Late night thoughts on medical education #7: Carousels

In the previous post I laid out some of the basic structures of the ‘clinical years’ of undergraduate medical degrees. In this post I want to delve a little deeper and highlight how things have gone wrong. I do not imagine it was ever wonderful, but it is certainly possible to argue that things have got a lot worse. I think things are indeed bad.

When I was a medical student in Newcastle in 1976-1982 the structure of the first two clinical years (years 3 and 4) were similar, whereas the final year (year 5) was distinct. The final year was made up of several long attachments — say ten weeks medicine and 10 weeks surgery — and there were no lectures or any demands on your time except that you effectively worked as an unpaid houseman, attached to a firm of two or three consultants. The apprenticeship system could work well during these attachments. The reasons for this partly reflected the fact that all parties had something to gain. Many if not most students chose where they did their attachments (‘if you like fellwalking, choose Carlisle etc), and had an eye on these units as a place to do your house jobs the following year. The consultants also had skin in the game. Instead of relying on interviews, or just exam results, they and all their staff (junior docs, nurses etc) got a chance to see close up what an individual student was like, and they could use this as a basis for appointing their houseperson the following year. If a houseman was away, you acted up, and got paid a small amount for this. At any time if you didn’t turn up, all hell would break out. You were essential to the functioning of the unit. No doubt there was some variation between units and centres, but this is how it was for me. So, for at least half of final year, you were on trial, immersed in learning by doing / learning on the job / workplace learning etc. All the right buzzwords were in place.

Carousels

As I have said, years 3 and 4 were different from final year, but similar to each other. The mornings would be spent on the ward and the afternoons — apart from Wednesdays — were for lectures. I didn’t like lectures (or at least those sort of lectures) so I skipped them apart from making sure that I collected any handouts which were provided on the first day (see some comments from Henry Miller on lectures below [1]).

The mornings were ‘on the wards’. Four year 3 students might be attached to two 30 bedded wards (one female, one male), and for most of the longer attachments you would be given a patient to go and see, starting at 9:30, breaking for coffee at 10:30 and returning for an hour or more in which one or more of you had to present you findings before visiting the bedside and being taught how to examine the patient. The number of students was small, and there was nowhere to hide, if you didn’t know anything.

For the longer attachments (10 weeks for each of paediatrics, medicine and surgery) this clinical exposure could work well. But the shorter attachments especially in year 4 were a problem, chiefly because you were not there long enough to get to know anybody.

The design problem was of course that the lectures were completely out of synchrony with the clinical attachments. You might be doing surgery in the morning, but listening to lectures on cardiology in the afternoon. Given my lack of love for lectures, I used the afternoons to read about patients I had seen in the morning, and to cover the subject of the afternoon lectures, by reading books.

I don’t want to pretend that all was well. It wasn’t. You might turn up to find that nobody was available to teach you, in which case we would retreat to the nurses canteen to eat the most bacon-rich bacon sandwiches I have ever had the pleasure of meeting (the women in the canteen thought all these young people needed building up with motherly love and food :-) ).

The knowledge of what you were supposed to learn was, to say the least, ‘informal’; at worst, anarchic. Some staff were amazingly helpful, but others — how shall I say — not so.

Year 5 envy

In reality, everybody knew that years 3 and 4 were pale imitations of year 5. The students wanted to be in year 5, because year 5 students — or at last most year 5 students — were useful. The problem was that the numbers (students and patients) and the staffing were not available. It was something to get through, but with occasional moments of hope and pleasure. Like going through security at airports: the holiday might be good, but you pay a price.

Present day

The easiest way to summarise what happens now is to provide a snapshot of teaching in my own subject at Edinburgh.

Year 4 (called year 5 now, but the penultimate year of undergraduate medicine) students spend two weeks in dermatology. Each group is made up of 12-15 students. At the beginning of a block of rotations lasting say 18 weeks in total, the students will have 2.5 hours of lectures on dermatology. During the two week dermatology rotation, most teaching will take place in the mornings. On the first morning the students have an orientation session, have to work in groups to answer some questions based on videos they have had to watch along with bespoke reading matter, and then there is an interactive ‘seminar’ going through some of the preparatory work in the videos and text material.

