Unless you are a cinder block or some other inanimate object, you have implicit bias. Just like body odor or flatulence, it is part of being human and can be a problem if you don’t recognize it. A study just published in JAMA Network Open should be further evidence of the implicit gender bias that health care professionals have. The question then is not whether implicit bias exists, but instead what to do about it.
Bias is when you prefer or avoid, favor or hinder, like or dislike something, some idea, someone, or some group of people without real justification. Therefore, you can’t really say that you are biased against sleeping in a bed of manure because there are clear reasons why that would be a bad idea. By contrast, there is usually no justifiable basis for bias for or against a particular sex, race, ethnicity, or other socio-demographic group. After all, every such group is very diverse, and talent, ability, morals, and personality have no real association with such demographics.
Your biases can fall into two general categories based on your awareness of them. Explicit bias is when you exhibit prejudice knowingly. For example, you may not want to associate with Asians or Blacks or want to associate with women just in certain ways. While this type of bias is certainly not good and prevents organizations and society from reaching their potential, at least it is clearer or more obvious.
More insidious and thus more difficult to deal with is implicit bias. This is bias that you have unconsciously, without you even realizing it. For example, you may claim or even believe that you are “open-minded” but a closer look shows that your behaviors (e.g., the composition of your friends or employees) don’t match your words. If you ever claim that you are completely unbiased, then you are either extremely self-unaware, lying, or a cinder block. If it is the first possibility, then you really need to take an Implicit Association Test (IAT), a validated measure of implicit bias that measures how quickly you sort and thus associate different words or ideas. If it is the last possibility, then you may need to see a doctor.
The trouble is many professions, organizations, institutions, leaders, and other people still deny that they exhibit different types of bias. Explicit bias, of course, still and will probably always exist. That’s why there are still very exclusive social clubs. What makes things even more complex is the pervasiveness of implicit bias without enough effort to address it. That’s the problem with being unaware of something. Medicine and health care are certainly no exceptions.
This prompted a team from Washington University in St. Louis (Arghavan Salles, MD, PhD, Michael Awad, MD, PhD, Laurel Goldin, MA, Kelsey Krus, BS, Jin Vivian Lee, BA, Maria T. Schwabe, MPHS, and Calvin K. Lai, PhD) to conduct the two-part study published in JAMA Network Open, estimating how much implicit gender bias was present among health care professionals and specifically the surgeons.
The first part was an analysis of data on 953, 878 people who had completed a Gender-Career IAT as part of Project Implicit over a 12-year-period from January 1, 2006, through December 31, 2017. Of these, 42, 991 self-identified themselves as health professionals. Project Implicit had measured both the explicit and implicit gender-career bias of participants.
To measure explicit bias, the project asked participants directly “How strongly do you associate career with males and females” and “How strongly do you associate family with males and females”? For each question, each participant then could respond on a seven-point scale, with one being “strongly female” and seven being “strongly male.” The difference between the answers to the two questions then measured the level of explicit gender bias. For example, if you answered a one for the family question and a seven for the career question, the difference would be the maximum possible score, a six. If this is your score, congratulations, then, you’ve got a bleep-load of gender bias and you know it.
To measure implicit bias, the project had participants complete a Gender-Career IAT. This involved sorting words such as “male” and “female” with the words such as “career” and “family” to different sides of a computer screen. The speed at which someone paired one word with another measured how strongly that person associated the two. For example, if you quickly paired “female” with “family” rather than “career,” you strongly associate females with family rather than careers. If this is the case, then congratulations, you’ve got a bleep-load of gender bias and you don’t even bleeping know it.
Results from this first part of the study confirmed that both explicit and implicit gender biases existed among health care professionals. Take a wild guess as to which direction these biases faced. Yes, even though there are plenty examples of women with careers and men taking care of families, health care professionals tended to both explicitly and implicitly associate men with career and women with family. In fact, they found such associations to be slightly stronger among health care professionals than non–health care professionals. Of note, while female health care professionals were less likely than male health care professionals to manifest such explicit associations, they tended to more strongly exhibit such implicit associations. Thus, many women may have biases against themselves baked into them without realizing it.
