Archive for January, 2008

Drinking water – or bathing in it – can be deadly (not)

January 31, 2008

One of the staples of the Alt Health world is to convince people that all sorts of rather commonplace things are, in fact, bad for you.

The scariest is if something “increase your risk of cancer”. Although heart disease kills more people, cancer tops the list of the diseases people are most scared of, and the mention that something “increases cancer risk” is a sure-fire scary scare.

Once you have your marks, sorry, patients, scared, and doubting the safety of hitherto commonplace and unscary things, then the Magisterial Authority of the Nutritional Expert, or Holistic Practitioner, will give them something to cling to. And pay for.

Resulting, of course, in a pronounced lightening of the “patient’s” wallet.

If you work it really well, the people are so grateful that they never realise they have been sold the cure to a non-existent illness.

So let’s move on to the actual scares. Many of the lines Alt practitioners use to make people nervous, and foster the belief that “everything is bad for you”, are actually flat-out nonsense – the farrago over mercury in vaccines and autism being one famous (or infamous) example. However, sometimes the Alties’ lines may be based on at least a grain of evidence.

Should this, then, convince you that you have been lied to by the mainstream, and that the Alties are revealing important truths that were kept from us?

Well, not necessarily.

The statement may be true, but it is important? Or, to put it more scientifically, Is it significant?

The Devil, as they say, is in the detail.


Something nasty in the water…

Let’s discuss a real example.

Dr John Briffa is a medical doctor and a well-known writer about nutritional topics in the UK, both in the media and via his blog at For some years now, Briffa has been telling his readers that they should drink bottled water rather than tap water. His stated reason is (taken from a recent post on his blog):

“…there is quite a body of evidence linking the consumption of tap water with an increased risk of cancer” [1]

and from a few years back, when he wrote an Observer column:

“The sanitation of tap water taints it with chlorine and by-products of the chlorination process that have toxic potential. Studies have linked the consumption of tap water with an increased risk of certain cancers (such as those of the bladder and rectum).” [2]

Now this second statement is more accurate, and adds some information – here you are told what cancers, not just “cancer”. But still, on its face, this sounds really scary. I suspect a certain fraction of the people that read this, particularly those least scientifically and medically literate, will see this as:


“Don’t drink tap water – it will give you cancer”


When Briffa repeated the “avoid tap water” comment on his blog recently, one reader asked him to give the evidence to back up the statement about cancer risk. In response Briffa reproduced the abstracts of several scientific studies and meta-analyses (see comment 3. in the thread at Briffa’s blog, [1]). There was no explanation of what the studies meant, and no context – see below – but they were real enough, published in proper scientific and medical journals.

So there is evidence to back it up! It’s true! Tap water increases your cancer risk!

Well – yes, and no.

The increase in risk is consistent between studies. Indeed, it is sufficiently well recognised that some mainstream health sites dealing with bladder cancer repeat it. For instance, the Cancer Research UK site says quite plainly:

“Drinking chlorinated water (water with chlorine added) may cause a small increase in bladder cancer risk compared to water disinfected in other ways.” [4] [Emphasis mine]


– Note that this means this can hardly be: “A terrible secret they tried to keep from you” – more “something that might be true, although we’re not sure, and with the causation not settled, and viewed on balance as an acceptable risk, as well as not being easily avoidable”. We should remember here that 120-odd years ago, before we had safe drinking water, bacterially contaminated water regularly used to cause epidemics of nasty diseases like cholera and kill loads of people. This is still true in large parts of the developing world, as you can read here.



Context, context, context

SO… what is the snag with the statement as presented on John Briffa’s blog?

To my way of thinking, the problem is the lack of explanation, and particularly of context. The basic statement “there is quite a body of evidence linking the consumption of tap water with an increased risk of cancer” leaves out several crucial pieces of information that you need to put this information in context.


