Richard Lehman’s journal review – 5 March 2012

Richard LehmanNEJM  1 Mar 2012  Vol 366
777   There is no JAMA this week, and the best things in the New England Journal come right at the start. Whether you are a British GP contemplating another set of humiliating idiotic directives and the imminent destruction of the NHS, or an American physician wondering how your crazy health system can ever be turned into something rational and sustainable, or an academic wondering how much more futile research you have to grind your way through and pretend to be interested in, here is the boost you need. Goal-Oriented Patient Care — An Alternative Health Outcomes Paradigm. For the battle-weary GP, an affirmation of the central value of what you came into medicine to do – to help patients to achieve goals that they choose for themselves, in a continuing dialogue about the possible and the practical. For the American system, a model of care that puts the patient at the centre and emphasizes joined-up, affordable care. For the academic, a new research agenda based on what will best inform shared decision-making with patients and society. Download this article by David Reuben and Mary Tinetti at once and keep it under your pillow.

The NEJM lets you have this free, and also the companion pieces which follow – about shared decision-making and defining patient-centred care. Both good, but the Reuben/Tinetti is a gem.

787   Now back to the bizarre world of the medical-industrial complex and the way that it generates “evidence,” and large amounts of money, with the help of medical journals. Ruxolitinib is an inhibitor of Janus kinase (JAK) 1 and 2, and it costs about $85,000 for a year’s treatment. Here are two new trials of ruxolitinib in myelofibrosis, showing some symptomatic benefit, but no survival benefit compared with best available therapy. Myelofibrosis can’t be a particularly rare condition, since I have encountered it several times among the elderly patients of the little flock I once used to shepherd. Once you have it, you will probably die as a result: which is a pity, but then we all have to die of something in old age. So symptomatic benefit is a laudable goal. But how do you go about setting that against costs of this order? The editorial explains a bit more about the issues at stake, but points out that these studies add little to what an NEJM long-term study had shown a few months ago. It concludes: “Approximately 30% of patients with myelofibrosis present with ruxolitinib-sensitive symptoms, and the drug might be useful in a fraction of these patients who are not candidates for allogeneic stem-cell therapy or for clinical trials of potentially better drugs, including newer and more selective JAK inhibitors.” So what are these papers doing taking up prime-time space in the world’s leading medical journal?

808   Here’s a UK trial of nicotine replacement patches to reduce smoking in pregnancy. If this were a pharma-funded trial hoping to sell nicotine patches, we might expect it to have been conducted in 76 centres in 15 countries, but instead this government-funded study recruited from 7 antenatal clinics in the West Midlands, and still managed to be 4 times larger than any previous trial. Unfortunately, though, it showed the same result as all the others: that nicotine replacement in pregnancy has no effect on overall cessation rates.

Lancet  3 Mar 2012  Vol 379
815   The Lancet is back on one of its frequent excursions into China: small recompense from our clapped-out little island which once tried to destroy Chinese civilization by forcing its population to become opium-dependent and then strangling its trade. Alas, the imperialism of bad ideas continues. Western notions of chronic kidney disease have now bestowed silent ill-health on 120 million Chinese people, whose creatinine clearance is dangerously below 60mL/min per 1.73m2, or who have albumin in a single sample. What is to be done? Absolutely nothing, of course. The Chinese health system has far better things to concern itself with, as the preceding article on health coverage and catastrophic expenditure makes clear.

823   Another bad Western idea is that adverse prognostic markers are somehow valuable. If you happen to have had surgery for non-squamous, non-small-cell lung cancer, you can now undergo a practical molecular assay to predict survival, thanks to the efforts of a Chinese team. When pressed, researchers who uncover these harbingers of doom generally claim that they identify a subgroup of patients who deserve special further investigation, or closer clinical follow-up. But in fact all that this test will tell you is whether are likely to snuff it, which is rarely a comforting or useful thing to know.

833   “The pace of reform should be moderated to allow service providers to develop absorptive capacity. Independent, outcome-based monitoring and evaluation by a third-party are essential for mid-course correction of the plans and to make officials and providers accountable.” This is the conclusion of an examination of China’s huge and complex healthcare reforms. I suggest that we should send Andrew Lansley and David Cameron on a fact-finding mission to China. Three years should do it.

BMJ  3 Mar 2012  Vol 344
Here’s a fact we would rather not face: we cannot really know the safety of devices or drugs over long periods of time unless we test them over similar periods of time. It is certainly not a fact that drugs and devices manufacturers want to face, and more worryingly it is something that regulatory agencies are very poor at addressing on our behalf (see the account of European and US device regulation in NEJM). The bar for introducing new devices is set very low, and once the device is over the bar, post-marketing surveillance is often nugatory. If you think this doesn’t concern you, then you must be somebody who has never referred a patient for hip replacement. Metal-on-metal hip prostheses have become hugely popular in the last two decades, and metal resurfacing of the hip is a great British invention in the best Barnes Wallis tradition, perfected by an orthopaedic surgeon rolling his sleeves up and tinkering with cobalt and chromium in his garden shed. Deb Cohen’s piece tells it all. So do we really want to face the fact that these devices spread metal ions and particulate matter all over the body, with some obvious harms and others we do not know? Should we insist on a 15-year testing period for all permanent implantable devices? Or should we ignore such uncomfortable facts, and just carry on with a system which supports the great medical devices industry and harms patients?

