How implicit bias in medicine hurts women and minorities

by Carolyn Thomas  ♥  @HeartSisters

It’s discouraging. I’ve read (and written) far too much about how the gender gap in cardiology has resulted in women heart patients being at higher risk of being both under-diagnosed compared to our male counterparts, and then under-treated even when we’re appropriately diagnosed (here,   here   and  here, for example). Studies even suggest that when physicians review case studies in which patients present with significant cardiac symptoms as well as a recent emotionally upsetting event (identical except for the patients’ male or female names), the doctors are significantly more likely to determine that a man’s symptoms are heart-related, but a woman’s symptoms are just due to the emotional upset.(1)

But what’s been missing in this acknowledged gender gap seems to be the most important part: why is this happening, and what can we do to actually address it?

It turns out that we may need to expand our attention beyond an existing gender gap that affects how women vs. men are treated, to how many other societal groups are treated, too.

Dr. Elizabeth Chapman is asking these questions. She and her colleagues at the University of Wisconsin-Madison are interested in how physicians demonstrate what’s called implicit bias in medicine. This is the trickiest kind of bias, apparently, because – unlike explicit bias in which we know and accept the fact that we’re biased against certain kinds of individuals or groups – most of us aren’t even aware of our own implicit bias.

Dr. Chapman reminds us that we all have some degree of implicit bias in a study published in the Journal of General Internal Medicine:

“All of society is susceptible to these unintended biases in decision-making, so-called implicit bias – including physicians.”

Cardiologist Dr. Sharonne Hayes is the founder of the Mayo Women’s Heart Clinic, and Director of Diversity and Inclusion at Mayo Clinic. When I asked her about employee surveys on implicit bias that I’d heard had been done at Mayo, she began by explaining that implicit bias is not all bad:

“Implicit bias actually serves us well in some circumstances. If we’re in the path of an oncoming vehicle or a snarling dog, for example, our built-in implicit bias about the dangers of speeding cars or vicious dogs can save us.

“But I now tell my medical colleagues that if you’re unaware of implicit bias and can’t control it, it can result in impaired decision-making.”

She then told me about a landmark study published in the Archives of Internal Medicine that exemplified the concept.(2)

In this study, groups of physicians viewed videotapes of the same actress performing the role of a female patient with chest pain in a scripted physician-patient interview, but in two distinct styles: one group saw a “histrionic” or excitable patient explaining her symptoms, the other saw a professional, business-like patient. A third group read a word-for-word transcript of the patient interview. Despite their conclusions that the cardiac risk for all examples was similar, 93% of physicians recommended cardiac referral when a patient was portrayed as business-like, compared to just 53% for the histrionic portrayal.

That conclusion isn’t about men and women. It’s about women and women.

But Dr. Hayes added that the goal in helping health care professionals to identify this type of implicit bias is not to make them feel defensive, blamed or shamed. The goal is not even to get rid of implicit bias (which may never be entirely possible anyway). We know that physicians want to offer the best possible care to patients. But they also don’t want to engage in impaired decision-making.

Like our doctors, most of us may believe that we’re smart enough/enlightened enough/ aware enough to outrun implicit biases. But, apparently, the mere existence of cultural stereotypes out there about various groups (e.g. women, men, blacks, whites, immigrants, the elderly, etc.) can subtly influence behaviour toward and judgment of individuals from that stereotyped group.(2)

For example, we know that black patients are systematically under-treated for pain, are less likely than white patients to receive pain medication – and, if they do, they receive less.(3) This is not so much because pain is recognized but untreated, by the way, but more that pain is not even recognized in the first place.

For example, Dr. Chapman described the deep cultural roots of implicit bias in a 2016 interview with the journal, ACP Hospitalist:

“We know that if we test a person’s bias towards people of colour or any other kind of minority group, people tend to favor white people. They tend to favor the majority group in general, and that holds true for physicians also.

“Even if you are of the minority category yourself, you may have that preference. It’s not as strong, but even kids as young as three years old have a tendency to prefer white faces to black faces, regardless of their colour themselves.”

As opposed to explicit prejudices (e.g. those we freely admit to), implicit bias occurs without conscious awareness, and can even seem at odds with our own personal beliefs.(4)  So if you’re already thinking, “Who, me? Biased?” – this might be about you.