For the second part, the research team recruited 131 surgeons (64.9% of whom were men) at the American College of Surgeons meeting in October 2017 in San Diego, California, to complete a special surgery-oriented tests of explicit and implicit gender bias. The explicit portion of this part asked each participant questions such as “How strongly do you associate surgery with males and females” and “How strongly do you associate family medicine with males and females”? The IAT portion had participants sort words such as “family medicine” and “surgery.”
The second part of the study revealed that both men and women tended to implicitly associate men with surgery and women with family medicine to equivalent degrees. There was a difference between men and women when it came to explicit bias, though, with male surgeons demonstrating higher levels. Surprised?
Dr. Salles, the lead author for the study and an Assistant Professor of Surgery at Washington University, explained, “I would love to stop getting asked if there is gender bias in medicine. If we could start every conversation about gender equity with the baseline understanding that there is gender bias, then we could have more productive conversations.” Yes, please stop asking Dr. Salles and anyone else this question. Can we once and for all put to rest any debate that biases such as gender, racial, and ethnic biases exist extensively in medicine? Denying that they exist is like denying that you have diarrhea. It stinks, makes things messy, and prevents real problem solving.
As every health care professional should know, the first step to solving a problem is realizing and admitting that there is a problem. Salles added, “Unfortunately, there are still many people who point to, for example, the large number of women medical students as proof that gender bias is not a problem. Those people are missing the big picture.” The big picture is just because you are able to start a journey doesn’t mean that you won’t face more obstacles than others. Or that you will be even allowed to remain on the journey.
A huge set of obstacles come from both explicit and implicit biases. Everyone’s biases start to form at very early ages based on what’s seen on television, in movies, in advertising, in communities, schools, and workplaces, and throughout other aspects of life. If you don’t think that this is the case, close your eyes and imagine a surgeon, a leader, a cool person, a good friend, a President, or a CEO. What do you picture? If that person is of a particular gender or race, you have a bias, period. That’s because none of these roles should have anything to do with either gender or race as the following tweet emphasizes:
“You don’t look like a surgeon.”
Actually, I do. What you mean to say is I don’t look like a middle-aged white guy… Thank you.#ILookLikeASurgeon
— Jamie Coleman (@JJcolemanMD) January 27, 2018
Similarly, when you meet a person for the first time, you can’t help but form an impression even before that person speaks a single word. Invariably, that initial impression doesn’t end up matching reality because how can really know a person until you’ve actually interacted extensively with that person? Unfortunately, such initial impressions can be like Instagram filters and cloud any subsequent opinion of that person. For example, if you don’t picture a woman as a surgeon, you will look for reasons why that person should not be a surgeon.
So what needs to be done? In a word, lots. In three words, lots and lots. Admitting that gender and racial bias are prevalent will at least help women and minorities better prepare and resist internalizing or blaming themselves when faced with biases. I have seen Chairs of Departments try to label as “resistant” or “difficult” women or minorities who have raised concerns about unfair treatment or obstacles to their career advancement. There is a distinct risk of believing these labels if no one admits that bias is occurring. There is also the risk that many people won’t fulfill their potential to help society because they were effectively barred from positions to do so. But admission is simply the first step.
Salles said, “There are so many other changes that need to be made, of course.” She mentioned “using standardized interviews and work sample tests” rather than “the standard ‘do I like you?’ interview” when selecting people for positions. Also, Salles said, “Transparency in hiring and promotion is very important and yet is mostly lacking in medicine. Being intentional about who is nominated for important roles and awards is a key to trying to level the playing field.” Yes, hiring and promotion in medicine can often be as transparent as constipating poop.
The design of this study and the IAT offers insights as well. Salles emphasized, “One really important piece of this is to look beyond the binary (man/woman) world of the IAT. That classification doesn’t capture everyone. We also know that carrying more than one marginalized identity (e.g., minority race and gender) likely has a multiplicative or exponential effect. We need more research to understand how intersectionality plays out in academics.” This further emphasizes the importance of looking at people as individuals rather than just things in a bucket.
Ultimately, things won’t really change until people get used to seeing a greater diversity of people in different roles. People need to see more women and minorities as surgeons and in leadership positions. And these can’t just be tokens who just want to toe the existing party line and aren’t willing to voice differing opinions and admit that bias is occurring. After all, what do you want in a surgeon or a leader, someone who just “looks” the part or someone who can really do the job the best?