Increased Winter Mortality

In most north European countries more people die in the four months from December to March than during the rest of the year.

There were 2,060 “excess” deaths in Scotland during the winter of 2018/19.

Contributing factors include low income, inefficient heating systems, external temperature fluctuations, and excessive dampness and mould growth associated with poor housing stock.

In Scotland there are long and damp winters, which are worse for health than dry Scandinavian winters. Factors affecting excess winter mortality are varied and complex, but there is a strong relationship between thermal standards in housing and excess winter deaths.

The UK has much higher winter deaths rates than other countries with more severe winter climates, implying that it is not outdoor exposure to cold that is the key determinant. Northern Finland, where winter temperatures regularly drop to -20°C, has a significantly lower rate of excess winter deaths than the UK.  However, Finnish buildings have historically had much higher levels of insulation and whole house central heating has been the norm for years.

It is generally accepted that the number of excess winter deaths could be reduced if everyone could be kept warm in their homes during the winter months.

Statistics on increased winter mortality

The Increased Winter Mortality (Excess Winter Deaths) figure for Scotland in 2018/19 was 2,060.

Increased Winter Mortality or Excess Winter Deaths are recorded by the National Records of Scotland (NRS).

There is no single cause of ‘additional’ deaths in winter. Very few are caused by hypothermia. Most are from respiratory and circulatory diseases such as pneumonia, coronary heart disease and stroke. In only a small proportion of deaths is influenza recorded as the underlying cause.

The seasonal increase in mortality in the winter has been defined as the difference between the number of deaths in the four ‘winter’ months (December to March) and the average of the numbers of deaths in the two four month periods which precede winter (August to November) and follow winter (April to July).

Statistics of the seasonal increase in mortality in the winter inform public debate and the development of government policy on matters such as the health of the elderly population, fuel poverty and whether there is a need to improve the housing stock in terms of central heating and thermal insulation.

Key documents:

The annual statistics are available here from the NRS. The increased winter mortality (Excess Winter Deaths) figure for Scotland in 2018/19 was 2,060.

The archive on increased winter mortality is available here.

The Occasional Paper (published in April 2002) on “The Raised Incidence of Winter Deaths” is here

Key facts:

  • There were 20,188 deaths registered in Scotland in the four months of winter 2018/19 (December to March). Comparing this with the average for the two adjacent 4-month periods, the seasonal increase in mortality in winter 2018/19 was 2,060. This was 2,750 fewer than the corresponding value of 4,810 for the previous winter (2017/18).
  • The seasonal increase in mortality has been calculated for every winter from 1951/52. The seasonal increase of 2,060 in winter 2018/19 was the seventh lowest in those 68 winters, and below the level seen in seven of the previous ten winters, and in 15 of the previous 20 winters.
  • The long-term trend in the seasonal increase in mortality in the winter has clearly been downward, but it can fluctuate greatly from winter to winter. There have been unusually large figures in some years (including, latterly, 4,060 for winter 2014/15 and 4,810 for winter 2017/18, which was the largest value since winter 1999/2000). However, the height of the peaks has generally been falling.
  • The five year moving average, which smooths out much of the year to year fluctuation, had tended to decline, but has risen recently (due, for example, to winters 2014/15 and 2017/18 entering the calculation). While the latest value (3,300) is the twenty-sixth lowest ever, it is greater than all of the previous 15 values and clearly above the level that had applied since the early 2000s. However, it is too early to say whether there has been a change in the long-term trend: the recent rise in the 5 year moving average could just be a short-term one, like that which was seen around the middle of the 1990s, after which the 5-year moving average fell for several years.
  • The latest 19 winters have had eight out of the ten lowest seasonal increases in mortality ever recorded. Over the 68 years covered by these statistics, the lowest seasonal increase in mortality was for winter 2011/12 (1,420), the second lowest was for winter 2013/14 (1,600), the third lowest was for winter 2005/06 (1,780), the fourth lowest was for winter 2001/02 (1,840), and the fifth lowest was for winter 2012/13 (2,000). The most recent winter (2018/19) had the seventh lowest value (2,060). The ninth and tenth lowest seasonal increases in mortality were for winter 2007/08 (2,180) and winter 2000/01 (2,220).
  • On the basis of the Increased Winter Mortality Index, winter mortality in Scotland is lower than in England & Wales.

A separate document, Increased Winter Mortality – Background Note, published in October 2010 (available on the NRS website), gives information about some of the medical causes of the seasonal increase in mortality in the winter, describes some research studies’ findings on factors that influence it, reports on a comparison of the figures for a number of European countries, mentions previous publications on this topic, and provides references to the sources of the material.

The main points to note are:

  • high cold-related mortality is associated with low indoor temperatures, and with people not wearing appropriate clothing when outdoors in cold weather;
  • increased winter mortality was at the same level in Scotland as the overall mean for the 14 European countries covered by a comparative study; and
  • the seasonal increase in mortality in the winter is higher in countries with a warmer winter climate, probably because their homes tend to be poorly insulated and their populations tend not to dress well for cold weather