A sense of foreboding descended on Narjust Duma as she sat at a presentation on drug-induced toxicities. The year was 2018, Duma was a 31-year-old second-year fellow at Mayo Clinic, and her discomfort stemmed from something other than the subject matter discussed.
Duma was attending a panel at the annual meeting of the American Society of Clinical Oncology, and on the stage were three physicians—one woman and two men.
The session chair, a man, introduced himself and the other man presenter by name and title. The woman—whom Duma knew to have the most experience and deepest understanding of the subject—was introduced by first name only.
Duma was taken aback.
Is oncology a field where men are introduced as “Drs. X, Y and Z,” while women get a “Jenny” or “Rashmi,” or “Heidi”? If this woman, despite being renowned for her expertise, can be so casually stripped of her title, how can a young oncologist like Duma hope to earn the same respect as a man with the same credentials?
“As she was being introduced by her first name—and she’s a full professor—I was thinking, then what are the hopes for me?” Duma said recently to The Cancer Letter.
The thought set off a firestorm, inspiring policy change throughout academic oncology, but it came at a price—harassment.
Duma was familiar with a 2017 study published outside oncology: The study gauged bias in introductions of speakers at grand rounds at Mayo, Duma’s institution at the time. (Another study later exposed this behavior in introductions at the 2017 annual meeting of the American Society of Colon and Rectal Surgeons).
Could something similar be happening at ASCO, the largest international oncology conference?
As a Latina in medicine, Duma felt this all the more acutely. “On top of being a woman, and also a minority—that really scared me.” Of 688,468 practicing physicians, only 5.2% identify as Hispanic, according to a study published in JAMA Internal Medicine in 2015.
“I’m a fellow. I’m almost done doing all this training that takes forever, and now this is stripping me of my title,” thought Duma, now an assistant professor and thoracic oncologist at Carbone Cancer Center at the University of Wisconsin-Madison.
“What is the future if this is happening right now? Is this just me?”
How pervasive is this bias, and how high does this problem reach in the field’s hierarchy?
At 9:05 a.m. CT, June 5, 2018, with the session still in progress, Duma tweeted out a question: “Do you think/observe women at #ASCO18 were more likely to be called by their first name than their male counterparts? (e.g., Instead of Dr. X).”
The Twitterverse answered. In this initial straw poll, the majority—63%—said yes.
The answers inspired Duma to dig deeper. In response to her poll, other researchers approached Duma with suggestions to study this question with scientific rigor.
A year after her tweet, Duma presented their results at a session titled “Evaluating Unconscious Bias: Speaker Introductions at an International Oncology Conference” at the 2019 ASCO annual meeting.
The Duma et al. study, published Oct. 11 in The Journal of Clinical Oncology, confirmed the results of the informal Twitter poll.
Women leaders in oncology said to The Cancer Letter that the Duma et al. findings confirm an open secret, amplifying it from whispered anecdotes and online chatter to robust data capable of informing policy and changing culture.
Informal introductions signal a lack of respect, which perpetuates gender disparities in health care. Unconscious or not, this bias could well have been holding women back for decades, even as more women than men continue to enroll in medical schools.
“Differential formality in speaker introductions may amplify isolation, marginalization, and professional discomfiture expressed by women faculty in academic medicine,” Duma et al. wrote, citing the Mayo study, which was published in 2017.
In 1960, 6% of all practicing physicians were women; in 2000, the proportion of women rose to about 30%. In 2017, the number of women enrolling in medical schools in the United States exceeded the number of men, at 50.7%.
Despite this increase, women make up only 42% of faculty at U.S. medical schools, 25% of full professors, and only 19% of department or division chairs. In the 2018 Association of American Medical Colleges report, the average rate of promotion of women to associate professor was significantly lower than that of their men peers, 41% versus 59%.
Nine of the 71 NCI-designated cancer centers are led by women directors—less than 13%.
“One of the things I think is probably underappreciated is how few women cancer center directors there are of the 71 NCI-designated cancer centers,” said Karen Knudsen, director of Sidney Kimmel Cancer Center, professor in the Department of Cancer Biology at Jefferson Medical College of Thomas Jefferson, and chair of Cancer Biology.
The Duma et al. paper led to immediate change.
