“No amount of careful writing can overcome careless reading.” Students, please note.

Clive James

This is from the US, but adds a different flavour to what I posted yesterday from Japan. I think it was Martin Wolf in the FT who pointed out that pharma runs the risk of being held in the same regard as many of us view the ‘banks’ and much of modern finance.

“The support by a consortium of major drug companies shows they’re clearly interested in obtaining the ability to do off-label marketing themselves, even if they haven’t been front and centre of the fight,” he [James Beck, attorney at Reed Smith] said.

The decision could have big implications for the pharma industry, because it paves the way for drug companies to push their medicines and treatments for a variety of illnesses without doing costly clinical trials to convince regulators that they work.

Instead, companies could choose to market their drugs for a wide range of illnesses, pointing to evidence with a much lower bar, such as the anecdotal experiences of patients and doctors.

FT

Nature:

Patients are willing to pay, and pay dearly: the HeartSheet treatment costs nearly ¥15 million (US$122,000). Last month, the health ministry added it to the procedures covered by national health insurance, which will help. But patients still pay 10–30% of the cost for a drug that is not known to be effective. As they do so, they basically subsidize the company’s clinical trial.

Japan has turned the drug-discovery model on its head. Usually, the investment — and thus the risk — is borne by drug companies, because they stand to gain in the long run. Now the risk is being outsourced. By the time it is clear whether a treatment works or not, the companies will have already made revenue from it.

“Besides inventing quantum theory, Planck had made another great contribution to science by welcoming and generously supporting the young Albert Einstein. In 1905, when Einstein, then an unknown employee of the Swiss patent office in Bern, sent five revolutionary papers to the physics journal that Planck edited in Berlin, Planck immediately recognized them as works of genius and published them quickly without sending them to referees. He did not agree with all of Einstein’s ideas, but he published all of them.”

Freeman Dyson in the NYRB. Real review by one’s peers, and not a checklist in sight.

From the land of choice, they may be less (NEJM). The issue: with expansion of med schools, there might not be enough residency slots ( a problem we are sort of seeing in the UK).

Potential effects on U.S. graduates have many members of the educational community worried, and this concern has been passed on to medical students through authoritative warnings that they may have trouble securing residency positions and that their choice of specialties will be severely curtailed. Congress’s unwillingness to legislate more Medicare GME funding is often cited as the reason for the perceived squeeze. The positions actually available and the trends reviewed here do not bear out this interpretation. The primary goal of public GME support, it should be noted, is to produce trained physicians to meet the country’s health care needs and not to fulfill the personal preferences of individual graduates for the specialties of their choice. Although the GME gap will narrow slowly, it appears likely that there will be ample positions for all U.S. graduates over the next decade. It would seem difficult to argue that Congress should fund more GME positions in order to create a larger margin for U.S. graduates. Disquiet among medical educators is understandable, but we believe that anxiety among students should not be amplified by well-meaning student advisors or national organizations.

I do not know enough of the economics of all of this, but I am puzzled as to why so much subsidy of ‘training’ is required. I find it hard to imagine that there are easy alternatives to what many junior docs do — even when you factor in the part of their jobs when they are learning or been explicitly surplus to service. I just wonder if this is rigging of the market by the medical certification lobby and, in the UK at least, the NHS monopoly of training and certification. Back of the envelope calculations would made me suspect that our (dermatology) trainees are grossly underpaid (compare what they deliver in the NHS, with their superior skills, to what some medics attract in the badlands of some private highstreet  dermatology).

‘Now Weinberg has added another credential to his crowded vita: historian of science. In his past writings, he had mainly concerned himself with the modern era of physics and astronomy, from the late nineteenth century to the present—a time, he says, when “the goals and standards of physical science have not materially changed.” Yet to appreciate how those goals and standards took shape, he realized he would have to dig deeper into the history of science. So, “as is natural for an academic,” he volunteered to teach a course on the subject—in this case, to undergraduates with no special background in science or mathematics. Then he immersed himself in the primary and secondary literature. The result is To Explain the World, which takes us all the way from the first glimmerings of science in ancient Greece, through the medieval world, both Christian and Islamic, and down to the Newtonian revolution and beyond.’

In a review, by Jim Holt,of ‘To Explain the World: The Discovery of Modern Science’, by Steven Weinberg.

The problem is that this is no longer natural or even encouraged of an academic. And if the writings of Weinberg I have read are anything to go by, this course must have been something special. I can remember the late John Ziman telling me that having been appointed to a lectureship in physics at Cambridge, he realised that there was no suitable text for his Cambridge undergraduates in the area that interested him. So, he spent two years writing such a text (which sold well for many years, he added). He observed that no longer would a UK university consider some behaviour appropriate: what about the REF! This tells us something about great thinkers, deep domain expertise, and how explanatory ability is the crux of great teaching. And about universities, and their troubled relation with teaching — and academics.

