Has it stopped rising? From what I can gather even Michael Marmot only really said that a) its rate of increase had slowed and b) that it was worth investigating whether austerity might be involved. None of which stopped the story being way more extreme on both counts, fuelled in some cases by people who really ought to know better.

So what to think? The first thing to remember is that Life Expectancy (LE) is a fairly artificial construct. It doesn’t actually describe the likely lifespan of anyone, given that it is constructed from contemporary risks – as if a baby born today were to experience today’s age specific risks throughout life. As a result, LE as a measure can change rapidly because of contemporary pressures.

Secondly, the excess winter deaths of 2014-15 make a substantial difference to the calculations and visual appearance of the data. We will only really know what is happening with that trend when we can see if that was a one-off event or something more sinister.

Thirdly, the appearance of slowed increase in LE may be more the result of an upward deviation from trend between 2009 and 2011 rather than a subsequent downward deviation, thus:

Without those better, earlier figures the furthest right points would not appear anomalous, notwithstanding the 2014-15 excess winter deaths.

This sort of pattern also invites speculation that there is a ‘frail survivor’ effect here, whereby older individuals, having avoided earlier insults, are particularly vulnerable when subsequently exposed to e.g. a new circulating viral strain.

On Radio 4 this morning, the ‘Thought for the Day’ speaker reflected the suggestion that we could be reaching a limit to extension of life expectancy. To which it is worth pointing out that people have been predicting that for many years and have found it necessary repeatedly to revise their assumptions (e.g. see Oeppen J, Vaupel JW. Broken Limits to Life Expectancy. Science (80- ). 2002;296(5570):1029–1031.)