Also Human

by Caroline Elton

  I have known Rahma for eight years. Today she is a fully qualified GP; a desperately distressed fledgling doctor has metamorphosed into a quietly confident clinician. Over the past eight years there have been a number of times when Rahma has come to see me. Initially she had hoped to specialise in dermatology; she passed her physician exams, but didn’t succeed in getting a training place. After a couple of attempts she changed track and became a GP but she carries a strand of sadness that her first choice of specialty didn’t work out. Currently, alongside her clinical work, she has a significant role in research. When she talked about her research interests at our recent meeting, her commitment to this work became clear. Above all else, she wants to ensure that junior clinical researchers, from all backgrounds, are able to realise their potential.

  *

  ‘You’re wrong,’ said the man at the back of the seminar room. ‘Completely wrong. It’s not the case that white students do better at medical school, or as junior doctors. Not at all. In fact they do worse than their BME peers.’

  The speaker, a middle-aged hospital consultant, was one of my students on a Master’s course in Medical Education. Despite the fact that I was the module leader, so vehement was his response that I began to doubt what I knew. Maybe I’d got it wrong when I told my seminar group that British students from BME backgrounds got poorer grades than their white peers, and these differences remained, even when you took social class, or the grades of entry into medical school, into account. The issue of race wasn’t central to my teaching that afternoon; I’d gone ‘off-piste’ in discussing the issue in the first place. I backed down.

  ‘I’ll check the data,’ I told my challenger. So that’s what I did.

  The following week, I returned to my seminar group. Armed with a lengthy reference list, I felt that I was on solid ground. Although my challenger didn’t respond with quite the same level of opposition, I don’t think I convinced him that differential attainment actually exists.

  To be fair to the consultant in the seminar as well as to the many other white medical students and doctors who remain in blissful ignorance, the problem of differential attainment in the UK is somewhat masked by the high proportion of BME doctors gaining entrance to medical school. According to figures from HESA14 – the government-funded agency responsible for collecting statistics on the higher education sector – just over one in five (20.6%) of UK-domiciled students come from BME backgrounds. However, when it comes to medical and dental students, the proportion of UK-domiciled BME students is significantly higher: 33.3%. With one in three UK medical and dental students coming from a BME background, it is perhaps easy to see why the myth persists that medical training in the UK has rid itself of racial discrimination. Unfortunately, it is nothing more than that. A myth. Just like the issue of sexism in medicine, the problem isn’t access to the first step on the ladder – gaining a place at medical school. It’s what happens after that.

  So how does the UK compare to the US? The National Center for Education Statistics (NCES) reported that amongst US students enrolled in university in 2014, 13.8% defined themselves as having a ‘black/African-American’ background15. But if you drill down and look specifically at medical school, the proportion of students coming from this group fell to 6.1%. In the US, black/African-American students are less likely to study medicine than other subjects. With Asian students in the US, the situation is reversed. NCES statistics reported that 6.6% of students enrolled in university in 2014 came from Asian backgrounds whilst over three times that number of medical students – 20.5% – were from this group16.

  A quick glance at the data might suggest that the UK has done better than the US in opening up medical training to BME students. But this would probably be misleading. The UK data groups together students from Asian, African and Afro-Caribbean backgrounds into the ‘BME’ category17. Other studies indicate that young people from Afro-Caribbean backgrounds are less likely to attend university – and are particularly poorly represented in medical school. The fact that a third of medical students in the UK are from BME backgrounds is largely accounted for by the high proportion of students with Asian backgrounds – not by those from Afro-Caribbean homes. Students from that ethnic group – like those from black and African-American backgrounds in the US – are poorly represented.

  *

  I open my filing cabinet and start pulling out the notes of doctors whom I have recently seen: Sunil, an international medical graduate from India who dreams of being an ophthalmologist, but never gets through the selection process. Then there’s Bindu, a UK graduate of Asian origin who is training as a GP, even though her heart had been set on surgery. She knew the competition statistics and in the end decided to play it safe and abandon her surgical ambitions. And there’s Tama who is about to be kicked off her training programme in anaesthetics. For each of these doctors, medical work has become toxic. The consultant in the seminar room would probably regard the contents of my filing cabinet as little more than anecdote. But there’s a hefty body of evidence to suggest he is wrong.

  One of the most comprehensive studies on ethnic differentials was carried out by Kath Woolf and colleagues, and published in the British Medical Journal in 201118. Analysing data from nearly 24,000 candidates in the UK, they reached the following conclusion:

  Ethnic differences in attainment seem to be a consistent feature of medical education in the UK being present across medical schools, exam types, and undergraduate and postgraduate assessments, and have persisted for at least the past three decades. They cannot be dismissed as atypical or local problems.

  In an accompanying editorial to the article, Aneez Esmail, a professor at Manchester Medical School, commented that all medical schools and all specialty royal colleges should analyse their assessment results by ethnic group19; they are already required by law to hold such data, but Esmail argued that they need to analyse the data and then place the results in the public domain.

