Book of the Month: And our Geek Lit winner is... BAD SCIENCE by Ben Goldacre (discussion night Tue 30 March)

[TillyBookClub]

BAD SCIENCE by Ben Goldacre has grabbed the winning title of March Book of the Month with a staggering 50 votes, streaks ahead of his competitors.

We will get together to chat about the book on Tuesday 30 March, 8-9.30pm.

For those that missed it, here were the choices and the March poll results.

And for anyone new to Bookclub, here is how it works.

I'm hoping Ben may be able to join us to answer all your questions - will keep you posted.

I can't find anything about humans transferring expectations to a dog - I only remember the part where they injected the dog with sugar water plus a immunosuppressant, and then afterwards the sugar water had an effect due to association.

I'm fascinated by the placebo effect, and I think it is the most mysterious and interesting part of the book.

Also, re: homeopathy, his argument isn't all homeopathy=placebo. He also spends quite a lot of time looking at concepts such as regressing to the mean. I.e, the chances are that even if you do nothing you'll eventually start to feel better.

With animals & children obviously this sort of concept is completely possible. If your dog has a self-limiting condition and you just happen to give the homeopathic remedy at the point where they'd have started to get better anyway you'll get a positive result. Or at least it'll seem like you did.

It also depends on who decides that the animal is "better".

As animals (and babies) can't generally tell us when they're feeling better, we interpret their behaviour to decide this. This is a subjective process. If you've given a remedy that you expect to work you're more likely to see what you believe is going to happen - recovery.

Of course, this depends on the condition. If you were using, say, a blood test or other biological measurement to judge whether there had been improvement then this would take away the subjectivity.

And (I'm sorry - this should really all be one post), if I remember rightly, this is one of his arguments against homeopathic research, that they're not controlling for this sort of subjectivity so the results can't really tell us anything useful.

SuziKettles: agreed, he does give good footnotes. And he makes a very valid point in the last chapter that anyone can understand very complicated science but only if they're motivated enough. Which usually means you are in hospital or have a sick child, or another strong impetus to take on all the information.

I think am now motivated enough to be sceptical, to think round an issue and to follow up research in one or two areas. But on the whole, I expect I will still be taking someone's word for it, whether that's my GP or Ben Goldacre.

But I think the fact that you realise you're taking someone's word for it is important.

Anything that makes you start asking questions has got to be positive.

The trouble is that you don't really know what questions to ask. Or understand the answers you are given.

Yes, i think regression to the mean might explain some of the placebo effect on children/animals. Also good point about the subjective interpretation of illness/recovery.

DH and I loved this book. The homeopathy chapter very good at explaining why it is rubbish. Interesting points about the MMR issue - no controversy in other countries, in this country the controversy took some years after the original, now retracted Lancet paper.

The things that I think about all the time are the points about terribly poor science/medical reporting and the contrast with the business pages where you get proper educated reporting. Also the stats - 50% increase in cancer risk if you do x, not that important if actually your risk goes from 2 in 100 000 to 3 in 100 000!

Yes, I fully support the regression to the mean argument.
Is the same theory applied to conventional medicine? Apologies if Ben covers this in the book - I am worried now that I will get another ticking off...

it is simply not true that there is no controversy about MMR in other countries.

Re: placebo effect in children. You can buy calpol-style placebo called obecalp ... read it backwards. Couldn't comment upon whether it works though.

I can't see why not really. If I understand correctly placebo works because of the belief that it will work. Children readily believe that medicine will make them feel better, and provided you can make a reasonable stab at convincing them it is real medicine, they are perhaps ideal candidates. Over the counter placebo is likely to be diminished in adults by the fact that you know that there are no active ingredients - this obviously doesn't apply to children.

I can't be the only one who has given Calpol as a last resort to soothe a distressed child and wondered if paracetemol could really have such a transformative effect, or whether it is in fact magic. Maybe it is just the placebo effect

It does occur to me though that the placebo still may not work as well - BG discusses in his book the fact that doctors' beliefs about whether the drug would work or not affected its effectiveness, this would seem to apply equally to parents. Think I'll just stick to Calpol

I really liked the book by the way. Really interesting to have statistical theories like 'regression to the mean' explained clearly. Has changed the way I 'read' news stories about science now. But I think the book/column does confirm my own prejudices, so I was always going to like it!

Agree about children who are old enough to 'get' medicine, i.e to know what it is and what it does. I was just astounded when a friend of mine with a 12 month old tried homeopathy for a terrible blocked tear duct (as diagnosed by the doctor and about to be operated on) - and it worked. Cleared the eye in 24 hours. It was most likely regression to the mean rather than placebo, but it made me wonder if babies that young could be affected by placebo psychology.

Not that it made me believe in homeopathy, which I think is brilliantly (and funnily) trashed in the book.

I feel I need to read it at least 10 times more, to absorb all the facts and prepare for informed arguments.

Well, i think there certainly are diseases which doctors are under pressure to 'do something' when actually they are self-limiting. Eg, the demand for antibiotics for viral diseases such as colds and the like which will get better by themselves (ie regress to the mean).

And the placebo effect does IIRC account for about 10% of the effect of conventional medicines. Hence research requiring double blind placebo trials.

V interested by Obecalp. Considered making my own placebos out of small sugary blobs. Then remembered Jelly Tots which have probably nailed most of my children's placebo needs.

Wondering what happened to all the 50 people who voted for the book? Perhaps bumped off by the evil big pharma...

Thanks to all who did make it and see you for the next bookclub chat on Tues 27 April.

I would be prepared to stick my neck out and suggest that the ONLY conditions that alternative therapies appear effective for are the self limiters. A really good example is Molluscum. Generally, people leave it,knowing that it will eventually go by itself unless they are motivated by it starting to bleed or look infected to try some remedy - which appears to work because molluscum almost always get "worse" just before they go naturally. Hey presto, the remedy worked!

The problem is it is utterly patronising to tell other people that ALL alternative therapies are nonsensical and that ANY postive effect is imagined.

I am sceptical about homeopathy myself. I can see how scientifically it appears nonsensical. But to apply this to all other alternative/complementary therapies [osteopathy, nutrition] is extremely blinkered and arrogant. The anti-inflammatory properties of fish oil, as well as other benefits, are mentioned here

that is not to say Gillian McKeith is not the anti christ

this a very recent study done by University of Southampton with patients in intensive care

But BG doesn't say that does he? He's just talking about Bad Science, specifically scientific claims made without any evidence to back them up.

So, if an osteopath or a nutritionist or someone promoting fish oils makes a claim about a benefit and produces research, properly conducted which minimises the risk of bias, subjective reporting of results etc etc etc then I really don't think there's a problem.

Of course all this shouting about woo that you seem to get everywhere now is deeply patronising I absolutely agree.

Again, I've never got the impression that BG is saying "fish oil is shite. It will always be shite and I don't care what you say I'm not changing my mind on that".

What he said was that the Durham Fish Oil "Trials" were actually a big advert for the company supplying the capsules, and I don't think Durham County Council (or whoever it was) did themselves any favours over that whole episode.

And actually, it doesn't matter if it turns out that fish oil is a panacea for all the world's ills, the Durham Fish Oil trials will still have been poorly carried out "research" which won't have added anything meaningful to our knowledge about fish oils.

Yes but there was better, peer reviewed research on fish oils that BG should have included in his book as a counter balance. But he didn't. And so loads of mothers go around saying 'I've stopped giving my dcs fish oil because BG says so'

Which isn't much better than following Gillian McKeith.