*
‘The hardest attitude to change is the one you don’t know you have,’ wrote John Dovidio, a professor in the Department of Psychology at Yale University3.
I’m not sure whether Professor only-interested-in-marriage or the Cardiff medical students were aware of holding racist attitudes. But I am sure that each and every one of us holds unconscious beliefs about people whom we perceive to be different from ourselves. These beliefs may be outside our conscious awareness, but they still influence our judgements.
The ubiquity of unconscious bias was brought home to me recently, when my son texted me a link to a live BBC interview with political scientist Professor Robert Kelly, who was speaking from the office in his home4. During the course of the interview the door opens and a little girl in a bright yellow jersey dances in, eager to see what her father is doing. Seconds later, a younger child in a baby walker follows his sister and enters the frame; moments after that a harassed woman of Asian appearance bursts into the office and bundles both children out of the room.
If you’d asked me whether I held racist assumptions about Asian women, I would have vehemently denied it. Yet when my son texted me the Twitter link I’m ashamed to confess that, like millions of other social media users, I too assumed that the woman in the clip was the nanny, rather than the wife of Professor Kelly, a white man. In fact I ‘corrected’ my son when he referred to the woman as the children’s mother and immediately texted back ‘Not mum – nanny.’ My automatic response was freeze-framed as a text message on my phone; the immediacy of modern technology caught me in the act of unconscious bias.
The cabin crew on Delta DL945 flying from Detroit to Houston in October 2016 made a similar mistake. Following an in-flight call for medical assistance, Dr Tamika Cross, an African-American doctor, offered her services. Initially the flight attendants refused to believe she was a physician. Even when she eventually persuaded them of her professional status, her assistance was declined in favour of that of a white man, who offered no evidence that he was medically qualified. When Dr Cross landed, she posted an account of her experience on Facebook; quickly it went viral, with the hashtag #WhatADoctorLooksLike. The Facebook post elicited a deluge of similar accounts by other women whose claims that they were physicians were initially discounted on the basis of their skin colour, gender, or both5.
But it’s not just psychologists or airline crew who make these sorts of immediate judgements. Doctors are not immune. Indeed they can’t be – because the tendency to reach snap judgements based on stereotypes is hard-wired into all of our brains. The only difference in the medical context is that the people on the receiving end of these automatic responses are typically other doctors and patients. So, for example, Damon Tweedy, an African-American psychiatrist who trained at an elite, overwhelmingly white medical school in the 1990s, was mistaken by a lecturer for a maintenance worker who had come to sort out an electrical problem in the lecture theatre6. When Tweedy explained that he knew nothing about the electrics, the lecturer asked ‘Then what are you doing here?’ All of this took place in front of hundreds of other students, causing Tweedy considerable embarrassment and distress.
And of course, doctors’ unconscious biases also get elicited by patients7. These biases are about more than race; they may also be about gender, sexuality, age, or stigmatised health conditions such as obesity. In turn, there is compelling evidence that the unconscious biases of clinicians impact not only on the quality of their relationship with patients, and the likelihood that patients will follow their medical advice, but also on the actual treatment decisions that doctors make. For example, a 2012 study published in the American Journal of Public Health reported that paediatricians with greater pro-white bias were more likely to agree with prescribing a narcotic medication for postsurgical pain for a white patient, but more likely to disagree with prescribing it for an African-American patient presenting with the same symptoms8. Stereotypes about the misuse of painkillers in the black community got in the way – even though the patients were all young children. The authors of this paper also emphasise how the intense pressure of medical work – clinical uncertainty, a high workload, and fatigue – increases the risk of over-reliance on snap stereotypic judgements.
This is the inescapable reality of medical students’ and junior doctors’ work. In overstretched hospitals and GP surgeries, as well as in seminar rooms and lecture theatres, they cannot help but see the day-to-day workings of unconscious bias in action.
*
Being on the receiving end of stereotypical perceptions (or fearing that one might be) impairs performance. This effect was brilliantly demonstrated in a set of classic psychological experiments carried out in the Department of Psychology at Stanford University, which used university students as study participants. When the students arrived in the psychology lab, the experimenter (a white man) explained that for the next half hour they would work on a set of verbal problems in a format identical to the SAT (the standardised college entrance test). All of the participants would have previously taken the SAT in order to gain admittance to Stanford.
Each participant was then given a piece of paper outlining the purpose of the study, describing the procedure for answering questions, and stating the fact that test was very difficult, so they should not expect to get many of the questions correct. (Information on the test difficulty was included so that participants’ expectations about the test were equalised across the two different test conditions.) In fact, the only difference between the experimental and control conditions were the inclusion of key phrases in the page that described the purpose of the study.