For the rest of the attachment students will attend a daily ‘teaching clinic’, in which they are taught on ‘index’ patients who attend the dermatology outpatients. These patients are selected from those attending the clinic and, if they agree, they pass through to the ‘teaching clinic’. The ‘teacher’ will be a consultant or registrar, and this person is there to teach — not to provide clinical care during this session.

Students will also sit in one ‘normal’ outpatient clinic as a ‘fly on the wall’, and attend one surgical session. At the end of the attachment, there is a quiz in which students attempt to answer questions in small groups of two or three. They also get an opportunity to provide oral feedback as well as anonymous written feedback. Our students rate dermatology highly in comparison with most other disciplines, and our NHS staff are motivated and like teaching.

The problems

When I read through the above it all sounds sort of reasonable, except that…

Students will pass though lots of these individual attachments. Some are four weeks long but many are only 1 or 2 weeks in duration. It is demanding to organise such timetables, and stressful for both students and staff

• each day a different staff member will teach the students
• it is unlikely that staff will know the names of most of the students. Students will usually not remember the name of the staff member who taught them in a previous week
• most teaching is delivered by non-university employed staff. Most of these staff have little detailed knowledge of what students are (now) expected to know. The majority will not be involved in any formal assessments, and reasonably view the teaching as a break from doing clinic after clinic.
• there is little opportunity to provide meaningful feedback on student performance, or to see student knowledge grow. Students find it easy to ‘hide’, and absenteeism is high and the rate of ‘illness’ seems higher than amongst hospital doctors.
• teaching the students plays second fiddle to service delivery. The terminology within NHS job plans is telling. When you see a patient it is called ‘direct clinical care (DCC)’. For maybe the remaining 10-20% of your time you have sessions allocated as ‘supporting professional activities (SPA)’. SPA time will include work relating to revalidation, CPD, hospital admin, teaching of registrars, and delivery of undergraduate teaching. Our overseas students pay in excess of 50K per year in fees, and each UK student attracts perhaps 50K from fees and government monies. Teaching undergraduates is merely a ‘supporting activity’ even when 50K is changing hands. Fettes or Winchester might be more careful with their terminology.

My critique is not concerned with the individuals, but the system. It is simply hard to believe that this whole edifice is coherent or designed in the students’ interest. It is, as Flexner described UK medical school teaching a century ago, wonderfully amateur. Pedagogically it makes little sense. Nor in all truthfulness is it enjoyable for many staff or many students. Each two weeks a new batch will arrive and groundhog days begins. Again. And again. And if you believe the figures bandied about for the cost of medical education, the value proposition seems poor. We could do better: we should do better.

[1] Lectures. Henry Miller, who was successively Dean of Medicine and Vice Chancellor at Newcastle described how…

“Afternoon lectures were often avoided in favour of the cinema. The medical school was conveniently placed for at least three large cinemas….in one particularly dull week of lectures we saw the Marx brothers in ‘A Day at the Races’ three times.”

Where’s the next frontier?

In a 1963 letter to molecular biologist Max Perutz, he wrote, “It is now widely realized that nearly all the ‘classical’ problems of molecular biology have either been solved or will be solved in the next decade…The future of molecular biology lies in the extension of research to other fields of biology, notably development and the nervous system.”

Sydney observed, and predicted, the flow of science: “Progress depends on the interplay of techniques, discoveries, and ideas, probably in that order of decreasing importance,” he said.

Man, the toolmaker. In this particularly case, a very special one.

Sydney Brenner (1927–2019) | Science [Obit of Sydney Brenner]

Direct URL for this post.

We will need even bigger prisons

Mr Kapoor’s co-defendants were Michael Gurry, Insys’s former vice-president of managed markets, Richard Simon, former national director of sales, and former regional sales directors Sunrise Lee and Joseph Rowan. Michael Babich, former chief executive of the company, and Alec Burklakoff, former vice-president of sales, had already pleaded guilty.

