Richard Lehman’s journal review—18 May 2015

richard_lehmanNEJM 14 May 2015 Vol 372
1887 Something strange seems to be going on in the New England Journal of Medicine. This week it publishes two trials of gene therapy for Leber’s congenital amaurosis, one at the start and the other at the end of its research section. Neither of them achieved more than fleeting benefit in one genetic variety of an uncommon form of blindness, which can be caused by mutations of at least 19 different genes. It takes a long editorial to explain to generalist readers what this is all about and why it may never work. I am all for publishing negative results, but I’m not sure most of us really needed to know this right now.

1877 Something strange seems to be going on in the New England Journal of Medicine. This week it publishes two open access pieces about the pharmaceutical industry. One is a closely argued piece by Jerry Avorn, questioning the old chestnut claim that it costs pharmaceutical companies $2.6 billion to develop a new drug. It doesn’t, and the article also points out that a lot of the preliminary development is often done at public expense. The second article is a continuation of Lisa Rosenbaum’s series about conflicts of interest and the pharma industry. It’s a stream of consciousness narrative in which she struggles to persuade us that all this talk of bias harming patients is a naughty lie and that we should demand more evidence. It’s quite sweet but I’m not sure what it is doing in a leading medical journal.

1937 For a Briton to refer to “appendectomy” rather than appendicectomy is still considered bad form. In my youth, “Americanisms” were regarded as almost unpardonable, but these days I lose track of what is and isn’t considered one. It also used to be considered Americanistic to question medical advice rather than doing as you were told. The narrative of this clinical review begins with a patient with appendicitis questioning whether this really mandates appendectomy or whether it wouldn’t be worth trying antibiotics first and converting to operation if needed. If looking things up on the web and discussing options with health professionals is becoming part of the American Way, then the sooner it crosses the Atlantic the better.

JAMA 12 May 2015 Vol 313
This week’s JAMA is taken up with a discussion of professionalism. Now this really does seem to mean something different on the two sides of the Atlantic. In the UK, there is a long and internalised ban on self-advertising, which means that anything that smacks remotely of the congratulatory is frowned upon. In American bookshops, by contrast, there is usually a whole shelf or section for books written by doctors who believe, very humbly of course, that they have discovered the secrets of empathy and many other things so wonderful that they simply have to tell the world about them. In England these narratives would have to be so heavily disguised and ironicized that nobody would get past the first page. Personally, I would rather it remained so.

Anyway, I am currently on a short break in Wales to help me become even better connected with the Universe in ways that I am too modest to share with you. So I decided to skip all these JAMA pieces, which are in fact all about the US context, except the one on governance and professionalism in UK medicine. It’s by Harvey Marcovitch, a rare treat from a superb writer, and you can access it for free.

JAMA Intern Med May 2015
OL A team of Spanish investigators report on cognitive test results from the Prevención con Dieta Mediterránea nutrition intervention trial, which ran from 1 October 2003, through 31 December 2009. It seems to show that over a four year period, older Catalans who were randomised to eat a “Mediterranean diet” supplemented with olive oil or nuts showed better preservation of cognitive function than those who were merely advised to reduce fat intake. Lisa Rosenbaum might well ask me why I believe their results are not biased by the fact that Spain is the world’s largest producer of olive oil. Why am I not saying the sample size was too small, the cognitive scores are of doubtful significance, the mechanistic explanation (the antioxidant hypothesis) is largely discredited, and that it is impossible to know what free living individuals actually had for their meals over a six year period? Actually, I am quite ready to say all these things. I don’t really think there could have been a financial bias in this trial, but there was almost certainly a strong bias towards the Mediterranean diet. And I don’t think this trial and a collection of observational evidence are really adequate to prove anything conclusively. We eat for need and pleasure, and if what we eat keeps us alive and thinking clearly for longer, well and good. The baseline traditional daily diet around the northern Mediterranean for the last 3000 years, by the way, was a loaf of bread, a litre of wine, some olives and cheese, and on lucky days a fish or a scrap of grilled meat.

OL Men have a rather uneasy relationship with testosterone. Apart from its effect on libido—which probably isn’t straightforward—there is a swirl of urban myth about its effects on energy, mood, and muscle bulk. And then there is the worry that too much increases cardiovascular disease and prostate cancer. The boundaries of the “normal” range are fuzzy, and all in all I think I can say that I’ve never prescribed it to anybody without a strong feeling that I didn’t fully know what I was doing. This is borne out by an impressively huge observational study of outcomes in over half a million men prescribed testosterone by different delivery methods in the US and UK. It turns out that testosterone injections were associated with a greater risk of cardiovascular events, hospitalizations, and deaths compared with gels, and that gels and patches have the same level of risk.