Several directors of cancer centers said they are talking about the study with their faculty and staff, and ASCO has implemented a rule that will require all chairs at scientific, educational, and policy sessions to introduce speakers formally, with their full professional titles.
The study brings about awareness, a crucial first step to achieving gender equity, Knudsen said.
“I view this study as part of that effort to try to enlighten women to say that they can and should be seen in a gender neutral way as capable leaders,” Knudsen said. “If that starts as part of an awareness, that there is an unconscious bias in terms of using titles, then it’s a small step toward the greater good of really achieving parity in the workforce.”
Seven prominent women leaders in oncology—Knudsen, Cheryl Willman, Reshma Jagsi, Nancy Davidson, Cornelia Ulrich, Caryn Lerman, and Sharon Stack—spoke with The Cancer Letter about their experiences with gender bias, implicit and explicit. Their stories appear here.
“It’s important that we are mindful, and are working towards a really true, equal environment,” said Ulrich, director of the Comprehensive Cancer Center at Huntsman Cancer Institute and the Jon M. and Karen Huntsman Presidential Professor in Cancer Research.
“That means that if there are multiple speakers, that all the men and women are equally introduced either by their first names or all by their last name—that there are no differences made. What we should say is that the professional titles are critical.”
In “Evaluating Unconscious Bias: Speaker Introductions at an International Oncology Conference,” Duma et al. analyzed 781 introductions from the 2017 and 2018 ASCO annual meetings, finding that women speakers were addressed less often by their professional title, compared with men speakers, 62% versus 81%.
Men were less likely to use a professional title when introducing women speakers, compared with women who introduced men speakers, 53% versus 80%, the study found. Men introducers were more likely to address women speakers by first name only, compared with men introducers, 24% versus 7%.
Duma’s name is now likely to be recognized at conferences.
“Shoot, she’s the one with the paper,” a man researcher at another oncology conference must have thought while introducing her, according to Duma. Unfortunately, he introduced her by her nickname, NJ, an abbreviation of her first name.
“He corrected it right away and said, ‘Dr. Duma.’ So, that was enough for me to say that our paper has an impact,” Duma said. “Because, yeah, I have been introduced as ‘NJ’ way too many times.
“If somebody introduces me as NJ in front of a conference instead of ‘Dr. Duma,’ the audience won’t take me as seriously,” Duma said. “And that’s actually even worse than introducing me by my first name—if I’m introduced by nickname.”
Have you ever felt disrespected or unsafe because of your gender in a professional setting in health care? The Cancer Letter wants to hear about your experiences as a health care professional. Please fill out our confidential short survey.
The Wolf Pack
When Miriam Knoll saw Duma’s Twitter poll on June 5, 2018, she immediately recognized its significance.
“Hey—Great poll,” Knoll, medical director of the Department of Radiation Oncology at Hackensack University Medical Center/Mountainside, wrote to Duma over direct message.
Duma’s Twitter poll reminded Knoll of a daily affliction—being just “Miriam” in the professional setting.
“This is probably something we can formally study—ex: review the videos from the plenaries,” Knoll wrote. “To get the actual data…”
Gender bias in introductions wasn’t an issue that had been formally recognized and discussed openly in public with other doctors—men doctors, Knoll said.
“Every woman physician that I had spoken to, ever, not just in oncology, has said that this is something that they’ve witnessed—that they are addressed, introduced, responded to by their first name and not by ‘doctor,’ and that’s in stark contrast to their male colleagues,” Knoll said to The Cancer Letter.
“It’s informally in the hospital. It’s in writing. In response to emails. And this, specifically, you’ll notice, is not by people that we know well. It’s not by friends and colleagues. It’s by people that we don’t know,” Knoll said.
Before Duma’s hypothesis, before her data, before the policy changes, there was a secret, invite-only Facebook group called the Wolf Pack. Today, with 1,563 in the pack, women in hematology and oncology continue to get together to discuss clinical questions, exchange career advice, and talk through issues of gender bias.
“Two years ago, you could talk about the Wolf Pack, but you couldn’t talk too much,” said Duma, a pack member since 2017. “It was like the Illuminati.”
In 2016, Knoll noted that child care wasn’t available at ASCO—an issue she voiced in the group, and again in a blog she published in ASCO Connection. She later turned her hypothesis into a scientific survey with another Wolf-Packer, Reshma Jagsi, and published the 2019 results, titled “Association of Gender and Parenthood With Conference Attendance Among Early Career Oncologists” in JAMA Oncology.
