Soviet Manchester

There's an exhibition of maps of Manchester at the John Rylands Library just now, and I'm hoping to have a look at it next week when I take a much-needed break. One of the exhibits in particular is attracting a lot of attention: a map of the city prepared by the Soviets, for use in the event of an invasion of the UK.

It shows the roads - familiar to many Mancunians - which the Soviets felt were wide enough to carry tanks including Washway Road, the Mancunian Way, and Princess Road.

I biked down Washway Road last night on my way home from a training ride, and it's quite hard to imagine columns of Soviet tanks clanking along it towards Stretford. Though I suppose if there ever had been a Soviet invasion, Washway Road would be even more of a post-nuclear wasteland than it currently appears. Apparently, this map is only 35 years old, and it's very strange to think that so recently there were plans for a Soviet Manchester.

Washway Road, yesterday

Society of Homeopaths defends ineffective treatment for condition that kills 2 million children a year...

...surprise!

I'm not going to write a huge amount about this, because it is ably covered elsewhere. But the World Health Organisation finally came out today, and clearly stated that homeopathy should not be used for life-threatening conditions such as AIDS, malaria, TB and childhood diarrhoea.

The Society of Homeopaths (SoH) respond by saying that treating AIDS, malaria and TB would contravene their ethical guidelines. These guidelines are a joke, as you can see by perusing the Quackometer and Gimpy's blog. But apart from that, they say:

The Society of Homeopaths, the UK’s largest body of registered homeopaths, is concerned to learn, in an online article by the BBC (“WHO warns against homeopathy use’), that the World Health Organisation (WHO) has issued caution against the use of homeopathy for childhood diarrhoea following a letter by the charity Sense About Science.

They go on to cherry-pick and distort the research that has actually been conducted on homeopathy for childhood diarrhoea.

Well, who's surprised by that? Regular readers will be aware of just a few of the ways that homeopaths try to distort the evidence that homeopathy doesn't work. The point here is that the SoH, a supposedly professional organisation, is trying to defend a totally ineffective treatment for a disease that kills 2 million children every year. Homeopathy can't do anything to help, and using it instead of proper medical treatment could cost lives. I don't really mind if homeopaths sell ineffective sugar pills to the worried well in Alderley Edge, but this press release is delusional and irresponsible, and reasonable people should have no truck with this sort of dangerous nonsense.

A long trip to a closed pub

Several years ago, I used to do a lot of time trialling. Given that, at my peak fitness, I weighed 8 and a half stone and looked like a bundle of pipe cleaners tied together, my favourite races were always hill climbs. Hill climbs are essentially short time-trials up a hill, and they usually happen at the end of the season in September and October. In a typical season I would ride several around Sheffield, Monsal Head and Curbar Edge being particular favourites. Then in 1999 I went to Canada to study for my PhD, and I didn't race again until this year.

This year I joined Stockport Clarion CC, and I've been riding the odd weekday 10-mile time-trial with no great success. But the event I was really looking forward to was the club hill climb championships. Hill climb courses are typically between several hundred yards and a couple of miles long, and they're usually steep, with gradients of 10-20%. Just to be different, ours is on the Cat and Fiddle road, between Macclesfield and the eponymous pub that stands, hemmed in by peat bog, at the road summit. That's about 6.5 miles of climbing, but at an average gradient of less than 4%. There are steeper railway bridges around here, but there's still about 335 m of height gain involved. The course starts in the outskirts of Macclesfield, opposite a bus stop, and the first 3 miles or so is a fairly steady 4.8%. The next mile and a half is very bendy, contains some short downhill sections, and is nearly flat on average. The final mile and a half takes you up onto Axe Edge Moor proper, climbing at 3.5% or so.

I had never ridden up the climb before, but I had a gameplan of sorts. The first section was the bit where my light weight would give me an advantage, so I would ride hard, but not flat out: the key is not to overdo it, what with there still being 3 miles to go. I would then use the fast section to take a bit of a breather, riding at slightly less than 10-mile pace, since you don't gain much time there anyway. Then it would be eyeballs out over the last mile and a half to the summit. What gears would I use? Hell if I knew. I would figure it out as I went along.

The race was last night, and I actually really enjoyed it. The conditions could scarcely have been better. It was clear and dry, about 18 C, and there was a moderate tailwind. There were only 8 riders, and I was off Number 2, with a slower rider starting one minute before me. I started well, settling in very quickly, and once I had emerged from the trees at the bottom of the climb I could see my minute man almost all of the time. I was clearly gaining on him, and that gave me a psychological boost, as I finally reeled him in on a sharp right-hand bend about a mile and a half in, just before Walker Barn. I was first on the road now, a nice feeling, and my legs were holding up just fine. After Walker Barn, into the fast part of the course, I switched into the big ring, and tried to maintain a slightly-slower-than-10-mile-time-trial pace. My biggest problem here was gear selection. The gradient changes so often and there are so many sharp curves that I felt like I was using nearly every gear on the bike, and I was changing front rings too often: I should have picked one and stuck to it. But I still felt good, and I knew I had something left for the last mile and a half. No strategy involved here; just eyeballs out until the summit. There was a guy out for a training ride in front, and that gave me something to chase. Round the last bend, and there was the welcome sight of the Cat and Fiddle Inn. One last leg-breaking effort, and I was past the timekeeper in 26 minutes dead, gasping like a freshly landed fish. It was about a minute faster than I'd hoped for, so I couldn't have been happier.