Lancet 16 May 2015 Vol 385
OL Here’s a gripping illustration of the problems of big epidemiology. “The Prospective Urban-Rural Epidemiology (PURE) study is a prospective cohort study of more than 154 000 community based individuals from 17 high income, middle income, and low income countries. The aim of the PURE study was to examine the association between societal factors, risk factors, and chronic non-communicable diseases across various sociocultural and economic environments.” As you will probably have read in the newspapers, the “study suggests that measurement of grip strength is a simple, inexpensive risk-stratifying method for all-cause death, cardiovascular death, and cardiovascular disease. Further research is needed to identify determinants of muscular strength and to test whether improvement in strength reduces mortality and cardiovascular disease.” Hang on, get a grip on yourselves, dear epidemiologists. By bunging a massive amount of data into your software you have come up with a plausible association between having weak hands and dying a bit sooner. Are you seriously expecting us to test the hypothesis that strengthening peoples’ hands will help them live longer?

OL The indefatigable Martin McKee and his team have come up with an 89 nation survey of how tax systems best support health services in low and middle income countries. Using cross-national longitudinal fixed effects models, they conclude that, “Increasing domestic tax revenues is integral to achieving universal health coverage, particularly in countries with low tax bases. Pro-poor taxes on profits and capital gains seem to support expanding health coverage without the adverse associations with health outcomes observed for higher consumption taxes. Progressive tax policies within a pro-poor framework might accelerate progress toward achieving major international health goals.” Somehow, I think Aneurin Bevan would have agreed and put it more eloquently.

The BMJ 16 May 2015 Vol 350
We owe a lot to the Nordic nations, whose great disease registers give us insights into the outcomes of 26 million people. As a result, we have a wealth of epidemiology relating to people who are mostly descended from the tribes mentioned in Beowulf—Geats, War-Danes, and so forth. As a devotee of the poem, once able to declaim large chunks of it in mead halls throughout the land, I find that rather thrilling. But it has to be said that other than Grendel’s Mum, there is very little feminine interest in the story. Nordic women of today sometimes conceive their bairns while taking antidepressants, which are usually serotonin reuptake inhibitors or venlafaxine. Fortunately, these carry no definable risk of birth defects, which is just as well since they can be very difficult to discontinue in a hurry.

A study of maternal outcomes in pregnant women taking antipsychotic drugs in Ontario is also reassuring. Compared with matched controls, they did not show higher rates of thromboembolism, gestational diabetes, premature delivery, or hypertension.

We travel back across the North Sea for a somewhat convoluted study based on Dutch data, which attempts to predict the direct benefit of vaccinating boys along with girls against oncogenic human papillomavirus, using bayesian evidence synthesis. I think the main point they are trying to make is that doing this will benefit some boys directly if they have sex with other men later on, as this vaccine should prevent anal cancer.

I can’t forbear to mention this week’s analysis piece, which launches the Choosing Wisely initiative of the Academy of Medical Royal Colleges. I am a co-author who came in quite late in a process which Aseem Malhotra had patiently championed for over a year. The idea of generating bin-lists of futile treatments is not new, but the idea of making this an open process involving patients at every point is game changing. I can’t wait for phase 2, which seeks to make sharing decisions with patients a reality rather than an aspirational slogan. I’m for a slow and humble approach, acknowledging the best in the clinical dialogue which already takes place, and gradually working out how we can improve on it. Should Choosing Wisely ever turn into a top-down rationing or tick boxing exercise, I would exit immediately, screaming. We all need to join in and make sure this cannot happen. We must work to make it the biggest move towards patient centred care in the UK. Get in touch with your Royal College and ask what they are doing and how you can help. If you’re a GP, join the terrific overdiagnosis group that Margaret McCartney and Julian Treadwell set up last year.

For many of us in the early 1990s, the appearance of the second edition of Clinical Epidemiology, a Science for Clinical Practice by Sackett et al proved life changing. Much of it was to do with the analytical skills required for the rational use of diagnostic tests, with exercise testing for coronary disease as a prime example. In the week that the great Dave Sackett has died (read a wonderful obituary by Richard Smith), an Uncertainties article revisits the topic. Kevin Barraclough is one of few people alive who can match Sackett’s analytical prowess and communicative ability, and here is an update which our departed hero would have admired and learnt from.

Plant of the Week: Crinodendron hookerianum
The Chilean lantern tree is a stunningly beautiful plant, which flourishes in dark moist places with little frost. Just now I’m staying in a house by the Menai Straits that provides an ideal habitat for it, so much so that it has had to be cut back to prevent it overwhelming a balcony which gives views over the great bridges.

Should you be blessed with a similar garden in some mild wet spot, try to find a space for this dark evergreen wonder with its hanging fleshy flowers of dusky red. It is one of the most exotic and showy plants that can be grown in Britain: your own little portion of Chilean rain forest.