Winter is approaching and many people are preparing by getting their boilers serviced, wearing warmer clothing and getting their flu jab. While colder weather can be a risk to health for older people and those with underlying health conditions, this article shows in recent years, higher levels of excess winter mortality are also related to the virulence of circulating flu strains.

The number of excess winter deaths in 2014/15 was the highest of the last 15 years

Excess winter mortality, the difference between the average number of deaths over the winter and the rest of the year, has decreased significantly since the 1950s. This is thanks to improvements in healthcare, home insulation and the introduction of the influenza vaccination programme. However, after a drop in 2013/14, deaths increased substantially in 2014/15.

Number of excess winter deaths, England and Wales, 1995/96 to 2014/15

Embed code
Download the data

Between December 2014 and March 2015 there were 44,000 excess winter deaths, 2.5 times higher than the record low of the previous winter, and the highest number since the winter of 1999/2000 when flu levels were very high.

Daily deaths peaked on 1 January 2015, 35% higher than the five-year average

Deaths peaked on 1 January last winter when daily deaths were 35% higher than the five-year average. The first part of 2015 (5 to 11 January) also saw weekly deaths at 15,000, the highest number in any given week since the last two weeks of December 1999 and first two weeks of January 2000, when flu levels were very high.

Daily deaths were above the five-year average on 304 out of 365 days in 2014/15. There were only two days during the winter period where daily deaths fell below the five year average, and on both occasions, the difference was less than fifteen deaths.

Number of daily deaths and 5-year average daily deaths, England and Wales, August 2014 to July 2015

Embed code
Download the data

Temperature was a bigger factor in excess winter deaths back in the 1950s and 1960s

Despite the rise in excess deaths last winter, recent times have seen excess deaths well below that of the 1950s and 1960s. Over the winter periods between 1950/51 and 1965/66 there were on average 62,000 excess winter deaths, more than twice the average over the winter periods between 1998/99 and 2013/14.

Average number of excess winter deaths, England and Wales, 1950/51 to 2014/15

Embed code
Download the data

Looking at 15-year averages from 1950/51 through to 2013/14, it is clear that excess winter deaths have shown a decreasing trend. However, the number of excess winter deaths in 2014/15 was far higher than the average seen in the previous 15 years.

Between 1950 and 1965 there was a strong relationship between the average winter temperature and winter deaths. Analysis of data from that era suggested that there were more deaths in colder winters than in recent years.

A large part of this may be explained by home improvements meaning that homes are easier to heat and keep warm now than they were half a century ago leading to more stable indoor temperatures.

In the 1950s and 60s, the majority of houses did not have central heating, and were instead kept warm by fireplaces and stoves. Today, over 90% of houses have central heating. In 2013, 80% of homes had full double glazing, up from 30% in 1996; and approximately 9.6 million dwellings had cavity wall insulation in 2013, up from less than 3 million in 1996.

Respiratory diseases such as flu are a major cause of excess winter deaths

Influenza is a respiratory disease caused by a viral infection affecting the lungs and airways, and is associated with potentially life-threatening complications such as bacterial pneumonia. This means although flu itself is not always mentioned on the death certificate, it is a contributing factor in a great deal more.

In the UK, pneumonia affects 1 in 1,000 adults each year, and was the underlying cause of 19% of all excess winter deaths in 2014/15, but was a contributory factor in many more. The predominant strain of flu in 2014/15 was A(H3N2). There were moderate levels of flu circulating in England and Wales in 2014/15, but excess mortality levels were far greater than the last notable A(H3N2) season in 2008/09.

Weekly number of deaths and Influenza-Like Illness incidence rate (ILI), England and Wales, 1999/00 to 2014/15

Embed code
Embed code
Download the data

Early estimates in the winter 2014/15 flu season suggested the flu vaccine was only effective in around 3 in 100 cases. However, updated estimates by Public Health England showed the effectiveness to be closer to 34%. In previous years the vaccine has been around 50% effective, coinciding with a far lower number of excess winter deaths.

During the most recent winter, fewer people went to see their GP about Influenza-like illness than in 2010/11, when there was a sharp peak in late December 2010-early January 2011.

Yet mortality in 2014/15 was much higher.

This could be for a number of reasons. The 2010/11 flu vaccine was estimated at 50% effective, far higher than the 34% effectiveness seen in 2014/15.

In addition to this, the predominant flu strain in 2010/11 was A(H1N1), whereas in 2014/15 it was A(H3N2).

In 2014/15 influenza A(H3N2) was particularly virulent in elderly people, an already at-risk group, and resulted in numerous care home outbreaks. By contrast, the 2010/11 flu strain was particularly virulent in younger people and has less of an impact on the elderly.

Despite the lower effectiveness of the flu vaccine in 2014/15, flu jab remains the best way to protect yourself and your family from flu. For more information about the flu vaccine, take a look at the winter 2015/16 flu vaccine information booklet and flu plan.

For more information, please see the full release on Excess Winter Mortality or contact: mortality@ons.gsi.gov.uk