How can physicians and other health care professionals overcome this kind of pervasive implicit bias – particularly since we know it’s affecting women’s heart disease diagnosis and care? And how easy will it be to convince physicians to examine their own implicit bias?

The first step, says Dr. Chapman, is for physicians and their colleagues to realize they likely do have bias in the first place:

“I think it’s often hard for physicians to admit it because, even though we know this implicit bias is part of being human, physicians may find it hard to admit there’s a problem. But we can break any habit. So, with practice, we think we can break the implicit bias habit.”

The University of Wisconsin-Madison developed “Breaking the Bias Habit,” a 3-hour interactive workshop for internists, family practice physicians, and internal medicine residents.(5) Their four strategies to reduce the risks of implicit bias are summarized in the acronym EPIC:

• Engaging in perspective-taking: a conscious attempt to envision another person’s viewpoint, and thinking of that person as a partner in care, using words like “we” and “our” when speaking with patients
• Practicing the right message: instead of just saying that everyone has implicit bias, say that everyone has these biases and most people are working to change them
• Individuating: look at patients as individuals instead of relying on stereotypes from the social category they belong to; instead of saying that a 66-year-old black female presents with chest pain, try coupling that capsule sentence with more information, such as her smoking status or family history, to round out that patient’s story
• Challenging stereotypes: challenge stereotypes and broad generalizations about groups of people with facts, such as how heart disease is the #1 killer of both men and women. A reminder: “Research shows that you’re more likely to fall back on stereotypes and bias habits when you’re tired, stressed, or hungry.”
  

Mayo Clinic has already started working with its own employees in this area. In 2015, Dr. Hayes and her team surveyed Mayo employees with questions about diversity inclusion and exclusion, and assessed steps being taken to prevent unintended bias. Staffers took an Implicit-Association Test to gauge their own implicit bias. Mayo is using these results as a baseline for future studies, with the aim of building on strengths, addressing gaps, and ensuring an equitable and welcoming environment – for both the Mayo workforce and patients alike. Dr. Hayes added:

“There is no cookbook recipe or gold standard for achieving this. We need to implement policies that support doctors, nurses and all healthcare providers to make good decisions for our patients.”

Curious about your own level of implicit bias? Project Implicit hosts a free version of the Implicit-Association Test used at Mayo Clinic. You may be surprised (as I certainly was!) to uncover your own implicit bias around topics like male/female roles, religions, disabilities, age, weight, skin tone, and many others.  Take this free test online at implicit.harvard.edu.

♥

1. Cardiovascular Research Foundation (2008, October 12). Signs Of Heart Disease Are Attributed To Stress More Frequently In Women Than Men.

2. Birdwell BG. “Evaluating chest pain. The patient’s presentation style alters the physician’s diagnostic approach.” Arch Intern Med. 1993 Sep 13;153(17):1991-5.

3. Green CR et al. “The unequal burden of pain: confronting racial and ethnic disparities in pain.” Pain Med. 2003 Sep; 4(3):277-94.

4. Nelson A. “Unequal treatment: confronting racial and ethnic disparities in health care.” J Natl Med Assoc. 2002;94(8):666–8.

5. Carnes M et al., “Effect of an Intervention to Break the Gender Bias Habit for Faculty at One Institution: A Cluster Randomized, Controlled Trial.” Academic Medicine : Journal of the Association of American Medical Colleges. 2015;90(2):221-230.

    

♥

Q: Have you identified an implicit bias that you didn’t know you had?

NOTE FROM CAROLYN:  I wrote more about how implicit bias can affect doctor-patient communication in my book, “A Woman’s Guide to Living with Heart Disease“ is available in bookstores (please support your local independent bookseller!) or order it online (paperback, hardcover or e-book) at Amazon – or order it directly from my publisher Johns Hopkins University Press (use their code HTWN to save 30% off the list price).

See also:

Yentl Syndrome: cardiology’s gender gap is alive and well

Cardiac gender bias: we need less TALK and more WALK

14 reasons to be glad you’re a man when you’re having a heart attack

Are women being left behind in cardiac research?

Heart Disease: not just a man’s disease anymore

How doctors discovered that women have heart disease, too

How a woman’s heart attack is different from a man’s