  How can one disagree with this suggestion? Knowing who gets a particular disease, and in what particular circumstances, forms the bedrock of medical research. One only has to think of the classic research carried out by Sir Richard Doll in the 1950s, in which he followed up 40,000 doctors over a two-and-a-half-year period and demonstrated that as the number of cigarettes smoked increased, so too did the risk of death from lung cancer20. Numbers matter not only in medical research – but also in medical education.

  In the UK, the GMC is now publishing data that examine the relationship between ethnicity and progression through medical school and beyond. Just as Doll put evidence about the links between smoking and lung cancer in the public domain – thus kick-starting public health campaigns to reduce smoking – the fact that data are publicly available on the GMC website is a vital first step. And the figures demonstrate, beyond all doubt, that there is a real problem.

  If one looks, for example, at progression through GP training, 10.3% of international medical graduates (IMGs) are graded as failing to make adequate progress. The comparable figure for UK medical graduates is 3.5%. Similar discrepancies between UK graduates and international medical graduates have been observed in other postgraduate training programmes such as anaesthetics and paediatrics. Being an IMG significantly increases the risk of failure21.

  The place of primary medical qualification, however, is not the only factor influencing lack of progress; the colour of your skin also counts. This becomes clear when one sees that, on average, British BME colleagues encounter greater difficulties in progression than their white British colleagues. They don’t tend to do as badly as doctors who qualified outside the UK – but on average they don’t do as well as white doctors. In fact it’s almost as if there is a dose effect as in Doll’s research on smoking. It’s better to be a non-smoker (white) than a smoker (have black or brown skin). But amongst smokers (those with black or brown skin), it’s better to be UK born (just puffing away at a few cigarettes a day) than be born in India, or Pakistan, or Africa (having a two
-packets-a-day habit). Doctors who fall into that category are at risk.

  *

  Trying to unpick differential attainment in the American medical system is much harder. One of the key players, the Association of American Medical Colleges (AAMC)22, puts data in the public domain. From open access data that the AAMC provides it’s apparent that black and African-American students tend to enter medical school with lower grades on the MCAT (Medical College Admission Test) than their white peers. It’s also clear from AAMC data that black and African-American students who successfully gain a place at medical school tend to come from significantly poorer socioeconomic backgrounds than white students23. For example, the largest proportion of black and African-American students come from the poorest backgrounds and the smallest proportion from the wealthiest income group. With white medical students the situation is reversed.

  But the AAMC doesn’t have responsibility for the United States Medical Licensing Examination (USMLE) that medical students and doctors have to pass, in order to work as a physician in the US24. And unlike the AAMC, the USMLE has a very different approach. Information about international medical graduates is shared – and, as in the UK, international medical graduates perform less well on each of the three stages of the USMLE sequence. But the USMLE doesn’t readily share information on how US medical students of different ethnicities perform.

  Studies of ethnic differences in USMLE examination performance are few and far between. An exception is a 2012 study of students who started medical school between 1993 and 2000, which found that black and African-American students were significantly more likely to fail the first examination (STEP1) than their white peers25. Given that students from black and African-American backgrounds enter medical school with lower MCAT scores, and MCAT scores are related to STEP1 scores, this result is not surprising26.

  The problem is, however, that STEP1 scores then take on a life of their own, as they are used as a screening tool for interviews into residency programmes. As Charles Prober, a physician at Stanford Medical School, explains27:

  On one hand we pride ourselves in teaching medical students to use diagnostic tests for their designed purpose, to be critical thinkers and to use evidence-based support to guide their decisions. On the other hand [residency] programs may make career-changing decisions about medical school graduates based on overweighting a screening test in a manner not supported by strong evidence and for which the test was not specifically designed.

  Not only is the wrong screening test (i.e. STEP1 scores) used as part of residency selection; it’s also a test that is likely to discriminate against African-American applicants28: ‘When STEP1 scores are used to screen applicants for residency interview, a significantly greater proportion of African-American students will be refused an interview,’ concluded one researcher based on applications to his particular medical school.

  But we simply don’t know how big the problem is across the US as a whole. And the reason we don’t know is because of another organisation, the National Resident Matching Program (NRMP). Each year the NRMP publishes a report on the outcomes of the match – a 211-page document in 2016. But the document tells you nothing about race or about gender. Across the whole 211 pages there isn’t a single analysis of these basic demographic criteria.

  Perplexed, I emailed the Director of Research at the NRMP and asked if the organisation conducted any research into these issues. The same day I received a ten-word response:

  ‘The NRMP does not collect any race or gender data.’

  *

  We’re a little bit ahead of the game in the UK. In 2015 the GMC described their commitment to:

  collecting and publishing a range of outcome data, and analysing it to better understand variation in performance and attainment … We’re committed to making sure training pathways are fair for all and have a focused work programme to investigate, understand and take action where we find evidence of unfairness or unsupportive environments29.

