Thursday, September 07, 2006

What's Good For You Is Poor Science

Billed as a sort of Australian Mythbusters, What's Good For You is a commercially produced health program with the ostensible aim of testing various ideas about health topics. EoR's only seen bits of the program occasionally and, when he did, it struck him as a bit light and not nearly as much fun as Mythbusters. WGFY weren't even blowing things up!

If a sample of the segments archived at the WGFY website are anything to go by then the program is really just another "current affairs" magazine style glitzy production with no real substance.

Take mobile phones, for example.

You probably can't remember not having one, but mobile phones have only been with us for 25 years. It's not long, particularly in terms of studying their effects on our bodies.

This clearly shows the target audience: those young enough to have lived in a world where mobile phones have always been a constant presence. EoR is a little older than that. It also clearly shows their lack of a proper understanding of science: 25 years is long enough to show any effect that takes less than 25 years to have an effect. For example, all the brain tumours people keep claiming are a result of mobile phone use.

Rather than test this particular "myth" the program opts for anecdotal "evidence". The program interviews an extremely heavy user of a mobile phone (who has noticed no side effects), the director of a "consumer protection group" (protecting from whom? from what?) who is so rabidly anti-radiation he's apparently screened his whole house (EoR wonders if he wears a full body suit made from tinfoil whenever he ventures outside), and a man who has developed something described as "electromagnetic hypersensitivity" which acts in a very strange manner (he can't use fluorescent lights, but radiant globes seem to be okay; he can use a speaker phone but can't sit in a cinema). We are told that this man's case is "extreme" and "contentious". Indeed. So why even bring it up, other than the predilection of such programs to run scare stories about common household objects?

Against the single scare story is weighed a proper scientific study of 6000 cancer cases. Even that, however, isn't proof enough.

"What we can say is there doesn't appear to be an increased risk of developing a brain tumour from using a mobile phone. However, that's not to say in any one individual case that we can completely rule that circumstance out," she says. That's not entirely reassuring, and the fact is, we can't be any surer than that. Many tumours take decades to develop and we haven't been mobile phone maniacs for that long.

Of course science can't tell definitively the cause of cancer in a single case. That's why the study looked at 6000 cases to determine if there's any perceptible trend. But people have claimed for years that they got their cancer from mobile phone use. If those claims were true there'd be more than enough cases to detect a causal link by now. The program's conclusion?

The good news is at this point in the communications revolution there is no convincing evidence that mobile phones will make you sick. But it always pays to be cautious!

Ah, yes. No evidence. But be scared anyway. Let's move to the bunker in the hills! But not anywhere near those dangerous mobile phone towers...

Meanwhile, what's an effective method of determining the sex of your unborn child? Five mothers (note the small sample size) test dowsing with a wedding ring, urine sampling with drain cleaner, Chinese astrology, and the mother's heart rate.

The results?

Two from five on the heart rate.
Two from five for the drain cleaner.
Four out of five for the Chinese lunar calendar.
The wedding ring pendulum didn't miss a trick with five out of five.
Maybe it's just one of those freaky things, but somehow, the ring seemed to know exactly what to do. So just remember: side to side it's a girl, round and round it's a boy.

It'll certainly save money on ultrasounds. The real Mythbusters don't always have perfect variation free tests, but at least they have a clearer idea of how to conduct an experiment with some fairly simple ideas like running sufficient trials, and eliminating outside variables.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.