Every woman physician that I had spoken to, ever, not just in oncology, has said that this is something that they’ve witnessed— that they are addressed, introduced, responded to by their first name and not by ‘doctor,’ and that’s in stark contrast to their male colleagues.Miriam Knoll
Jagsi is a professor, deputy chair, and residency program director in the Department of Radiation Oncology, and director of the Center for Bioethics and Social Sciences in Medicine at the University of Michigan. She is also a member of the ASCO board of directors.
“It is difficult as a parent, specifically, to attend conferences,” said Knoll, who is also the senior author on the Duma et al. paper. “Female oncologists really want to go, they want to be there. And they think that it’s important. And ASCO decided to offer onsite childcare, and they did. That policy implementation—which was so important—happened.”
ASCO offered free onsite child care for the first time at the 2019 meeting.
Social media was the key for women to connect and describe this form of bias, Knoll said.
Since Duma first joined, the Wolf Pack has become more well-known. At ASCO 2019, women in the Wolf Pack identified one another by placing a red circle on their badges. They were a team.
“[This study] is really a great example of building consensus on social media and recognizing yourself in others,” Knoll said. “Because what social media allows is to comment on, ‘I noticed something that you may have been feeling and wondering about for a long time, but thought maybe it was only yourself, and then you realize that, no, this is something that other women, other doctors are feeling as well.’”
On June 3, 2019, a full year after Duma ran her poll on Twitter and first connected with Knoll, they presented their findings together in a Clinical Science Symposia session at ASCO.
“This abstract deserves attn of all at this mtg”
On June 2, a day before Duma’s presentation, a tweet from Tatiana Prowell drew the ire of online trolls.
Prowell, a breast oncologist at FDA and associate professor of oncology at Johns Hopkins Sidney Kimmel Comprehensive Cancer Center, was one of four who retweeted Duma’s initial Twitter poll.
Prowell is part of the Wolf Pack and is active in the #MedTwitter community.
Aware that sessions on diversity-related issues are often not well-attended at conferences, Prowell, at the time the chair-elect of the society’s Annual Meeting Education Committee, wanted to drum up attendance for Duma’s presentation.
At 7:46 A.M. CT on June 2, Prowell tweeted: “This abstract deserves attn of all at this mtg, whether an #ASCO19 session chair or in the audience. Dr. @NarjustDumaMD et al analyzed assoc of race/gender & use of professional title during speaker intros at @ASCO mtgs. Hear full results Mon 2p in S100BC. cc @HemOncWomenDocs.”
Prowell’s tweet, which contains the abstract, has received 228 retweets and 554 likes.
“I wanted to be sure people came to the session, so I tweeted about it before the session took place to try to get people in the room,” Prowell said to The Cancer Letter. “Because I thought—the authors have done this important work, everyone in this meeting is going to see speakers introduced, or introduce speakers, or be introduced—or all of the above. They need to be in the room to be made aware of this—acutely, during this meeting. Before it’s over.”
Initially, there was a lot of support for the tweet and a lot of support for the results. “Men and women were congratulating [the authors] for doing the painstaking work of going back and watching all of the videos from all of the sessions from two years of annual meeting,” Prowell said.
Prowell’s tweet offered no further interpretation of the data. She didn’t point fingers, didn’t make accusations. No men were attacked, but some might have thought they were:
“#metwomovemrnt. Enough with political correctness.”
“… And amazingly, these are supposedly the country’s intellectual elite, MDs and PhDs. What a mess the rest of the country must be. Or who knows, maybe it’s better…”
“When did we as scientists begin addressing criticisms of supposed scientific studies by name calling? And where does medical science go in an environment where this attitude is commonplace?”
“Is it a good idea to encourage documentation of gender inequality?”
Duma continues to receive online hate mail.
“I want to call it how it is—harassment—because it was crazy,” Duma said. Prominent faculty members from other institutions sent Duma messages advising her to “stick to lung cancer” and telling her that she “was a good researcher until now.”
I want to call it how it is—harassment— because it was crazy.Narjust Duma
“People were saying, this is an insult to intellectuality, this isn’t science,” Duma said. “Some of them were comparing me to the anti-vaccination movement. It came from everywhere, from a lot of different types of people. I did delete a lot of the messages on Twitter because I didn’t want to see them.”