The defendants face up to 20 years in prison. Andrew Lelling, US attorney for Massachusetts, said it was “the first successful prosecution of top pharmaceutical executives for crimes related to the illicit marketing and prescribing of opioids”.

Insys founder convicted in opioid bribery case | Financial Times

Direct URL for this post.

A diagnosis not to miss: email apnea

A phenomenon that occurs when a person opens their email inbox to find many unread messages, inducing a “fight-or-flight” response that causes the person to stop breathing.

James Williams, ‘Stand Out of Our Light’

I wonder when this will be recognised as a bona fide occupational disease.

Direct URL for this post.

You need a wallet biopsy

“However, if a wallet biopsy – one of the procedures in which American hospitals specialise – discloses that the victims are uninsured, it transfers them to public institutions.”

In Paul Starr, ‘The Social Transformation of American Medicine’.

Direct URL for this post.

Late night thoughts on medical education #6: Structures

In the previous post in this series (Late night thoughts #5: Foundations) I wrote about the content or material of medical education, hinting at some of the foundational problems (pardon the meta). We have problems distinguishing between knowledge that is essential for some particular domain of medical practice, and knowledge that is genuinely foundational. The latter is largely speciality independent, less immediate than essential knowledge, and is rightly situated within the university. The expertise necessary to teach foundational knowledge lies within universities.

What I have not made explicit so far in this essay is also important. The best place to learn much essential knowledge is within the hospital, and during a genuine apprenticeship. There are various ways we can hone a meaningful definition or description of apprenticeship but key is that you are an employee, that you get paid, and you are useful to your employer. Our current structures do not meet any of these criteria.

How we got here

Kenneth Calman in the introduction to his book ‘Medical Education’ points out that medical education varies enormously between countries, and that there is little evidence showing the superiority of any particular form or system of organisation. It is one of the facts that encourages scepticism about any particular form, and furthermore — especially in the UK — leads to questioning about the exorbitant costs of medical education. It also provides some support for the aphorism that most medical students turn into excellent doctors despite the best attempts of their medical schools.

Across Europe there have been two main models of clinical training (I am referring to undergraduate medical student training, not graduate / junior doctor training). One model relies on large lectures with occasional clinical demonstrations, whereas the UK system — more particularly the traditional English system — relies on ‘ clerkships’ on the wards.

At Newcastle when I was a junior doctor we used to receive a handful of German medical students who studied with us for a year. They were astonished to find that the ‘real clinical material’ was available for them to learn from, with few barriers. They could go and see patients at any time, the patients were willing, and — key point— the clinical material was germane to what they wanted to learn. The shock of discovering this veritable sweetshop put some of our students to shame.

The English (and now UK) system reflects the original guiding influence of the teaching hospitals that were, as the name suggests, hospitals where teaching took place. These hospitals for good and bad were proud of their arms length relationship with the universities and medical schools. The signature pedagogy was the same as for junior doctors. These doctors were paid (poorly), were essential (the place collapsed if they were ill), and of course they were employees. Such doctors learned by doing, supplemented by private study using textbooks, or informal teaching provide locally within the hospital or via the ‘Colleges’ or other medical organisations. Whatever the fees, most learning was within a not-for-profit culture.

 Scale and specialisation

It was natural to imagine or pretend that what worked at the postgraduate level would work at the undergraduate level, too. After all, until the 1950s, medical education for most doctors ended at graduation where, as the phrase goes, a surgeon with his bag full of instruments ventured forth to the four corners of the world.

This system may have worked well at one stage, but I think it fair to say it has been failing for nearer a century than half a century. At present, it is not a system of education that should be accepted. There are two reasons for this.

