Off Script: Implicit Bias Can Be a Matter of Life and Death—Let’s Do Something About It

Fair-minded people can discriminate.

Implicit bias describes the phenomenon that occurs when we unconsciously make negative or positive associations about certain groups of people. These associations occur outside of our conscious control or awareness, but can influence our behavior. When implicit bias happens in the education system, the criminal justice system, or in employment, it can lead to certain demographic groups of people being treated unfairly: lower pay for the same work, harsher punishments for the same infraction, or restricted opportunities for higher education. Now imagine what happens when a physician is unconsciously biased against the patient in front of them. A therapeutic decision influenced by implicit bias can have dire, even life-or-death consequences.

At every step in the continuum of care for the acute coronary patient, research has long shown that underrepresented minority patients (largely African-Americans and Hispanics) and women are often delivered a lower quality of evidence-based care. African-Americans presenting to the emergency department with chest pain are less likely than whites to be evaluated thoroughly and admitted to the hospital; once admitted, they are less likely to be evaluated with cardiac catheterization; if catheterization is done and critical CAD is demonstrated, they are less likely to be treated with state-of-the-art therapies (CABG and PCI with drug-eluting stents); and they are less likely to be discharged with instructions to follow up with a formal cardiac rehab program and with the appropriate, guideline-directed medications. When they are found to have refractory, severe left ventricular dysfunction and congestive heart failure, their inpatient, ICU-level care is less likely to be overseen by a cardiologist, and they are less likely to be referred for implantation of an automatic implantable cardioverter-defibrillators (ICD) or to have existing devices upgraded to biventricular pacemakers. Perhaps the most shocking example of these racial and gender cardiovascular healthcare disparities involves patients with critical lower-extremity ischemia. Multiple studies have shown that African-Americans are more likely than whites to be treated with limb amputations and less likely to be treated with mechanical procedures to restore lower-extremity circulation. In several studies these findings persist after controlling for severity of disease at presentation. Finally, African-Americans, Hispanics, and to a lesser extent women are embarrassingly underrepresented in research trials to evaluate new treatments, like transcatheter aortic valve repair and drug therapies like sacubitril/valsartan.

Could explicit or conscious biases like racism and sexism drive these disparities? Yes, but that would mean that a physician encounters a patient and, because of the patient’s race, gender, religion, sexual identity, etc, consciously withholds important therapies. The optimist in me wants to believe that this is extremely rare. There are many culprits underlying disparities in care; however, research makes a solid case for implicit or unconscious bias influencing physicians’ therapeutic decisions.

Studies dating back more than a decade show that doctors with implicit negative racial bias, such as “implicit white race preference” on the Implicit Association Test, are less likely to recommend thrombolysis for black versus white acute MI patients with the same simulated presentation, tend to spend less time with their black patients in office visits, and tend to overtalk them. This is consistent with social psychology research that details how we interact with persons from a group against which we hold negative implicit biases: we tend to direct the conversation, allow them less time to speak, provide less eye contact, and smile less. Successfully obtaining patient consent for the procedures we perform requires gaining the patient’s understanding and trust, so if this is how we interact with them, it is not hard to imagine patients being wary of consenting to a cardiac catheterization, an ICD implantation, or an atherectomy procedure of the lower extremity! But patients declining procedures due to uncomfortable interactions is but one result of physician implicit bias. The other, more troubling outcome is when the doctor, due to an implicit negative association with the patient’s demographic group, decides against evidence-based therapies. To give a haunting example, black and Hispanic children who have pain associated with cancer are less likely than white children to be treated with narcotics for the same level of pain. And how might this play out when consenting patients to participate in research studies? Ask yourself, would you consent to have an investigational device placed inside your chest if the discussant charged with obtaining your consent does more talking than listening, doesn’t look you in the eye, or smile? Uhhh . . . no thanks, I’ll pass.

It's worth emphasizing that even well-meaning doctors with egalitarian beliefs can unknowingly discriminate—we are all vulnerable to the insidious effects of implicit bias. Fortunately, it is possible to override implicit biases with a few mental exercises. The following are a short list of examples of research-proven strategies to neutralize or mitigate implicit biases.

• Common identity formation Focus on a shared, common identity between YOU and the patient; do you have common hometowns or common interests in food, music, sports teams, etc? Such discussions not only put the patient at ease, research shows that they also blunt the impact of the physician’s negative implicit bias because now you and the patient have a shared common group identity. They are no longer “other” but “one of the gang.”
• Perspective-taking Take the perspective of the patient; what did they go through today before this interaction? What was their life like 6 months ago? Five years ago? What will happen in their household when they go back home? This exercise develops empathy for the patient that can oppose implicit bias.
• “Consider the opposite” When data seem to point to one conclusion, briefly look for data supporting the opposite conclusion before making a final decision. Example: at first pass the patient does not seem to be a good candidate for organ transplantation because of a history of medication nonadherence and lack of reliable transportation. Before making a decision, re-review the information looking for support for the opposite conclusion. Has the patient held the same job for years? This implies that they can follow instructions and complete tasks. Does the patient have a stable family life, ie, same spouse or significant other for years, children raised to independence? This implies reliability and that the patient keeps commitments. After this exercise, make a final decision. Research shows that this exercise can blunt the impact of implicit bias.

Many cardiologists will say that they already do these things. However, the current reality of racial and gender healthcare disparities in our field suggests that we could recommit to doing these mental exercises in all of our patient interactions. As I write this, I am being paged to the cath lab to meet my next patient and lead the “time-out” and patient consent process. I pledge to consciously use two or three of these strategies now, and in all patient encounters until it becomes my unconscious standard. Won’t you join me?

Off Script is a first-person blog written by leading voices in the field of cardiology. It does not reflect the editorial position of TCTMD.