National Records of Scotland

Preserving the past, Recording the present, Informing the future

Probable Suicides: Deaths which are the Result of Intentional Self-harm or Events of Undetermined Intent

Probable Suicides: Deaths which are the Result of Intentional Self-harm or Events of Undetermined Intent

Main points

Overall number, and trends

  • There were 728 probable suicides registered in Scotland in 2016, 56 (8%) more than in the previous year. These figures are based on the new coding rules that apply in Scotland with effect from 2011 (more information below). It is estimated that only 697 of these deaths would have been counted as probable suicides under the old coding rules: 41 (6%) more than the corresponding estimate for 2015. Further information can be found in Table 1.
  • Chart 1 shows the number of probable suicides in each year from 1974, using the old coding rules for 2011 onwards. From time to time, there have been big changes in the figures (like the large drop from 746 in 2013 to 659 in 2014), some of which have been followed by a large change in the opposite direction. Because the number of probable suicides may fluctuate from year to year, the chart also shows the 5-year moving annual average (which uses the old coding rules for 2011 onwards), as an indication of any overall trend, and the likely range of statistical variability around it (which is explained in the Definition of the statistics section on this web site). Some clear trends can be seen, despite what may be large percentage fluctuations in the numbers for individual years. In particular, the 5-year moving annual average rose fairly steadily from about 750 (in the period centred on 1990) to a peak of slightly under 900 (in the period centred on 2000). Since then, the 5-year moving annual average has fallen back to a level that was last seen in the early 1980s: its latest value (of 704, for the period centred on 2014) is the lowest value since that seen for the period centred on 1977 (which was 698).
  • As there may be large percentage fluctuations in individual years’ numbers, the figures for a period of a few years may not reflect the overall trend. For example, although there has been a general downward trend since around the start of the new century, year to year fluctuations during much of the first decade led to little change in the 5-year moving annual averages centred on the years from 2004 to 2009 (all were between 791 and 811). More recently, there has been a clearer downward trend with the 5-year moving average (using the old coding rules for 2011 onwards) falling from 796 for the period centred on 2009 to 704 for the period centred on 2014 . It should be noted that, despite being 6% more than in 2015, the ‘old basis’ figure for 2016 (697) is still below the latest 5-year moving average value (704). Further information can be found in Table 1.
  • Over the longer-term, broadly speaking, the annual number of probable suicides (using the old coding rules) was roughly 650 in the mid-1970s, rose during the rest of the decade, was around 700-750 during the 1980s, increased in the 1990s to almost 900 at the start of the new century, then fell: it was about 750-800 between 2009 and 2013, then dropped to just over 650 in 2014 and 2015 before the rise in 2016 (further information can be found in Table 1). Again, Chart 1 shows that there were many year-to-year fluctuations: over the period covered by the figures, they have ranged from 642 in 1974 to 912 in 1993.
  • In 2009, how National Records of Scotland (NRS) obtains information about the nature of death changed. Since then, there has been a large increase in the percentage of poisoning deaths described as accidental, and a fall in those described as being due to events of undetermined intent. This caused part of the fall in the number of probable suicides between 2010 and 2015: more information is available below.

Probable suicides by sex and age

  • Roughly three-quarters of all probable suicides are men: 71% in 2016 and between 70% and 77% in every year from 1986 (further information can be found in Table 1).
  • The likelihood of suicide varies with age. In 2016, using the figures on the basis of the new coding rules, the 45-49 year old age-group had the largest number of probable suicides (93, or 13%), followed by 40-44 year olds (86 or 12%), people aged 50-54 (85, or 12%), the 25-29 age-group (69 or 9%), 35-39 year olds (62, or 9%), and those aged 30-34 (61, or 8%). However, the number of suicides by age fluctuates from year to year. Using the figures based on the old coding rules, the largest numbers of suicides over the latest five years have been in the following age-groups: 45-49 (93 per year, on average); 40-44 (85 per year, on average); 50-54 (79 per year, on average); 35-39 (66 per year, on average); 55-59 (63 per year, on average); 30-34 (60 per year, on average); and 25-29 (59 per year, on average). The pattern has changed over the years. In the second half of the 1990s the largest numbers were in the following age-groups: 25-29, 30-34 and 35-39, for which the annual averages (over the period from 1995 to 1999) were 108, 106 and 93, respectively. The corresponding figure for 45-49 year olds was only 78 (further information can be found in Table 3M and Table 3F.

Area of residence

  • Only a couple of per cent of the probable suicides in Scotland each year are people whose usual residence was outwith Scotland (further information can be found in Table 2).
  • Tables 4 and 5 give figures for each Health Board and council area, which can fluctuate markedly from year to year, so the tables include 5-year moving annual averages, which should indicate better any overall trend. 

