“Rudi Giuliani, aspiring US President, had prostate cancer. His chance of surviving in the US, he said in August, was 82 per cent; [but in the UK], “under socialised medicineâ€, about half as good. There was even a graph to prove that the NHS was a killer….
“Was Rudi right? Here’s a clue. Is it remotely likely that prostate cancer strikes three times as many in the US, or just that three times as many instances are diagnosed? Bump up the number of diagnoses — when deaths stay much the same — and, presto, there’s your massively higher survival rate. Innumerate, but brilliant.”
The most amusing thing about this is the title of the article in which it appears: The worst junk stats of 2007. So The Times intends this to be read as an example of a ‘junk stat’, a piece of incorrect or misleading information. How is it?
Even if what the piece suggests (without supporting fact) is correct – that it is the number of diagnoses of prostate cancer that is higher in the US than in the UK – how does that make anyone feel better about the UK National Health Service, or even invalidate Giuliani’s point in any way? Giuliani never speculated about what exactly it was about the UK system that made it worse than the US system: he simply made that point that it was worse (in order to argue that universal healthcare constitutes worse healthcare, a point I made myself here). Isn’t a suggestion that the UK may be diagnosing less cases of prostate cancer than the US just as severe an indictment on the UK system as some other explanation of the figures?
Or let’s take the assertion of the article that prostate cancer may simply affect three times as many in the US as the UK. Why would this be the case? What fact contributes to this affect? Where is it documented, evidenced, established?
And The Times has entirely missed what was really wrong with the quote. As I reported here, Giuliani simply got his figures wrong. The Times doesn’t even mention that, which seems to me the most pertinent fact to mention. But, as I also pointed out, even that doesn’t affect Giuliani’s argument: the point of citing the statistic is not affected, since the correct figures still go his way.
Clearly The Times put a lot of thought into this article… for a piece which purports to expose the shoddiness of thinking in the stats of others, it’s not doing very well at getting it right itself. Who are the junk staticians behind it? Andrew Dilnot and Michael Blastland. Anytime you see those names in the future, you’ll know what to expect.