As a taxpayer-funded employee, Prowell believes she should be accessible to the public. Her Twitter settings allow anyone to send her private messages, which were often more hurtful than the public replies.
“I was honestly surprised that the tweet generated that much attention, let alone controversy. It’s science and not really that different from any other study I might tweet about to my mind,” Prowell said.
Duma received hundreds of direct messages on Twitter. She received emails to her institutional address. People who didn’t even know her, who declined to reveal their names publicly, took the time to dig for her information and contact her directly.
“People were a little bit more cruel by email. Because it is not out there, so you feel more free to say whatever,” Duma said.
She was weeks away from finishing her training—“I was still a fellow, and I think trainees are considered to be a protected population,” Duma said. “We’re learning. We are still young.”
Before the study was published, Duma was most-commonly accused of dividing the ASCO community. “The other most common message was that we were cherry picking the data—when we reviewed all of the videos. Not one, all of them.”
Some of the messages received by women in the Twitter conversation appeared to be threatening.
“I got a string of emojis that included a dagger, a heart, and a crying face,” Prowell said.
On June 3, Duma was in no position to put the uproar in perspective. This was her first-ever oral presentation.
There were about 40 people in attendance. About 25 of them were women from the Wolf Pack. There were about half-a-dozen men, Prowell estimates.
Of 2,511 videos reviewed from ASCO’s 2017 and 2018 annual meetings, 781 met inclusion criteria. Women were addressed less often by their professional title, compared with men speakers (62% v 81%; P , .001). Men were less likely to use a professional address when introducing women speakers, compared with women introducing men speakers (53% v 80%; P , .01).
When women performed speaker introductions, no gender differences in professional address were observed (75% v 82%; P = .13). Women speakers were more likely to be introduced by first name only (17% v 3%; P , .001). Men introducers were more likely to address women speakers by first name only compared with women introducers (24% v 7%; P , .01).
In a multivariable regression including gender, degree, academic rank, and geographic location of the speaker’s institution, men speakers were more likely to receive a professional address compared with women speakers (odds ratio, 2.43; 95% CI, 1.71 to 3.47; P , .01).
The study’s conclusion: “When introduced by men, female speakers were less likely to receive a professional address and more likely to be introduced by first name only compared with their male peers.”
The presentation seemed to have gone smoothly. After the talk, while Duma was in conversation with ASCO staff members about reporting the harassment, she noticed two security guards stationed by the door.
“I learned that the ASCO team had brought some guards, just to be safe. I think it was two people. It wasn’t the detail of Queen Elizabeth,” Duma said.
ASCO brought in security guards specifically because of the harassment she received.
“’Ay dios mio, thank God you told me after my presentation,’” Duma thought, relieved she hadn’t been distracted while focusing on the data. “I didn’t think anybody would get physical with us, but we were just afraid that somebody would be aggressive. This is the first time this type of data was ever presented at ASCO.”
ASCO also sent cease-and-desist letters to the perpetrators.
“ASCO staff spontaneously reached out to let me know they were monitoring the conversation taking place on Twitter,” Prowell said. “They were very focused on ensuring a safe environment for everyone, which I appreciated.”
ASCO has since implemented directives for how to introduce speakers at all future meetings. The society plans to train all chairs on unconscious bias ahead of the 2020 annual meeting, Jamie H. Von Roenn, vice president of education, science and professional development at ASCO, said to The Cancer Letter.
“The results of this study and some discussions that were ongoing regarding ASCO’s commitment to gender equity across its programs—which the board has been following for a couple years—really led to this change,” Von Roenn said.
ASCO’s action on gender bias is part of a larger cultural shift, including the recent introduction of free onsite child care and private nursing rooms during the 2019 meeting, and a larger focus on patient-first language. The Duma et al. study is directly cited in ASCO’s recently published instructions written by Prowell and ASCO staff on meeting language: “The Language of Respect.”
ASCO will provide training for abstract reviewers, Von Roenn said, which will make them aware of how gender can influence how a reviewer evaluates an abstract.
“The more we can do to level the playing field, the better care we can provide,” she said.
How unconscious is “unconscious”?