First, medicine has (rightly) splintered into multiple domains of practice. Most of the advances we have seen over the last century in clinical medicine reflect specialisation, specialisation as a response to the growth of explicit knowledge, and the realisation that high level performance in any craft relies not solely on initial certification, but daily practice (as in the ‘practice of medicine’). Second, what might have worked well when students and teachers were members of one small community, fails within the modern environment. As one physician at Harvard / Mass General Hospital commented a few years back in the New England Journal of Medicine: things started to go awry when the staff and students no longer ate lunch together.

Unpicking the ‘how’ of what has happened (rather than the ‘why’ which is, I think obvious), I will leave to the next post. But here is a warning. I first came across the word meliorism in Peter Medawar’s writing. How could it not be so, I naively thought? But of course, historians or political scientists would lecture me otherwise. It is possible for human affairs to get worse, even when all the humans are ‘good’ or at least have good intentions. The dismal science sees reality even more clearly: we need to only rely on institutions that we have designed to work well — even with bad actors.

Biology is just messy

Some traits, such as adult height, are readily measured. The heritability of this trait is ∼60 to 80%. Attempts to characterize “height genes” have resulted in the identification of tens of thousands of genes, each of which contributes a small amount to this heritability. The plethora of factors is almost inevitable, given the vast number of cellular and physiological steps involved in the development of an adult human being. A model that accounts for ∼40% of height variability predicts individual heights to within 4 cm for 50% of people, but with errors of more than 10 cm for 5%. Thus, a sophisticated genomic analysis can predict height to some extent, but not well enough for use in ordering tailored clothing. Most direct-to-consumer genomic results are based on much less detailed analyses and many involve complex traits, so considerable skepticism is appropriate.

But such sensible comments, will not stem the hype — or the investors.

Consuming personal genomics | Science

Direct URL for this post.

It’s not all physics

This is why I have doubts about mechanical theories such as disruptive innovation. Too often, they’re presented as a type of physical law: You drop a glass of wine, it always falls to the ground with an acceleration of 32.17405 ft/s2. This truth is indisputable…but it ignores the drunken clumsiness of the oaf who knocked the glass over, and discounts the quick reflexes and imaginative solutions you only get when there’s a human nearby.

Jean-Louis Gassée. A nice summary of why human agency matters, and also why companies fail.

First Winning Wars, Only To Lose Them Later – Monday Note

Direct URL for this post.

Late night thoughts on medical education #5: Foundations

We sought out an examiner who would understand that anatomy was being taught as an educational subject and not simply for the practice of surgery. I thought I had found such a man in an old colleague. I listened while he asked the student to name the successive branches emerging from the abdominal aorta in a cadaver. When we got to the inferior mesenteric he asked what viscera were supplied by that vessel. The student gave a complete and correct answer but did not know the exact amount of the rectum supplied. The examiner asked me what I thought and I said that I thought he was very good, that the only question he had missed was the last one, which in my opinion, was trivial. No, said the anatomist, by no means trivial. You have to know that before you can excis the rectum safely.

My mind still boggles at the thought of a newly graduated doctor undertaking the total excision of the rectum on the faint remembrance of the anatomy he learned as a student.

George Pickering, “Quest for Excellence in Medical Education: A Personal Survey”

When I was a medical student I read this book by Sir George Pickering. It was published in 1978, and I suspect I read it soon after the Newcastle university library acquired it. Why I came across it I do not know, but at the time ‘new volumes’ were placed for a week or two on a shelf adjacent to the entrance, before being assigned their proper home (or ‘final resting place’). It was a way to find things you didn’t know you might enjoy. I liked this book greatly, and have returned it on many occasions. Parts of it are wonderfully dated (and charming), but it remains a wonderful young man’s book written by an old man. Now I am an old man, who read it first as a young man.

Roger Schank summarise the problems of education this way:

There are only two things wrong with the education system:

1. What we teach, and
2. How we teach it

George Pickering’s quote relates to ‘what we teach’ — or at least what we expect students to know — but in clinical medicine ‘what we teach’ and ‘how we teach’ are intimately bound together. This may be true for much  education, but the nature of clinical exposure and tuition in clinical medicine imposes a boundary on what options we can explore. The other limit is the nature of what we expect of graduates. People may think this is a given, but it is not. If you look worldwide, what roles a newly qualified doctor is asked to fill vary enormously (something I discovered when I worked in Vienna).