– It doesn’t tell you how common the cancers for which this has been shown actually are;

– It doesn’t tell you how much the chlorinated water is said to increase the risk;

– It doesn’t tell you if we know for certain that drinking the stuff causes the increased risk, as opposed to, say, bathing in it;

– It doesn’t tell you how well the studies eliminated possible “confounding factors” – for instance, if all the people drinking chlorinated water lived in polluted cities, while those drinking artesian non-zapped water lived healthy lives in the country.

– It never makes the basic point that it is notoriously difficult to infer causation from epidemiological studies of this kind, partly because of the confounding factors. A famous saying is “correlation is not causation”


– It doesn’t tell you that all sorts of other things will slightly increase, or decrease, your risk of these kinds of cancer.

[This last one is the source of numerous black jokes in science about everything in the world either increasing, or decreasing, your risk of cancer. Sometimes the same thing increases the risk of some cancers and decreases the risks of others. Swings and roundabouts.]



Anyway, without this kind of information, which you are not given by being told “There is scientific evidence that cholorinated water increase cancer risk” you cannot make a sensible judgement about whether to be worried.

Instead, you just worry. And then drive off to the supermarket in your 4×4 SUV to get a huge multi-pack of bottled mineral water. Ker-chhhinnng.

So what is the truth about chlorinated tap water and cancer risk? If I was putting in the context, I would state it roughly as follows:


“In studies, people who drinking chlorinated tap water all their lives seem to have a very small increased risk of bladder cancer – about 20% – and perhaps of rectal cancer too. However, we are still not totally sure this is due to the chlorinated water, and not to something else that is different about the groups of people, like maybe the ones who drink bottled water simply drinking more than the ones who drink tap water. And even if we accept the idea that it is really the chlorinated water that does it, we don’t know for sure whether the risk is due to drinking the water, or bathing and swimming in it – probably both. And you have to drink or bath in the stuff for at least 35 years to get this 20% increase in risk. There is no suggestion chlorinated water increases the risk of other kinds of cancer, including those which are far more common than bladder cancer, like lung cancer and prostate cancer.

It is also important to remember that risk factors do not actually tell you if you are going to get a disease. Most people who have all the risk factors still don’t get the disease. And conversely, some people who have no risk factors do get the disease.

Next, even if we accept the “chlorinated water hypothesis”, the increased risk of bladder cancer from drinking chlorinated water is probably considerably LESS than the increased risk of this cancer you seem to have if you don’t eat the recommended five daily portions of fruit and veg. It is definitely far, far less than the increased risk of this cancer that you have if you are a smoker (which doubles or triples the risk – smoking is thought to account for about 50% of all bladder cancer cases). And if you are a non-smoking man, simply being male is your main “risk factor” for bladder cancer (it is three times more common in men). (For the source of the relative risk figures, see [4])

Bladder cancer, by the way, is usually diagnosed when you are past 70, and is one of the more treatable cancers. You are statistically about as likely to die of bladder cancer as you are to be killed in a road accident. You are about twenty times MORE likely to die from heart disease, eight times more likely to die from a stroke, and about three times more likely to die of prostate cancer.

PS Finally, to retain a sense of proportion, it is also necessary to bear in mind that life has a 100% mortality rate, so something is definitely going to get you in the end.”

I didn’t know about the chlorinated water / bladder cancer work, so it came as a slight surprise when I heard about it from one of Briffa’s readers. However, having looked it up, I will not be giving up drinking chlorinated tap water.

What I dislike about John Briffa’s way of presenting this, as will by now be apparent, is the lack of context. Another example – no absolute risk figures. Bladder cancer is not a rare cancer – in Europe it is about the 5th most common cancer overall– but that still doesn’t make it common, in the sense most people fear, namely “you are quite likely to get this and die of it before your time”.