  Although the GMC had started researching the causes of racial disparities prior to 2014, undoubtedly the issue gained significant momentum when the British Association of Physicians of Indian Origin (BAPIO) sought legal redress in the High Court. The matter at stake was whether the final stage of the GP licensing examination should be declared unlawful because it discriminated against BME doctors.

  The judge’s final verdict was that the examination did not directly or indirectly discriminate against BME doctors. However, following the court case the GMC commissioned two major studies30. The first was a comprehensive literature review published in 2015. This review concluded that, although there isn’t yet agreement as to the causes of differential attainment, an adequate explanation is likely to involve multiple factors. The authors went on to say that any attempt to mitigate or address differential attainment will have to take account of factors operating at the macro level (policy), the meso level (the institution) and the micro level (individuals and small groups). No simple answers, it seems.

  A second study commissioned by the GMC using a focus group and interviews also highlighted the complexity of the issues, but through the use of two words – ‘psychologically risky’ – the whole debate was radically reconfigured31. Differential attainment can’t be attributed to deficits in the students, the study concluded; it stems from the fact that whilst medical training is ‘psychologically risky’ for everyone, the risks are heightened if you don’t happen to be white.

  The risks that all postgraduate doctors face include ‘dysfunctional and highly-pressurised environments, bullying, lack of autonomy, lack of work–life balance, and lack of confidence’. That’s the day-to-day diet for all junior doctors. But BME doctors – whether from the UK or abroad – encounter ‘additional’ risks to their progression, including ‘difficulties fitting in, unconscious bias in assessments, recruitment and day-to-day working, and occasionally overt prejudice and greater chances of social isolation’. If that list isn’t long enough, some special risks faced by international doctors were also highlighted, including ‘difficulties forming relationships because of cultural differences and lack of experience of the NHS’.

  And the study continued: ‘More risks could reduce confidence and motivation, which could make these doctors less enjoyable to teach, making them receive poorer quality teaching reducing still further their confidence and motivation.’

  In other words, accumulated risks produce a potent vicious cycle.

  *

  Khalid was a surgical trainee who grew up in Saudi Arabia and attended medical school in London as an international student. Tall, traditionally dressed and with a long beard, it was apparent from his appearance that he was a devout Muslim. Although he had an excellent academic record – prizes at medical school, cleared his postgraduate surgical examinations at first attempt and had papers published in prestigious journals – he failed to secure a surgical training position in London, and ended up accepting a place in the west of England. Not only was he separated from his wife and children, who remained in London – he also knew nobody in the town where he ended up working and living.

  On the first day the senior surgeon told Khalid that he didn’t want him in the operating theatre because his beard was ‘unhygienic’. Surgeons with beards can wear beard hoods in just the same way that they can cover head hair with surgical caps. There is no evidence that having hair on one’s chin poses any greater threat to patients than having hair on one’s head; nobody is suggesting that surgeons need to be bald. Masquerading as infection control this senior surgeon’s claim had no basis in scientific fact. Khalid successfully argued his case – but things got off to a terrible start.

  On another occasion he was huddled around the computer with a group of colleagues, looking at a patient’s X-ray. Apropos of nothing, one of the consultants turned to Khalid and said, ‘You people blow people up – you’re terrorists.’ Nobody said anything, and Khalid felt too frightened to speak out. He didn’t want to rock the boat, and find himsel
f labelled as a troublemaker – let alone a supporter of terrorists.

  Having been made to feel so unwelcome, Khalid became resentful as well as scared that he would end up with bad reports from this rotation. Desperate to win over his consultants, his solution was to work even harder. But rather than seeing him as a committed surgical trainee, his consultants disliked Khalid’s lack of small talk and ostracised him more. Khalid’s future in surgery had become risky.

  I also struggled when Khalid came to see me. During the first part of the session he talked at me in such a direct and continuous manner that I was left feeling that I couldn’t think straight. After about twenty minutes I realised that his manner stemmed from desperation; he was terrified his surgical ambitions could unravel. But just as somebody struggling in the water makes their situation worse by thrashing around, Khalid’s frantic attempts to overcome his difficulties were increasing the risk that he would drown.

  I had a hunch that the impact he had on me was repeated on a regular basis with his colleagues. So I decided to seize the moment, and talk about it. Choosing my words carefully, I began by acknowledging the racism and emphasised that anybody on the receiving end of such hatred would feel distress. I also told him that whilst I respected his decision not to make a formal complaint, if at any point he changed his mind, I would help find people who could support him through the process. But I went further; I told Khalid that I thought his response (to put his head down and work even harder), whilst understandable, was leading him to be misread by his colleagues. I didn’t doubt his commitment to his patients, but I suspected that his colleagues saw him as somebody who only cared about his career. Khalid listened intently to what I was saying and we then discussed ways in which he could better convey to his colleagues that he wanted to be a surgeon because he wanted to treat patients – rather than simply to advance his ambitions.