Wednesday 18 May 2011

Health Not Scare Story Du Jour

From the BBC:

Coffee has been linked to a reduced risk of dying from prostate cancer in a study of nearly 50,000 US men.

Those who drank six or more cups a day were found to be 20% less likely to develop any form of the disease - which is the most common cancer in men. They were also 60% less likely to develop an aggressive form which can spread to other parts of the body...


That's great news, isn't it? Oh, wait...

"But charities say the evidence, reported in the Journal of the National Cancer Institute, is still unclear. They do not recommend that men take up coffee drinking in the hope of preventing prostate cancer."

At the risk of writing what you're already thinking, do we not suspect that if the study had shown that drinking lots of coffee increased the risk of prostate cancer, these cancer 'charities' would find the evidence to be crystal clear?

PS, the National Cancer Institute is a department of the US government, if that's relevant.

18 comments:

dearieme said...

If that's old-fashioned US coffee, you'd be as well off drinking your washing-up water: it'd be cheaper, taste just the same, and be subject to equally ropey science.

Macheath said...

In the land of the free, where everyone has the right to life, liberty and a large Starbucks macchiato, they might be better advised to ask why the non-coffee drinkers didn't touch what is effectively their national drink.

Several of my relatives have suffered from an identical form of aggressive prostate cancer - almost certainly with a genetic link. To a man, they actively loathed coffee, claiming it was too bitter to drink, while the coffee-drinking family members have - so far - escaped the disease.

At least the scientists ivolved have the decency to admit, "At present we lack an understanding of risk factors that can be changed or controlled to lower the risk of lethal prostate cancer."

But my guess is that if you aren't drinking coffee bcause you hate it then you've lost the genetic lottery as far as prostate cance is concerned and no amount of coffee is going to help that.

Anonymous said...

Less of the talk about genetics please. If word got out that obesity, alcoholism or damage from smoking were to any extent governed by genetics, it could seriously damage the healthscare promotion and nagging industry, which provides so many badly needed jobs for the boys and girls.

Macheath said...

Anon, I don't believe for a minute that it would stop them in their tracks, if the experiences of some of the family members mentioned above are anything to go by.

Apologies for self-linking, MW, but it would take too long to explain.

Mark Wadsworth said...

D, that's as true as ever.

McH, good summary, good maths.

Anon, but it would be a great excuse for a massive 9and massively expensive) genetic screening program, and extra terrorisation of 'at risk groups', as McH outlined in his own post.

JJ said...

Has anyone ever proved that genetic malfunctioning in our DNA isn't responsible for cancer.

Or is that a double negative?

View from the Solent said...

JJ,
Nope, no double there. isn't = is not.
But you can't prove an "is not", or negative, connection. That's why the precautionary principle bandwagon hurtles on.

View from the Solent said...

So we can add prostate to ovarian, skin, breast , ...

See the index at heroku to a well-known source of medical expertise.

dearieme said...

"At present we lack an understanding of risk factors that can be changed or controlled to lower the risk of lethal prostate cancer."

Are they stupid or liars? A "risk factor" is merely a positive correlate. There's no good reason to believe that fannying with correlates would change anything. To effect change you need to identify causes.

P.S. It's my suspicion that almost all medical causes are genes or germs. That's the inclusive "or".

Mark Wadsworth said...

JJ, no they haven't, in fact, I thought that they had. And not it isn't.

VFTS, I love that list. It's a bit like Homer Simpson's comment on alcohol: "The cause of, and solution to, all of mankind's problems."

D, yes, but in theory a risk factor is sort of a pointer as to what the cause might be. And when they say 'controlled', they might mean it in the statistician's sense of 'controlled for', i.e. comparing absolutely like with like except for one single variable.

Anonymous said...

Coffee can also help prevent Alzheimer's.

It also kills bacteria, suppresses appetite, acts as a diuretic and interferes with sleep.

On the downside, it can apparently cause heart attacks.

It's really a pretty powerful drug. Expect attempts to regulate it soon.

DNAse said...

Has anyone ever proved that genetic malfunctioning in our DNA isn't responsible for cancer.

This is effectively a definition of cancer. Specifically malfunction of the regulation of cell growth and turnover. The malfunction is either caused by DNA errors or faults with the components involved in "reading" the DNA within a cell. Typically since these processes are multi-factorial there is redundancy such that a "double hit" is required to trigger a cancer. External factors (chemicals, virus, UV light) can be a cause. But if one "hit" has already been acquired by an inherited DNA error then such people are more susceptible to cancer since just a single "hit" is needed from external factors.

Mark Wadsworth said...

AC: "Expect attempts to regulate it soon."

Yup.

DNAse, you're the expert :-)

dearieme said...

"but in theory a risk factor is sort of a pointer as to what the cause might be": no it isn't. It could be caused by the disease, for instance, rather than causing it; or the disease and the "risk factor" might both be caused by some third thing. It's a low advertising-man sort of stunt even to name it "risk factor": it's the sort of false reasoning we were all warned across in our beginners' lectures in statistics.

Mark Wadsworth said...

D, in which case I have clearly misunderstood 'risk factor', I though (in the case of cancer) it meant something like 'Your parents or grandparents had it' or 'You smoke' or 'You work with dangerous chemicals'.

Can you give examples?

James Higham said...

Studies show that reading blogposts after 11 p.m. is injurious to total amount of sleep.

Ian B said...

Well, datamining. Nobody would have run a 20 year survey on coffee and prostate cancer. It's one of those general datasets like the infamous Harvard Nurses Study, no doubt. So this is just a "find a slight correlation, publish" thing.

My bet FWIW is that 150 years from now, everyone will be saying how cancer turned out to have always been caused by viruses (papilloma et al) and how funny it is that back in the ignorant 21st century everyone thought it could be avoided by Social Purity programmes.

They won't ever get cancer by then. Or suffer infections. Or go bald. Or get fat. Or have heart attacks, or strokes. Or grow old. I sometimes feel really quite angry at having missed the Eternal Utopia by so few years, really I do. We're in the last few generations of Homo Sapiens 1.0. That's quite a weird feeling. For me, anyway.

dearieme said...

It's well known that drinking diet coke is a risk factor for obesity.