Spotted a really interesting article in The Times over the weekend (I can’t include a link to the article because you have to pay these days to view The Times online – but if you want to read it, and you don’t mind splashing out, the headline is: Genetic testing companies admit: our risk assessments are confusing).
Basically it’s about three genetic testing companies and how they may allegedly be misleading people about their risk of developing disease. So basically you send off a sample for DNA testing, then they look for risks for certain diseases that you have inherited from your family (don’t ask me how they do that, it’s far too clever for my tiny brain).
So, by taking your relative risk as signalled within your DNA, they then compare it to the average risk – or the risk of that particular disease seen in the population. By multiplying one by the other they get your personal risk. Again, it’s all a bit complicated and I’m hardly the world’s best mathematician (if anything, I’m the worst). But the gist of the article is that all three companies use different statistics for their population risks.
Let me explain with an example from the article. The reporter who wrote the article had his DNA analysed and found he had a higher-than-average risk for developing exfoliation glaucoma – an eye disease. But while one company calculated his lifetime risk of developing it at 36.2 percent, the others put it much lower (one said it was one in 15, the other said it was one in 45). All three are not even in the same ballpark.
The problem is that thecompanies were each using different population statistics to do the calculation (only one of the companies declared what statistics they were using).
All of which just goes to show that while most people think statistics – when they come from a scientific source – are gospel. Well they aren’t. And that’s something all riskfactorphobes should keep in mind when the next big health scare story comes along. There’s nearly always more to it than meets the headline.