Monday, May 18, 2009

Honey, I shrunk the cold (and the evidence)

I love honey.

In fact, I love it so much, I became a beekeeper (which, even when things go wrong and honey levels are low, is still an amazing, rewarding and worthwhile pursuit). I met the Honey Monster once, but it turned out to be an actor in a honey monster outfit. Gutted.

Nonetheless, honey ticks all the boxes for being hijacked by quacks as a cure-all - it is natural, available and has historical & religious provenance;
And thy LORD taught the bee to build its cells in hills, on tree and in men’s habitations, then to eat of all the produce of the earth and find with skill the spacious paths of its LORD, there issues from within their bodies a drink of varying colors, wherein is healing for men, verily in this is a sign for those who give thought
(Qu'ran Surat Al-Nahl, The Bees, Aya 69)
Some time later, when [Sampson] went back to marry her, he turned aside to look at the lion's carcass. In it was a swarm of bees and some honey, which he scooped out with his hands and ate as he went along. When he rejoined his parents, he gave them some, and they too ate it. But he did not tell them that he had taken the honey from the lion's carcass.
(Bible, Judges 14, NIV)
Leaving aside the bee-zarre story of Sampson (he murders 30 people and gives his wife to his best mate a week after the wedding), honey has been used for squillions of years as an antiseptic - most micro-organisms are unable to survive in such a sugar-rich, water-poor environment. Hence, it never goes mouldy in your cupboard.

(One important exception is clostridium botulinum whose spores can survive in honey, meaning that it shouldn't be given to infants, although there have been no cases in UK since 2001. Of course, there will be people telling you that feeding honey to a 6 month old baby is fine, such as Unhindered Living, but then they are also anti-vaccination so I guess the stupid goes with the territory.)

So history, religion, antiseptic qualities and naturalness are all part of the honey story. (Naturalness may not be a real word). Throw into the mix the intriguing biology and evolution of bees (matriarchal society, highly beneficial to flowering plants, pheromone behaviour, relaying of information through the medium of dance, to name a few) and you have a perfect background for flogging honey as a natural panacea. (Although let's not forget the tragic case of Russell Jenkins who died after trying unsuccessfully to treat gangrene with honey).

Obviously, not any old honey will work to cure every disease in the world ever, it has to be special expensive branded honey, like Life Mel (still no follow up trial?) or Manuka Honey.

A story promoting the health benefits of honey is always going to make certain sections of the media. The fawning, uncritical, regurgitating cut-and-pasters at the Daily Mail will always print your honey story - no questions asked. It'll cure cancer, kill superbugs, fight heart disease and counter ageing.

The most recent story to be CtrlC/CtrlV-ed into the People's Journal of Science (Daily Mail) is that
Eating honey shortens colds by two days

The first line of copy tempers this a bit (naughty attention-grabbing sub-ed - so out of character) by saying it 'can help' reduce the duration of colds. 'Can help' is one of those weaselly marketing phrases like 'may', 'reduce' and 'up to' which allows them to infer much more than they *actually* can. Still, it's possibly worth further investigation. Like all dead-tree media, the reference to the original paper is never given (no idea why not) so after a brief hunt, I found the paper here. (Unfortunately, it's paywalled, but this is only one of a number of minor hurdles presented to the modern bloggist.)

If you click on the link to the paper, something very obvious and quite odd should become apparent. A cheeky tactic employed by some researchers is to write to the editor of a journal, have your letter printed, and then by neatly side-stepping peer-review and due process claim you have 'published' a paper. And so it is in this case - not actual published research, just a letter to the editor - how nice. Never mind, let's perservere.

The trial involved 60 people, who had developed cold symptoms within the previous 24h. They were divided into two groups of 30, with the first group getting paracetomol, naproxen and chlorpheniramine (Piriton) and the second group getting the same drugs, but with 50g of natural honey per day as well. So, poor randomisation, no blinding and a small sample - the results are already next to worthless. They were visited by researchers every day to examine the symptoms (rhinitis, muscle pain, fever, throat congestion, cough and sneezing), and these researchers were unaware of which group the patients were in.

The only result the paper gives is that
In the group given honey, duration of signs and symptoms was 1-2 days less than control group.
And how long was the duration of symptoms in the control group? Doesn't say - which is mighty important if the results are to be put into context - was it 4 days in which the honey group halved the time of recouperation or was it 14 days in which the honey (along with all the flaws in the methodology) did almost nothing?

Now don't get me wrong, this mightn't be useless research - it could be what some people call 'skunk' work - a little trial on the side just to see if it's worth investigating further under more suitable conditions (possibly for funding reasons). However, it's important to look at it for what it is - a correspondence to the editor of a journal, not peer reviewed, not blinded, not placebo-controlled, not randomised, lacking in detail and information, and hence completely unreliable.

What it does not show is the remotest glint of reliable evidence that demonstrates that honey is any good for colds - I hope I have persuaded you of that. Compare this again to the headline "Eating honey shortens colds by two days" - not only is it bollocks, but the Daily Fail have overegged the pudding by using the '2 days' rather than the woolly and meaningless '1-2 days' that was in the not peer-reviewed, not blinded, not placebo-controlled, not randomised, lacking in detail and information, and hence completely unreliable letter.

The article goes on to describe
a recent trial at the Dubai Medical Centre, [involving] 16 adults with a history of recurrent cold sores.
After a hunt, I think I found the paper - if you call 2004 recent, and if by cold sores you mean "labial and genital herpes lesions". Still they got the Dubai bit right. A unblinded, unrandomised, non-placebo controlled trial of 16 people is, yet again, not worth a jot.

This is the state of UK media science journalism. If this is the level of effort being put in by the hacks at the Daily Mail for their science stories, what does that say about the level of truth, fact and effort in the rest of the paper?

I think it needs some honey.
Many thanks to Samuel Eaton.BPSDB

3 comments:

  1. People's Journal of Science, love it.

    ReplyDelete
  2. "The most recent story to be CtrlC/CtrlV-ed into the People's Journal of Science (Daily Mail)..."
    Nice.

    "A cheeky tactic employed by some researchers is to write to the editor of a journal, have your letter printed, and then by neatly side-stepping peer-review and due process claim you have 'published' a paper."
    I think this is something I've come across before. Very cheeky indeed.

    ReplyDelete
  3. Great little blog! It's amazing how few scientific studies of honey there appear to be. I swear it helps my stomach when I have troubles but maybe that is just placebo... anyway you missed that it also cures eye infections: http://www.dailymail.co.uk/health/article-2178816/Man-cures-painful-eye-infection-99p-jar-honey.html
    Or perhaps it's another ctrl-c/ctrl-v!

    ReplyDelete

Share it