Why are Scotland's mortality rates rising?

Elderly man

Jeremy Peat OBE

Jeremy Peat
Visiting Professor, International Public Policy Institute
jeremy.peat@strath.ac.uk

11 January 2017 

In a recent article Will Hutton produced some data showing that, in his words: - ‘The British are in the midst of a critical health epidemic’. He based this very strong statement on data which showed not only a major jump in the number of deaths recorded in the UK in the 12 months to June 2016 (the biggest absolute and relative amount since 1940) but also that this increase was spread across age groups. There had been increases for all age groups from 55-74 upwards.

This seemed of importance to me, so I have dug out the latest official data for Scotland, which cover up to the calendar year 2015. Again there are clear signs of a disturbing trend. That final year, 2015, showed a trend change.

Established patterns

One way of examining the data[1] is by 5 year groupings from 1946. These data show a remarkably consistent trend, for males and females, for the key age groups. Mortality rates per 1000 of the population have been declining steadily quinquennial on quinquennial – with just the odd blip.

This trend applies to 2011-2015 as compared to 2006-2010 just as it applies to other post war periods. For males aged 45 upwards the mortality rate declined, measured over that five year period, for each age band covered by the data. Exactly the same finding applied to females. (As expected the mortality rate in each age band was lower for females than for males, albeit the difference is diminishing, as male life expectancy has been increasing at a more rapid rate than that for females.[2])

An unwelcome change

A different picture emerges when looking at the very latest available data. In 2015 male mortality rates were higher than in 2014 for each of the five highest age bands – 45-54, 55-64, 65-74, 75-84 and 85+. For females this was the case in four of the five age bands, not for those aged 55-64 where there was another small reduction.

These increases were of some substance. For those aged 85+ the increase on the year was 6.7% (male) and 9.5% (female). For those aged 75-84 the increases were 2.8% (male) and 2.0% (female); and for the 65-74 age bracket 3.5% (male) and 6.0% (female).

For those aged 85+ the mortality rate was the highest since 2006 (for males) and likewise for females. For the four lower age bands the change was not as extreme, but I repeat that the mortality rate (and recall this is adjusted per 100 of the population so not due to the impact of an ageing population) was higher in 2015 than in 2014 in all these age bands for males and in three out of four for females.

Taking the population as a whole, the increase in the age-standardised death rate from 2014 to 2015 was 5%. By way of comparison the reduction from 2005-2015 was 12% and from 1994-2015 25%. For males the equivalent figures are +5%; -15%; and -31%; for females +6%; -11%; and -21%. In other words the increase in mortality rates is roughly similar for both genders, and the change in trend more significant for males than females.

Why are mortality rates rising?

The available data show the major underlying causes of death and how mortality rates have changed by each cause[3]. The increase from 2014-2015 is across the board. The biggest % increases were in deaths due to ‘diseases of the respiratory system’ (16%); ‘chronic obstructive pulmonary disease’ (12%); and accidents (12%)[4]. Whereas most causes have shown a declined mortality rate from 1994-2015, there was an increase for ‘alcohol-related’ (+29%). This applied to males (+29%) and females (+25%). For females there was also a 45% increase (again from 1994-2015) in death from ‘chronic obstructive pulmonary disease’.

So what to conclude? Two immediate points should be made. First these are data for only one year. Those for 2016 could show a reversion to trend – and indeed the 2015 data may be subject to revision. Second, no causation for the change can be implied from the data. Will Hutton referred to a ‘critical health epidemic’. I am not rushing to a similar judgement for Scotland. But I would suggest further examination of these data, by statisticians and health care experts; and examination of the 2016 data as soon as feasible. At first blush these figures show a real cause for concern.

Tags: Health


References

[1] All data quoted hereafter come from the National Records of Scotland.

[2] I chaired the BBC Pension Trust from 2005-2012 and consequently monitored these life expectancy data quite carefully. There was a period when male life expectancy for those in my (then) age band appeared to be increasing faster than I was ageing. I saw this as my own personal Tardis.

[3] Op cit.

[4] This is under the new definition of accidents. Under the old definition the increase was 10%.