Participants in the experimental group were told that the study was concerned with ‘various personal factors involved in performance on problems requiring reading and verbal reasoning abilities’. They were also told that, following the test, they would be provided with feedback which ‘may be helpful to you by familiarising you with some of your strengths and weaknesses’. Thus for participants in the experimental group, the test was framed as a test of individual verbal ability.
In contrast, participants in the control group were told that the purpose of the research was to better understand ‘the psychological factors involved in solving verbal problems’. Control group participants were also told that they would receive feedback after the test, but it was justified as a means of familiarising them with the kinds of problems that appear on tests, as opposed to giving them an indication of ‘personal’ strengths and weaknesses, as in the experimental group.
So what happened?
Black participants in the experimental group, where the test was framed as a test of individual ability, performed significantly worse than black participants in the control group – even though participants had been randomly assigned to both groups. The study authors, Steele and Aronson, explained the discrepancy between the two groups in terms of ‘stereotype threat’, which they defined as ‘being at risk of confirming, as a self-characteristic, a negative stereotype about one’s social group’9. In other words, if people fear that their individual performance might be viewed through the lens of a negative stereotype – their performance is likely to be impaired.
In a subsequent experiment reported in the same paper, Steele and Aronson examined whether increasing the saliency of race (by asking participants to classify themselves according to race, prior to taking the test) amplified anxieties. It was predicted that the impact of stereotype threat would be increased in the ‘primed’ group, where the issue of race was raised prior to participants taking the test, in contrast to the control group, where participants were asked to record their race after they had completed the test. The results confirmed this hypothesis, as black participants in the ‘primed’ group performed significantly worse than black participants in the control group.
Stanford University is a world-class institution. In the 2017 Times Higher Education league table it was ranked the top university in the US, and third in the world. Wit
hout exception, all of the students who took part in these psychology experiments would have been academic high-flyers. You don’t get into Stanford without an exceptional performance at secondary school level – including extremely high SAT scores. Yet tiny manipulations such as emphasising whether the test assesses individual ability, or highlighting cues about race, has an impact on performance. And this even happens to students who have already proven their academic credentials by gaining entrance to an elite university.
Since the publication of Steele and Aronson’s study in 1995, there’s been a veritable growth industry of research. If you type ‘stereotype threat’ into the main psychological database (Psychinfo) a list of 923 papers comes up. And whilst the initial study focused on the impact of racial stereotypes on academic performance, subsequent research has investigated many other stereotypes (gender, sexuality, social class, etc.), as well as outcomes beyond academic test results.
Despite the enormous amount of research into stereotype threat, there’s been minimal interest in exploring whether it might apply to medical training. Yet given that experiments have shown its impact in so many situations where stereotypes are pervasive – white men compared to Asian men in mathematics; men compared to women on social sensitivity; students from low socioeconomic backgrounds compared to those from wealthier homes on intellectual tasks, etc. – it would be impossible to construct an argument as to why stereotype threat wouldn’t occur amongst doctors. And as the initial demonstration of its influence was in a study of academic high-flyers, the notion that doctors are somehow too bright to be influenced by such things simply doesn’t hold water.
*
Moving beyond the psychology laboratory, the conversations I have had with doctors over the years often reference, explicitly or implicitly, what it feels like to be on the receiving end of negative stereotypes; feeling that one isn’t good enough, or one doesn’t quite fit in, or that one has to work harder to prove to other people that one has got what it takes to succeed or that the whole medical training system is desperately unfair. Of course, these conversations haven’t only been about race; gender, class and sometimes sexuality are also mentioned. And these different dimensions intersect; consultants would have responded differently to Rahma had she been a man, or from a privileged background.
The notion of ‘not belonging’ is one that crops up repeatedly. It also runs through much of the wider literature on stereotype threat. And what is one likely to do if one feels that one doesn’t belong to the wider professional culture? Stick with one’s own. This is certainly what happened to Rahma. She told me how her closest friends at medical school were all young women who came from Pakistani backgrounds, had grown up in her part of London and were living at home. The group had gelled naturally; in the first days at medical school one member of the group had been at secondary school with the cousin of another; there were family and geographic links as well as cultural familiarity. After lectures they tended to be travelling home in the same general direction – so a close-knit group was quickly formed. And the fact that Rahma and her friends didn’t feel comfortable socialising in the pub may also have contributed to their relative isolation from the broader medical school year.
Rahma’s experience is repeated across the country; students tend to form social networks with others of the same gender and ethnicity. But students don’t only learn in lectures – they also learn from their peers. Friends may share hints about new resources, or how best to prepare for an examination. Students within a network also pick up study habits – for good or for worse – from those around them and there is evidence that medical students’ choice of friends has a significant influence on their examination grades, even after taking their previous grades into account. Another educational risk, in other words, for marginalised BME students.
