Cancer: ‘Mortality’ and ‘survival’ rates are not the same

Cancer – and particularly Breast Cancer – has become a political football with questions being asked over whether very large increases in investment by the last government have led to better results. This morning the BBC and other news sources are headlining a new report in the BMJ showing trends in Breast Cancer ‘mortality’ throughout 30 European countries.

The major findings for UK readers are that whilst mortality has declined in most countries, the UK starting from a high base has achieved one of the biggest improvements—with about a third less deaths.

This study starkly contrasts with the equally well-reported EUROCARE study that found relatively poor improvements in ‘cancer survival’ in the UK. Why the difference? Well, ‘mortality’ and ‘cancer survival’ are different indicators each with their own problems.

Mortality as the name suggests measures the rate of death from a disease within the whole population; e.g. 28.2 breast cancer deaths per 100,000 in the UK according to the BMJ study. Mortality figures were highlighted in the election when David Cameron questioned the value of the extra cancer investment.

He said:

“Now what Gordon Brown is not telling you about the situation with cancer, cancer drugs and cancer outcomes is after all of the things that he has talked about, all of the money that has gone in, our death rate from cancer is actually worse than Bulgaria’s so all that has happened has not actually improved the outcome which is what matters.”

Well – so what? Whilst I’m reluctant usually to rely on anecdotal evidence, I‘ve been in a Bulgarian hospital and I’m pretty confident that their low cancer mortality has nothing to do with standards of care.

At a guess I would say it might be their record keeping or the fact they die on average much earlier of other causes. Mortality is probably a useful metric for following cancer trends but it has problems comparing between countries.

Meanwhile survival is the percentage of people that are alive sometime after diagnosis – e.g. 49.6 per cent overall cancer survival after five years in the UK according to EUROCARE. However it is as malleable to changes in diagnosis as it is to changes in the success of treatment.

The authors of the BMJ report point out that the countries that have reported the greatest success in improving EUROCARE4 survival rates are also those that have reported the greatest increases in incidence.

In short they suggest these countries have instigated intensive screening programs and have identified many small relatively benign cancer – many of which may never have progressed or proved life threatening at all. So perversely intensive screening can greatly improve survival without improving mortality very much.

Furthermore, in an accompanying editorial (behind paywall), to the BMJ study it is claimed that UK survival statistics are also lowered due to an administrative gap between diagnosis of cancer and the often-later registration of diagnosis. D’oh!

On the whole I trust the BMJ study. The improvement reported is consistent with the experience of most cancer professionals. There is a larger message here however over just how slippery statistics can be, and particularly international comparisons. Evidence based policy is seldom simple.