Two other studies have found the same form of bias in other medical specialties:
A study published in the Journal of Women’s Health in 2017, “Speaker introductions at internal medicine grand rounds: Forms of address reveal gender bias,” honed in on speaker introductions at Mayo Clinic’s grand rounds. The Mayo study found women introducers were more likely to use professional titles when introducing any speaker during the first form of address compared with men introducers (96.2% vs. 65.6%).
Another study, published in Diseases of the Colon & Rectum in 2017, found women moderators were more likely to use formal introductions, compared to men moderators at the 2017 American Society of Colon and Rectal Surgeons’ annual meeting (68.7% vs. 54.0%). Men moderators were significantly less likely to formally introduce a woman versus a man speaker (36.4% vs. 59.2%).
“Studies like this are so important to show people these small unconscious biases, and how they can cumulatively have a significant effect, such that we’re never going to achieve equity unless we’re mindful of them and target them,” Jagsi said to The Cancer Letter. “Kudos to Dr. Duma for extending this to the setting of ASCO speaker introduction.
“It’s a brilliant idea to take a look at this and she, again, followed the very rigorous approach that Dr. Sharonne Hayes [senior author of the Mayo study] had established when she had done the Grand Rounds study, where she had mixed gender coders who were coding the introductions to remove the impact of any kind of bias. And, really, she did a well done study in a really important setting to study,” Jagsi said.
The retrospective analysis used by Duma et al. solidified the study’s findings, Sharon Stack, Ann F. Dunne & Elizabeth Riley Director of Harper Cancer Research Institute, Kleiderer-Pezold Professor of Biochemistry, Department of Chemistry & Biochemistry at the University of Notre Dame, said to The Cancer Letter.
“Things can’t be changed if you say, ‘Oh, I think there’s an issue with the way people are being introduced,’ … the participants were not aware that this study was going on because this was retrospective data. [The researchers] could really go back and analyze it in a completely unbiased way,” Stack said. “I think that makes the data even stronger. The data were unequivocable, and the ranges were not subtle.”
On its face, “unconscious” bias may amount to a benefit of the doubt. If the issue is so pervasive, how could men—and less often, women—not be aware of it?
“It’s not that we think or suspect that there is overt discrimination, overt sexism, or a concerted effort to disrespect women,” Knoll said. “What is happening, clearly, is an unconscious bias. It’s unconscious inequity. It’s not that someone is looking to discredit women, but clearly there’s a difference between how men and women oncologists are being introduced.”
These behaviors are not intentional, said Jagsi, who is a founding member of Time’s Up Healthcare, a part of Time’s Up, an organization that advocates for causes that include gender equality in the workforce. “By and large, these are unconscious biases, and they’re called unconscious, because they are not conscious,” she said. “And the purpose of studies like these are to identify them so they can be targeted.”
What is happening, clearly, is an unconscious bias. It’s unconscious inequity.Miriam Knoll
Bias and prejudice have been the subject of extensive research in psychology, sociology, and anthropology. Gender bias in medicine and health care is now receiving increasing scrutiny.
So, is the gender discrepancy in introductions truly “unconscious,” in psychological terms?
In the social sciences, bias is generally considered to be either “implicit” (i.e. subconscious or unconscious) or “explicit” (conscious). Implicit biases are thought to develop from the continual association of a social group (such as “men”) with either a trait (such as “competent”) or attitude (good/bad), experts say.
“People reveal [implicit bias] unintentionally through their actions, without being aware of it, to culture, the set of meanings, symbolisms, and expectations based on what we are conditioned to give meaning to,” Dawne Moon, an associate professor of sociology in the Department of Social and Cultural Sciences at Marquette University, said to The Cancer Letter.
Moon provides an analogy:
In the same way that a driver would stop (habit of thought) when the traffic light turns red (culturally-defined meaning), perceptions of someone’s sex and gender can trigger psychological processes (chains of associations) at a lower level of awareness.
“What level of non-consciousness that operates on, I couldn’t really say. It wouldn’t be as a result of repression, though,” Moon said. “It would have to do with the social construction of symbolic associations in the brain.”
Introductions are “an objective measure of formality, an objective measure of professional respect. It’s really a great way to document and show unconscious bias,” Knoll said.
Every time a woman physician is stripped of a title her men colleagues get to keep, it perpetuates the falsehood that men are better, UMich’s Jagsi said.
