must be considered a legal obligation on part of the clinicians and as part of the standard of care provided to patients. 

Healthcare providers, clinicians, ethicists, and lawyers need to take more affirmative responsibility to identify instances of misconduct, bias and microaggressions. These instances must be studied and identified. Proper training needs to be given to healthcare providers to eliminate such biases in patient care and that should be the standard of care that patients deserve. 

Recommendations and Suggestions for Mitigation of Implicit Bias

Much research and deliberation need to be undertaken among medical community, lawyers, ethicists, patient communities, and underrepresented minority communities on an ideal strategy to police substandard patient care related to implicit bias. Medical Review Boards, Legal System and Patient Advocacy Organizations are some of the relevant institutional bodies that can play a role in suppressing implicit biases in patient care. Training sessions must be organized, not just for the clinicians but also for patients so that they can recognize and identify certain biased behavior that they may experience during their treatment. Patients must be made aware of the standard of care they deserve and the steps they can take if such care is not provided to them. Medical institutions, hospitals and universities must undertake community based participatory research methods and other methods such as public deliberation with patient communities to get insights from patients on mitigating such implicit and unconscious biases. For clinicians, organizational strategies such as cultural competency training and training in cultural humility can help clinicians recognize and eliminate unconscious biases in patients’ care. Furthermore, clinicians may also undertake individual or group deliberations to share their understanding and experiences of unconscious biases and reflect on ways to tackle them. Training for medical students must also include exposure to staged interactions with patients that involve biased and stereotypical dialogues, which could help them recognize and avoid such behavior in real life. 

Training and research can help the medical community realize their ethical and clinical obligations to recognize and eliminate implicit biases in patient care as part of the standard of care to be given to patients. The popular attitude towards ethics in medical community, especially among medical students, is to leave ethical dealings with the ethicist and to only care about legal obligations towards patients, since violating those obligations will have repercussions. While racism and ethnic biases are explicitly forbidden, there are many other kinds of biases which may not fall under the legal purview. In fact, systemic racism is often aided by colonial laws and health policies. Although, as stated earlier, the legal definition of standard of care is broad enough to include elimination of implicit biases in patient care as part of standard care. Clinicians and medical students must be actively given knowledge of this fact so that they feel legally obligated to eliminate unconscious biases in patient care.