Here is another quote, this time from the philosopher, Ian Hacking, who has written widely on epistemology, the nature of causality and the basis of statistics (and much else).

Syphilis is signed by the market place where it is caught; the planet Mercury has signed the market place; the metal mercury, which bears the same name, is therefore the cure for syphilis.

Ian Hacking | The Emergence of Probability

Well, of course, this makes absolutely no sense to the modern mind. We simply do not accept the validity of the concept of entities being ‘signed‘ as a legitimate form of evidence. But no doubt medical students of the time would have been taught this stuff. Please note, those priests of Evidence Based Medicine (EBM), that doctors have always practiced Evidence Based Medicine, it is just that opinions on what constitutes evidence change. Hacking adds:

He [Paracelsus] had established medical practice for three centuries. And his colleagues carried on killing patients.

I am using these quotes to make two points. The first, is that there is content that is correct, relevant to some clinical practice and which medical students do not need to know. This may seem so obvious that it is not necessary to say it. But it is necessary to say it. Pickering’s example has lots of modern counterparts. We could say this knowledge is foundational for some medical practice, but foundational is a loaded term, although to be fair I do not know a better one. The problem with ‘foundational’ is that it is widely used by academic rent seekers and future employers. Students must know this, students ‘must’ know X,Y and Z. I once started to keep a list of such demands, but Excel spreadsheets have limits. You know the sort of thing: ethics, resilience, obesity, child abuse, climate change, oral health, team building, management, leadership, research, EBM, professionalism, heuristics and biases etc. Indeed, there is open season on the poor undergraduate, much of which we can lay blame for at the doors of the specialist societies and the General Medical Council (GMC).

My second point, stemming from the second quote, is to remind that much of what we teach or at least ask students to know is wrong. There is a feigned ignorance on this issue, as though people in the past were stupid, whereas we are smart. Yes, anatomy has not changed much, and I am not chucking out all the biochemistry, but pace Hacking, our understanding of the relation between ‘how doctors work’ and ‘what underpins that knowledge’ is opaque. We can — and do — tell lots of ‘just-so’ stories that we think explain clinical behaviour, that have little rational or experimental foundation. Clinicians often hold strong opinions on how they arrive at particularly decisions: there is a lot of data to suggest that whilst you can objectively demonstrate clinical expertise, clinicians often have little insight into how they actually arrive at the (correct) diagnosis (beyond dustbin concepts such as ‘pattern recognition’ or ‘clinical reasoning’).

What is foundational knowledge?

If you are a dermatologist, and you wish to excise a basal cell carcinoma (BCC, a common skin cancer) from the temple, you need to be aware of certain important anatomical structures (specifically the superficial temporal artery, and the temporal branch of the facial nerve). This knowledge is essential for clinical practice. It is simple to demonstrate this: ask any surgeon who operates in this area. Of course, if you are a lower GI surgeon, this knowledge may not be at your finger tips. Looked at the other way, this knowledge is in large part specialty specific (or at least necessary for a subset of all medical specialties). What happens if you damage these structures is important to know, but the level of explanation is not very deep (pardon the pun). If you cut any nerve, you may get a motor or sensory defect, and in this example, you may therefore get a failure in frontalis muscle action.

This knowledge is not foundational because it is local to certain areas of practice, and it does not form the basis or foundation of any higher level concepts (more on this below). The Pickering example, tells us about what a GI surgeon might need to know, but not the dermatologist. Their world views remains unrelated, although the I prefer the view of the latter. There is however another point. We should be very careful about asking medical students to know such things. So what do we expect of them?

Beyond essential

I find the example of anatomical knowledge as being essential compelling. But only in terms of particular domains of activity. Now, you may say you want students to know about ‘joints’ in general, and there may well be merit in this (Pickering, I suspect, thought so), but knowing the names of all the bones in the hand or foot is not essential for most doctors. If we move beyond ‘essential’ what is left?