Because so many things alter the risk, it is hard to give a precise number for “how likely is it I might get bladder cancer” As a ballpark, as a non-smoking man eating a reasonably healthy diet, your risk of getting it by the time you are 70 would seem to be in the order of 1 in 200 at the most. As I have said, there are many other diseases you should be far more worried about. A marginally increased relative risk of 20% for one particular kind of cancer, or even two kinds if one adds rectal cancer, is not going to keep me awake at night.

ESPECIALLY since, even were I to decide to stop DRINKING chlorinated tap water, I am not in a position to stop BATHING in it, which probably accounts for 2/3 of your total “chlorination byproduct exposure”. Mrs Dr Aust wouldn’t stand for the not washing, even if I could.

So, much as I would like to say:

Avoid increased cancer risk – DON’T WASH! EVER!”

…I shall refrain.


Let’s INVERT that risk factor

Risk factors are interesting. They are the meat and drink of medical and scientific papers about causes of disease, but are notoriously poorly understood. And peoples’ reaction to them is quite strange and unpredictable. The same “risk rate” means different things to people in different contexts.

An interesting way to probe how you view a “risk factor”, I find, is to “invert” it – to turn it back to front. This is because “increase” always sounds more scary, and hence more useful to someone trying to sales-pitch you, than “decrease”

Thus, instead of saying “XYZ increases your risk of ABC cancer by a fifth” (itself an advance on just saying “XYZ increases your cancer risk”, but still scary-sounding because it has “Increases” – MORE cancer, hence frightening), try inverting this to:

“Did you know that if you stopped doing XYZ you could reduce your risk of ABC cancer by an Nth?”

That way it avoids sounding scary (no “increase”) and also the focus tends to be on the XYZ, the behaviour, rather than the scary stuff. I would argue this means you can consider the risks in a more sensible “cost-benefit” way, rather than through a prism of “primal fear of nasty disease”.

Let’s take a real example:

“Do you know you can reduce your risk of bladder cancer by about one-sixth if you never drink, or bathe in, or swim in, chlorinated water?”

[Note again that this accepts the relationship as “causation proven”, which it isn’t]

Presented with it that way, I would bet money that most people, for whom bathing in asses’ milk or Evian water is financially out of reach, and who don’t have a Malibu Colony beach house with a solar-heated sea water shower, would probably not be that bothered.

Which seems rational to me. Because by living – not to mention eating, and drinking, and walking and sleeping, and driving a car – you are accepting some intrinsic risks. C’est la vie – literally.


More risks in context

It turns out that reading about bladder cancer is a smorgasbord of interesting papers about risk. Here is another example. Studies report that you actually seem to have an increased risk of bladder cancer if you are a lorry driver – strange but true. In fact, the reported increased risk of bladder cancer for spending a lifetime drinking and abluting with chlorinated tap water (around 20% in several studies) is a little more than the risk increase if you drive a lorry (about 17%), but a bit less than if you drive a bus (33%) [5].

Again, I find it difficult to imagine many lorry or bus drivers quitting on account of this, even if they all knew. They have a living to make, so they will probably accept the marginally increased risk – if they care at all.

So to reiterate: with risk, it seems to me, you need to ask whether we are reasonably sure about the causation; you need accurate numbers, preferably absolute ones, rather than relative risk increases; and you need context. Only then can you think sensibly about whether the risk is a real concern, perhaps enough to want to change your behaviour. On which basis, there some health-risking behaviours that are self-evidently worth avoiding, if you possibly can. Smoking is one. Being grossly overweight is another.

But as for avoiding drinking water out of the tap – give me a break.

My advice: save yourself some money, and a ton of empty plastic bottles.




[4] Murta-Nascimento C et al. (2007). Epidemiology of urinary bladder cancer: from tumor development to patient’s death. World J Urol 25: 285-95. PMID: 17530260.

[5] Boffetta P, Silverman MT (2001). A meta-analysis of bladder cancer and diesel exhaust exposure. Epidemiology Jan;12(1):125-30. PMID: 11138807