Acute cannabis consumption and motor vehicle collision risk. A meta-analysis, pointing out that you should not drive if you are stoned, just in case you kill somebody.

It was nice of the BMJ to publish this paper led by two physiotherapy PhD students from Sweden, but it would have been better to wait until they had some useful data. It was a 12-week study examining subacromial impingement syndrome. The right kind of exercise helps to avoid surgery during this time period. Aha. And then?

“Profiles for psychological, vasomotor, and sexual discomfort symptoms relative to age at menopause could help health professionals to tailor their advice for women with natural menopause” claim the authors of this study. I’m not entirely sure what the advice might be. Generally it is better to move up social classes as much as possible, and also to get a degree, but only if you do this in your 20s rather than your 50s. Don’t marry if you want to avoid late menopausal symptoms. But nobody has done a randomised controlled trial of divorce for symptom control (blinding would be only one of many problems). In all, the main value of this study is to characterize a number of discrete patterns in menopausal symptoms, rather than identify any modifiable factors.

The Uncertainties Page this week has a nice account of the evidence around the use of probiotics to prevent antibiotic-associated diarrhoea. If your patient has ever had diarrhoea associated with antibiotics before, probiotics are certainly worth advising. It may be going too far to recommend their routine use by all healthy individuals taking antibiotics, but then Greek yoghourt is nice to eat anyway.

Arch Intern Med  27 Feb 2012  Vol 172
312   Doctors and patients love to share simple mechanistic explanations of diseases and cures. Angina pectoris is caused by blocked pipes: unblock the pipes, preferably leaving behind something to keep them open, and you have cured the problem; whereas if you just keep on pushing tablets, you’re just treating the symptoms without dealing with the cause. Most of us shared this belief with cardiologists and with patients, until the COURAGE trial came along in 2007 to prove otherwise (and the trial was nearly never done because many interventional cardiologists considered it unethical). This meta-analysis looks at this and 7 other trials and confirms the counter-intuitive conclusion that “Initial stent implantation for stable CAD shows no evidence of benefit compared with initial medical therapy for prevention of death, nonfatal MI, unplanned revascularization, or angina.” Collusion in mechanistic certainty is one of the great barriers to progress in medicine, and it may cause direct harm to patients, as the next paper illustrates.

There isn’t much else of note in this week’s printed Archives, but hidden in the Online First section awaiting paper publication is a remarkable study of patient satisfaction in relation to outcomes. This is worth spending some time on. In the USA, the highest quartile of patient satisfaction is associated with 9% higher  inpatient care and drug costs, and – wait for it – 26% higher mortality. What is killing America’s most satisfied patients? Is there some hidden confounding here (as the editorial hints)? The authors adjust for everything obvious, such as age and health status; so perhaps the kind of medicine these patients are getting really is 26% more dangerous. But this exceeds even the most pessimistic predictions of those of us who have long held that the overprovision that American patients are programmed to demand might be dangerous. Costly, futile, yes: but as lethal as this? I intend to spend an hour with my brilliant group of young overseas doctors at Yale picking over this paper. If nothing else, it will have some lessons on what to avoid in their own health systems. I firmly believe that all health provision should be judged by the experience of patients, but the simple criterion of fulfilled expectation is clearly inadequate, and may be positively dangerous in societies which ratchet up demand for medical services while ignoring costs and harms.

The same set of Online First papers includes a Research Letter from the UK which is relatively reassuring about the value of patient opinion, at any rate about hospital in-patient episodes. British patients have been encouraged to rate their experiences of all 166 acute NHS hospitals on the NHS Choices website. When compared with routinely gathered measures of clinical outcome and hospital-acquired infection, patient opinion showed a high degree of correlation with quality.

Plant of the Week: Hamamellis mollis

We can’t grow witch-hazels on the limy clay of our garden at home, but here in New Haven they thrive on the local acid sand. Many of the colleges have planted them in borders by the road, and they don’t seem to mind fairly high levels of motor exhaust. Passers-by in late winter are hit by a sudden wall of spicy scent, and become aware that these medium sized shrubs are covered in thin tassel-heads of dark yellow or (in other species) dull red. Older, larger specimens make a fine sight beneath the blue skies we sometimes enjoy here.

Hamamellis mollis is really a small tree, so you might do better to look for H japonica or the hybrid of the two, called H x intermedia, which comes in colours ranging from pale straw to dark brick. The yellow ones tend to have the most scent. Also, if you happen to have bleeding piles that require immediate attention, you can rush out and use the bark or leaves as an astringent.

  • mark aley

    re the stents and COURAGE:

    As I
    commenting hitherto on this- what is lost from the data is the potential effect of
    multiple RX’s on the individual in terms of side effects/ RX load, interactions etc etc: Maybe the stent
    is justified in this regard for an individual, though ‘benefit’ would be lost in
    trail data looking at outcomes.

  • mark aley

    re the wealthy and unhealthy amaericans…

    maybe it really is
    the medicine killing them off… we know CT scans increase cancer risk, we know
    lots of drugs do harmful things, we know if people pay for things they are likely
    to be given what they want, as opposed to what they need… Michael Jackson being
    a extreme example.