I also suspect that the assaults on one’s sense of belonging are particularly potent in medicine. If one compares medicine, say, to law – another high-status profession with competitive entry – one finds that medical students spend a huge amount of time in hospital seeing patients (under supervision), whilst law students don’t spend a comparable proportion of their time with legal clients. But interactions with patients open up a myriad other ways in which one can come to the conclusion that one has no place in the medical profession. I have never forgotten a brilliant young man from an extremely impoverished white working-class background; his mother was a school cleaner and his father a manual labourer, unable to work because of alcoholism. When he was observed examining patients – particularly if they came from similar backgrounds to his own – he was criticised by his consultants for his accent, or for his less than hyper-formal grammar. The fact that these patients were delighted to be treated by somebody from their neck of the woods didn’t alter the negative feedback he received from these senior clinicians.
And it’s the same with race. BME doctors will have a different experience of looking after patients than their white peers. One of the incidents that emerged from the inquiry into Cardiff Medical School concerned a patient who refused to be examined by a particular medical student because the student was black. The student was unsupported. More worryingly, some staff felt that this incident should be treated in exactly the same way as if the patient had simply said that they preferred not to have any medical student present. Should it really need spelling out, to somebody who trains doctors, that a patient saying they don’t want you to examine them because you are black is not the same as a patient saying they don’t want you to examine them because you are a medical student?
Then there’s the potential for comments from relatives, or from other members of the healthcare team to corrode (or occasionally boost) a fledgling doctor’s sense of professional identity. Damon Tweedy, one of the few black medical students in his elite medical school in the 1990s, described his fears that colleagues’ views about the black, overweight, diabetic patients they encountered in a rural free clinic would also influence how they saw him – the only black member of the team10.
‘I had a foot in both worlds – but didn’t have two feet in either,’ was how he put it.
Of course there is a potential upside. Tweedy also noted the delight when the black people he encountered – patients, relatives, nurses, receptionists, cafeteria workers, cleaners – saw him in his white coat. ‘They shared in my achievements and promise like an extended family,’ he wrote. But Tweedy went on to stress that, even with the positive feedback, there was a potential sting in the tail: ‘Along with the racial pride that came with their praise and adulation, however, I felt an added weight, as if my success or failure would reflect not just on me, but on those who had come before and on those who would follow me.’
*
When we first met, Rahma informed me that she would never be able to work as a doctor. On her first day in her first job, when it became clear to her consultant that she was too depressed to work, she was signed off sick. After a short period of sick leave she returned to work but her self-doubt and depression gradually built up to the point that once again she was unable to work.
The cumulative effect of being viewed through a stereotypic prism probably wasn’t the only reason she became so distressed as soon as she started work. Rahma’s family background was complicated and, as we saw in an earlier chapter, when a person assumes the role of carer, the residue of how they were themselves cared for in childhood can ripple through their psyche. Faced with the responsibility of looking after sick and needy patients, Rahma’s difficult feelings related to her own experiences in childhood may well have been activated.
There’s also the issue of wider family responsibilities. Many junior doctors – particularly from BME backgrounds – have described their significant roles in supporting other family members. Understandably, when somebody is sick, if a family member is a medical student or junior doctor, their opinion or advice is likely to be sought. But there is evidence that both the size of the group who are considered ‘family’ and th
e weight of the expectations placed on the student/doctor may mean that BME doctors are expected to become more involved in family crises than their white peers11. This certainly happened to Rahma; increasingly other family members turned to her for advice and support with their growing list of health problems.
Then there was the issue of living at home. Apart from a short period during her elective, Rahma had never lived away from home. This meant that the quantum leap from medical student to junior doctor coincided with her having to live independently for the first time. ‘I’m an adult, why am I feeling like this?’ she asked herself. Being surrounded by people who appeared excited about the prospect of starting work made it even harder. ‘All I felt was fear,’ she told me. Rahma was not alone; other studies have shown that BME medical students in the UK feel less well prepared for the transition to work, although the underlying reasons for this (not having left home before/erosion of confidence in medical school) haven’t yet been teased out12.
If I look at how Rahma has been transformed, I am clear that her referral to the specialist psychotherapy service for doctors was crucially important13. She formed an excellent therapeutic relationship with one of the psychiatrists, who helped her understand why she had been so acutely distressed when she started work. Moreover, Rahma was not alone in finding this specialist psychotherapy service beneficial; a recent follow-up study of 124 doctors found that over 95% who had used the service continued to work and progress in their medical careers, that they took less medication for physical ailments and became more aware of their psychological needs, were less likely to set themselves unachievable targets at work and were better able to access help from colleagues.
Also Human Page 19