It was glorious at the summit, with views across the Cheshire Plain and the Peak District. Unfortunately, the pub is closed on Monday nights, so there was no chance of a swift half before the 25 mile ride back home. At least the first seven or so was downhill.

Publication bias: medical science is looking for it. Are you?

Via the mighty Ben Goldacre's miniblog, here's a study that's just been published in the BMJ on the subject of publication bias. Publication bias typically occurs when studies with positive results are preferentially published over studies with negative results. There are many reasons why this might happen; drug companies don't want to publish negative trials of their drugs for obvious reasons, but there is also a tendency for journals to be uninterested in publishing negative trials, and for busy authors to be reluctant to write them up. There's nothing sexy about a negative trial.

In the BMJ study, Santiago Moreno and colleagues look at anti-depressants. This is a good area to look at, because of the availability of data that was submitted to the Federal Drug Administration in the United States. Legal requirements enforce submission of ALL data to the FDA, so the authors consider the FDA dataset to be unbiased (although not necessarily complete). This unbiased dataset can then be compared to the data that is available in published journal articles.

The comparison is done with our old friend the funnel plot. This plots the standard error for a trial against the size of the effect that the trial found. The authors of the BMJ study added a new twist by contouring the funnel plot for statistical significance: at a glance it can be seen where studies fall in terms of statistical significance.

The results are dramatic. Of the 74 trials registered with the FDA, 23 were not published. In the FDA data, there is a wide spread of results across the contours marking 1%, 5% and 10% levels of significance. When only the published data are considered, there is a clear cut-off at the contour for the 5% significance level, which is typically used in clinical trials to establish statistical significance. That strongly suggests that publication bias is a serious problem in the set of published trials: trials with statistically non-significant results have been systematically excluded.

What this means is that the published literature over-estimates the benefit of anti-depressants. It doesn't show that anti-depressants don't work: meta-analysis of the FDA data still shows a beneficial effect. The point is that the real benefit (as shown by the FDA data) is less than the benefit you would expect if you looked only at the published literature. Anti-depressants work less well than you might think, but they still work.

The problem is, of course, that most of the time we only have the published literature to work with. So the BMJ paper adds a useful visual technique for identifying publication bias as a likely problem, even if we don't have access a bias-free dataset for comparison. There is no doubt that biases exist in published data; the response from medical science, as seen here, is to try to identify and account for these biases. Regular readers will know that the response of CAM research is to manipulate the data in order to pretend that the problem doesn't exist.

I have a bad feeling about this...

I was idly browsing the front page of the Guardian's website, when I really ought to be doing increasingly tedious corrections to a paper, and I saw this. Eczema sufferer Peter Beech, after many years of decreasingly effective steroid treatments, is trying homeopathy. Beech also claims to be a sceptic of alternative medicine.

First things first. Is there any evidence that homeopathy is effective for treatment of eczema? A quick search on Pubmed showed this to be the most recent (2003) relevant review. Incidentally, Pubmed is a freely accessible service, unlike many of the journal databases used by academics, and anyone with an internet connection can do the same thing as me and come up with the presently available evidence in a couple of minutes. Here's what the summary says:

Alternative methods are commonly used in patients with dermatologic diseases, with homeopathy being one of the most common. Homeopathy was developed by Samuel Hahnemann (1755–1843) and is based on the law of similars and the law of infinitesimals. It is a regulatory therapy where high dilutions of particular compounds are thought to induce a counterreaction in the organism. In dermatology, homeopathy is often used in atopic dermatitis, other forms of eczema, psoriasis, and many other conditions. To date, however, there is no convincing evidence for a therapeutic effect. There are only a few controlled trials, most of them with negative results. The few studies with positive results have not been reproduced. Acceptance by the patient seems largely based on counseling and emotional care rather than on objective responses to the homeopathic drugs.

Not particularly convincing, is it?

Beech is at least unlikely to do himself any harm by trying homeopathy. However, real harm can result when people use homeopathy and avoid real medicine, as this tragic case in Australia shows. Nine-month old Gloria Thomas died after her eczema allowed an infection to get out of control. Her parents had been treating her with homeopathy instead of real medicine: they were later convicted of manslaughter by gross criminal negligence.

But the main reason that I have a bad feeling about this, is that even if Beech gets better, it won't prove anything about homeopathy (and the same goes if he doesn't see any improvement, to be fair). This is essentially an uncontrolled case report, and there would be no way of showing that any improvement resulted from the homeopathy, rather than, say, the natural cyclicity of eczema, or the placebo effect. Whatever happens to Beech, it isn't going to trump the evidence from the most recent systematic review of the evidence from clinical trials. So Beech's experiment cannot add anything to what we laughingly call the "debate" over the efficacy of homeopathy.

Still, good luck to him. I hope he does get some relief from his symptoms, but if he does the chances of it having anything to do with homeopathy are vanishingly small.