Awhile back I was chatting to a student. We had met say half a dozen times over a few years, and we gossiped widely. He was now returning back overseas, setting out on a surgical career after deciding that the UK postgraduate system wasn’t as good as he could get elsewhere. As we parted, he thanked me for the various conversations, muttering that it was good to hear how things had been in the past. I might have taken it as putdown — like when my children were younger and any conversation about ‘olden days’, meaning anything from the past thousand years was viewed as something I had firsthand experience of. The more interesting points are: not all change is good (here is an amateur geneticist speaking), and knowledge of how things once worked is not irrelevant to any future; and second, that often in medicine our thoughts about undergraduate medicine are deeply influenced by our own experience. The problem for me is that this experience was close to 35-40 years ago.

Listening to students is hard. I don’t mean passive listening, but when you try and interrogate students about how they see things, trying to momentarily forget how you see things. And of course, why they see things in a particular way. Listening this way is an act of imagination, which is why it is tough, but worth doing.
One conversation I filed away was about one facet of lectures (and I am not going to digress into the ‘what is the value of lectures’ as I think it is a misinformed trope). A group of students were saying how tedious they found many of them, but when I said ‘why go’ (because I ‘never’ did) they pointed out the main purpose of lectures was to provide information on what might come up in the exam. They pointed out some potent examples of where trivial information was presented in a lecture, and that same factoid came up in an exam (and some of the stories they recall are disturbing: what chromosome is this gene on, for example). So, this got me thinking about how students work out what it is they are expected to know. And my conclusions are that it is at least worth considering that things are worse than they were once for me.

First, we need to set aside learning outcomes. Conceptually they make lots of sense, but as implemented operationally in medicine they are often dreadful (I have written before on this topic, so will not repeat myself). They are a giant pretense of apparent order.

Why do I think things may be harder than they were for me? Well, if you decrease the frequency of lectures, and lectures — however bad and inefficient for this purpose — serve as the index of what has to be learned, you are left with a problem. Then think about books. I do not have empirical data, but it seems to me there are more and more books on any  in undergraduate medicine, and if you compare different books the differences for a learner are not trivial. In any case, many books for dermatology that are said to be for students are aimed at multiple markets. It makes no sense to ‘learn the book’.

Then there is the web. So much there, so much good stuff, but lots of inaccuracy, and again not matching of content to learner’s stage. How do you treat recommendation about what is good, and the ‘search cost’ in finding that which is good, and the power of modern social media and communications to magnify and echo both what is good, and what is bad? You have to think about recommendation errors, errors in fact, and matching of content to your course. The latter point, matching to a course, is something we do not talk much about. You only have to visit two medical schools 50 miles apart and look at how curriculum time is allocated, to seriously question whether student’s can reasonably learn the same things. (Or perhaps consider the possibility that the real metacurriculum is the one the medical school publishes). Finally, the web is wonderfully anarchic, but format and approach to several different domains might benefit from some similarities in pedagogical approach — perhaps. And meaningful curation is not much in evidence.

And what about the teachers? Well, this is an easy one. There are more staff, more students, more compartments (as in specialties), and staff time is ever more fragmented. There may have never been much coherence to a medical school syllabus beyond thinking it was what you had yourself received 20-40 years earlier, but for many who teach medical students, the ‘course’ is an alien place situated somewhere between the GMC, the medical school, the NHS, and what interests — or bugs —the teacher. Education and scale are not easy bedfellows, even if all the people doing the teaching were professional educators.

In many things to do with teaching, to borrow some words from Larry Lessig (and Cass Sunstein), I am a libertarian paternalist. My take on this is that the oft quoted complaint about students — ‘they want us to tell them the exam questions!’— may reflect a genuine sense that students really know less about what they are supposed to know that I did all those years ago when the choices were not so legion.

Any real education is incapable of robust and widely accepted psychometric assessment that will satisfy professional regulators.

Another formulation: if you can reliably assess education within a standardised and regulated framework it is not real education.

Medicine: we do two things; educate and train. It may be possible to pull them apart, and we can assess and measure training simply (simply, in theory at least, if not at a certain cost in practice). Education is harder, and is something you can only meaningfully assess 10 or more years after graduation. Students will not be able to fully value and asses their education until a long time after graduation.

On glancing at Nature Genetics, a journal I used to read (and even publish in).

If two or more authors on a single publication have identical  Christian and surnames, you have left discovery behind.

Wonderful retrospective in the Economist on one of the intellectual triumphs of the twentieth century. A few quotes give the flavour of real advance without an impact statement in sight.

General relativity was presented to the Prussian Academy of Sciences over the course of four lectures in November 1915; it was published on December 2nd that year. The theory explained, to begin with, remarkably little, and unlike quantum theory, the only comparable revolution in 20th-century physics, it offered no insights into the issues that physicists of the time cared about most.

Some people might be satisfied just to let each theory be used for what it is good for and to worry no further. But people like that do not become theoretical physicists

A century ago general relativity answered no-one’s questions except its creator’s