At one time anatomy was both essential and foundational. And I am using the term foundational here to mean those concepts that underpin not just specialty specific medicine, but medicine in the round. A few examples may help.

Whatever branch of medicine you practice, it is hard to do so without some knowledge of pharmacology. How deep you venture , is subject to debate, but we do not think knowing the doses and the drug names in the BNF is the same as knowing some pharmacology. 

Another example. I would find it very hard to converse with a dermatologist colleague without a (somewhat) shared view of immunology or carcinogenesis. Every sentence we use to discuss a patient, will refer and make use of concepts that we use to argue and cast light on clinical decisions. If you want to explain to a patient with a squamous cell carcinoma (SCC) who has had an organ transplant why they are at such increased risk of tumours, it is simply not possible to have a meaningful conversation without immunology or carcinogenesis (and in turn, genetics, virology, and histopathology). And for brevity, I am putting to one side, other key domains such as behaviour and behaviour modification, ethics, economics and statistics etc.

To return to my simple anatomical example of the excision of the BCC. The local anatomy is essential knowledge, but it is not foundational. What is foundational is knowing what might happen if you cut any nerve.

Sequencing of learning

Let me try and put the above in the context of how we might think about medical education and medical training.

Foundational knowledge is specialty (and hence career) independent. Its function is to provide the conceptual framework that underpin much clinical practice. This not to say that the exact mix of such knowledge applies to all clinical domains, but we might expect most of it to be familiar to most doctors. But none of it will, years later, have the same day-to-day immediacy of ‘essential knowledge’ — think of my example of the temporal branch of the facial nerve for the dermatologist excising facial tumours on a weekly basis.

In this formulation, the core purpose of undergraduate medical education is to educate students in such knowledge. The purpose is not therefore to produce doctors at graduation who are ‘just not very good doctors’ but graduates who are able to pursue specialty training and make sense of the clinical world around them. The job of a medical school is to produce graduates who can start clinical training in an area of their choice. They are now in a position to — literally — understand the language of the practising doctors that surround them. They are not mini-doctors, but graduates, embarking on a professional career.

By contrast most specialty knowledge is not foundational, but essential for those within that specialty — not medical students. If you learn dermatology, you might come across things that help you learn respiratory medicine or cardiology but to be blunt, not very often. Specialties are not foundational domains of knowledge. You do not need to know dermatology to understand cardiology or vice versa.

Place of learning

The best place to learn the ‘foundations’ are universities. Anatomy, again may be an exception, but if you want to learn immunology, genetics, statistics or psychology you have, I think, no alternative. Hospitals simply cannot provide this.

On the other hand, using Seymour Papert’s metaphor, if you to want learn French you should go to Frenchland, if you want to learn maths, you should go to Mathland and if you want to learn doctoring, you need to go to doctorland. Medical schools are not the place to learn how to find you way around doctorland — how could they be?

NB: I will use the epithet TIJABP, but as subsequent posts will confirm, I am serious.

WD (Bill) Hamilton

Scope for recognizing and accommodating exceptional individuals has been diminishing in British universities ever since. Hamilton published relatively few papers, in generally low status journals, and gained only a handful of grants much later in life. Bureaucratic measures of performance are increasingly important and judge the impact of an article only by the journal it is published in. This seriously undervalues radical originality, which although extremely rare is utterly vital to science. It is disturbing that a young Bill Hamilton today would probably find an academic career even more difficult to pursue.

Alan Grafen, in his obituary of Bill Hamilton (Biogr. Mems Fell. R. Soc. Lond. 50, 109–132 (2004)).

I post this excerpt following a discussion with somebody who had never heard of him. Hamilton’s enormous contributions to biology are not well known. You also have to wonder if the lack of a Nobel for biology diminishes medicine in the long run. Some things do indeed get worse.

Direct URL for this post.