“If it’s sowing an unconscious seed that man is worthy of respect and dignity, it is suggesting in some way that the woman is less significantly accomplished as a professional,” Jagsi said.
These implicit gender biases can be changed, Moon said.
“Those chains of association can be examined, rethought, and challenged,” Moon said. “Just because we tend to think ‘doctor = man’ and ‘nurse = woman’ doesn’t mean we can’t—with either intrinsic or extrinsic motivation—catch ourselves and remake those chains of association.”
Heidi and Howard
It’s not a coincidence that Duma’s study is published in the #MeToo era, said Ally Coll, a lawyer, president and founder of The Purple Campaign, a Washington non-profit that seeks to prevent workplace sexual harassment by implementing stronger corporate policies.
“This is a moment where people across all industries and professions are recognizing the way that gender discrimination plays out in a professional setting,” Coll said.
[Disclosure: Coll is a step-daughter of Paul Goldberg, editor and publisher of The Cancer Letter]
When a moderator introduces a woman speaker without her title or biography, the woman’s expertise can be undermined, Coll said.
“Women get penalized when they have to share their own bio or say their own title to get that credibility—people often, because of unconscious bias, they’ll perceive that as negative—as too self-promoting or selfish. Or they’ll hold it against women in a way that they don’t with men,” Coll said.
For example, in a 2003 experiment that has become known as the Heidi / Howard study, Harvard Business School students were presented with a case study of a successful entrepreneur based on Silicon Valley venture capitalist Heidi Roizen.
Both cohorts were given the same profile: one named “Heidi” and the other “Howard.”
Students in both groups ranked Heidi and Howard as equally accomplished, but Howard was seen as an appealing colleague, while Heidi was regarded as selfish or unpleasant.
“When I’m involved in panels or other events, it is a better practice for the moderator or the introducer to do that work for the panelist, rather than my having to say that to prove my credentials to the room,” Coll said.
The audience responds differently if Coll were to introduce herself.
“They’re going to have a higher likelihood of seeing it as bragging, or selfish, or self-promoting in a negative way. And I think that the Heidi / Howard study shows that when men have to do that, they don’t have necessarily the same backlash effect,” she added. “This might seem like a little thing, but it’s not.”
Recognition by title at conferences have an effect on acknowledgement in other professional settings, Coll said.
“That’s why it’s so important for people in a workplace setting, whether that’s events or in hiring, in promotion decisions, to create neutral processes that everyone just has to follow,” Coll said. “When you leave things for individual discretion, that’s when unconscious bias can come out—and we see that in decisions around hiring a lot.”
Women directors of cancer centers who spoke with The Cancer Letter are addressing gender bias in their institutions.
Knudsen, who has previously been on many NCI site visits, said she had always noticed the discrepancy in who is addressed by name and title and who isn’t.
She was part of the decision to make sure all faculty and staff referred to each other by their title at a recent NCI site visit at Thomas Jefferson.
“We did it because we wanted there to be continuity in how we discussed and talked about each other,” Knudsen said. “So, we said, ‘Look, it has to be one or the other—we were either going to use first names or we were going to use titles. And we elected to use titles.”
NCI remarked on the consistent use of titles—regardless of gender—at Thomas Jefferson.
“I like to think we’re leading the way on this, and it just is a small step toward what we want to achieve—parity,” Knudsen said.
Despite ASCO’s new policies and heightened engagement on the issue, the problem persists, and is not unique to the “international oncology conference.”
On Dec. 11, Prowell spoke with this reporter while attending the San Antonio Breast Cancer Symposium.
“I saw multiple female speakers in one session addressed at the podium by first name only from the floor microphones. This did not happen to any of the men in the session,” Prowell said. “I don’t believe anyone doing this intends any disrespect. It’s unconscious bias.
“Fortunately, now that we are all aware of it, we can work towards solutions.”
Have you ever felt disrespected or unsafe because of your gender in a professional setting in health care?
We want to hear about your experiences. Please fill out our short survey.
The Cancer Letter will not use any identifying details without your consent. Any data that we publish will be de-identified and anonymized, including the names of any institutions.
Editor’s note: The Cancer Letter does not use titles and honorifics in its reporting, per the AP Stylebook.
Katie Goldberg and Matthew Ong contributed to this story.