Method of suicide

  • In 2016, using the figures based on the new coding rules, the most common method of suicide was ‘hanging, strangulation and suffocation’ (49%) followed by ‘poison’ (28%). In addition, 8% killed themselves by jumping or falling from a high place, 5% by drowning or submersion, and under 1% used firearms or explosives. However, the figures for 2016 using the old coding rules have a greater gap between the percentages for ‘hanging, strangulation and suffocation’ (51%) and ‘poison’ (24%).
  • Methods of suicide have changed over the years: in the 1970s, over half took poison, on average only about 13% hanged themselves, and almost a fifth drowned (further information can be found in Table 2).
  • As explained in the definition of the statistics section on this web site, 'probable suicides' are deaths which are believed to be due either to intentional self-harm or to events of undetermined intent. Chart 2 shows how the numbers for each of these ‘natures of death’ have changed: for example, each of the past ten or so years had roughly 500 to 600 deaths from intentional self-harm plus a number of deaths from events of undetermined intent which has fluctuated greatly in percentage terms (for example, the ‘undetermined intent’ figures for 2011 and 2016, respectively, were 245 and 94 based on the old coding rules, and were 362 and 125 based on the new coding rules). Table 2b shows the numbers for each nature of death broken down by the method of death. In the latest ten-or-so years, ‘hanging, strangulation and suffocation’ was the cause of most of the deaths from intentional self-harm, whereas ‘poison’ has been the main cause of most of the deaths from events of undetermined intent. The right hand side of Table 2b shows that the main reason for the fluctuations in the figures for undetermined intent deaths over the years since 1974 is large rises and falls in the number which were due to poisoning: in comparison, there are relatively few undetermined intent deaths by other methods, and their numbers have not changed as much.

The change since 2009 in how NRS obtains information about the nature of death

  • Procurators Fiscal (PFs) have a responsibility to investigate all sudden, suspicious, accidental and unexplained deaths. This includes all probable suicides that are reported to a PF. How NRS classifies the nature of a traumatic or suspicious death registered in Scotland is usually informed by the view of the PF who, at the conclusion of the investigation, will notify NRS as to whether such a reported death was due to an accident, assault, intentional self-harm or undetermined intent. The last category should be specified in cases 'where the evidence is insufficient for the PF to form a view, on the balance of probabilities, as to which of the other categories is appropriate'. The current procedure is described briefly at Mid-2009 Change in the Procedure Used to Inform NRS About Suicides on our website. It was introduced in 2009. Since then, there has been a large increase in the percentage of poisoning deaths which PFs have described as being the result of accidents, and a corresponding fall in the proportion which they have described as being due to events of undetermined intent. This contributed to the fall in the number of probable suicides between 2010 and 2015. However, it did not cause the whole of that fall: in 2016, NRS estimated that, had there been no variation in the percentage of poisoning deaths classed as accidental, there would still have been fewer probable suicides (based on the old coding rules) in 2015 (and 2014) than in the previous five or more years, and that the number in 2015 would still have been well below the level seen around the year 2000. A separate note on How the statistics may have been affected by changes in the views of Procurators Fiscal on the nature of deaths provides more information about this. 

The change to the coding rules for deaths which were registered from the start of 2011

  • As indicated earlier, new rules for coding the causes of death mean that the overall total numbers of probable suicides for 2010 and 2011 are not directly comparable. How ‘drug abuse’ deaths from ‘acute intoxication’ are coded has changed: in 2010 and earlier years, they were counted under ‘mental and behavioural disorders due to psychoactive substance use’; now they are classified as ‘poisoning’, so some of them will be counted as probable suicides. This follows a World Health Organisation (WHO) update to the International Statistical Classification of Diseases and Related Health Problems, which NRS implemented at the start of 2011. A note on the changes to the way in which causes of death are coded is available in the Death Certificates and Coding the Causes of Death section on this website. Examples of the cases concerned are people who were known or suspected habitual drug abusers, for whom the cause of death was certified as ‘adverse effects of heroin’, ‘methadone toxicity’ or ‘morphine intoxication’. Unless NRS has been informed that the overdose was believed to be accidental or deliberate, it classifies such deaths as being due to ‘poisoning by …, undetermined intent’, which means that they will be counted as probable suicides.
  • NRS has estimated what the figures for 2011 onwards would have been, had the data been coded using the old rules, in order that users of these statistics can see the underlying changes between 2010 and 2011 and the longer-term trends without the break in series caused by the introduction of the new coding rules. NRS hopes to continue to estimate the number of probable suicides on the basis of the old coding rules for at least a few more years.
  • Since 2011, there has been a marked fall in NRS’s estimate of the number of deaths that are counted as probable suicides using the new coding rules, but would not have been counted as such under the old coding rules (this is the new coding rules estimate minus the old coding rules estimate): from 117 in 2011, through 68 in 2012, 49 in 2013 and 37 in 2014 to 16 in 2015, although it increased to 31 in 2016. This is a consequence of the reduction in the proportion of poisoning deaths that PFs have described as being due to events of undetermined intent, which was mentioned earlier.
  • The fact that the difference between the ‘old coding rules’ and ‘new coding rules’ figures got smaller (over the years from 2011 to 2015) caused the ‘new coding rules’ figures to fall faster than the ‘old coding rules’ ones (over that period). For example, of the deaths that were registered in 2011, NRS estimates that 117 more were counted as probable suicides under the new coding rules than would have been counted under the old coding rules. The corresponding figure for 2012 is 68, which is 49 fewer - so the number of deaths which are only counted under the new coding rules fell by 49 between 2011 and 2012. That is why the drop between 2011 and 2012 in the number of probable suicides is 49 greater for statistics based on the new coding rules than for figures based on the old rules. The table below shows that there was a similar (but smaller) effect on the changes between later years (ending in 2015): in each case, the drop in the number of probable suicides was greater for the figures based on the new coding rules than for the numbers based on the old coding rules. However, both figures increased in 2016, with the rise in the number of probable suicides being greater for the figures based on the new coding rules than for the numbers based on the old coding rules. 
Year

Old coding rules figure

Old coding rules change from previous year

New coding rules figure

New coding rules change from previous year

2011

772

 

889

 

2012

762

-10

830

-59

2013

746

-16

795

-35

2014

659

-87

696

-99

2015

656

-3

672

-24

2016 697 +41 728 +56