Better safe than sorry?

I’m no friend of the Precautionary Principle. It is not a principle, but a rhetorical device, which can justify action and inaction, depending on one’s fears, rather than rational analysis.

A mathematics professor has done the maths underlying current programs for early screening of breast cancer[1], and his analysis is startling. For well-meaning politicians, it yields a most difficult conundrum.

Mammograms detect breast cancers: if one screens early, so the conventional wisdom, one can save lives. Assume that 0.4% of women have breast cancer. If we screen, we save 4000 lives. We screen – we do.

But one has to read the fine print. No test is perfect: let’s assume in 1% of instances it is wrong – it yields a “false positive” result. This is plausible: we’d err on the side of caution. Lacking further knowledge, we act on the Precautionary Principle and treat. The treatment was unnecessary. 9’960 women will have been treated without cause (the number is slightly less than 1% because we assume that we deducted all the true cases from the screened population).

The policy decision as to whether to screen early or not is to weigh the 4’000 lives saved against the 9’960 disturbed by false alarm. For each live saved, we blight over two lives for nothing – with medical costs added in. How would one react to the news: “The bad news is – you’ve tested positive to the screening. The good news is: you only have a chance in three that you actually do have cancer”? The self-evident application of the Precautionary Principle becomes less self-evident, once you include the costs of false alarms.

Add insult to injury. As we know more about oncology, we discover that cancers vary greatly. Some are so aggressive that even early detection would not help. Some are so slow-growing – the chances are the patient will die of something else. The sobering result of the US Services Task Force is to slow down on preventive testing. From a public health point of view its benefits are uncertain.

US Vice-President Dick Cheney has argued that if there is a 1% chance that a rogue country has WMD one should act preventively on that assessment: “Better safe than sorry”. I can see him madly swatting one country after the other with a nuclear fly swatter, while shouting: “One per cent”. But what if he were wrong in his estimate half the time?

[1] John Ammen PAULOS (2012): Weighing the positives. Breaking down the latest mammogram math. Scientific American, January 2012.

Link of original post