Suicides in the UK: 2011 registrations

• In 2011 there were 6,045 suicides in people aged 15 and over in the UK, an increase of 437 compared with 2010

• The UK suicide rate increased significantly between 2010 and 2011, from 11.1 to 11.8 deaths per 100,000 population

• There were 4,552 male suicides in 2011 (a rate of 18.2 suicides per 100,000 population) and 1,493 female suicides (5.6 per 100,000 population)

• The highest suicide rate was in males aged 30 to 44 (23.5 deaths per 100,000 population in 2011)

• The suicide rate in males aged 45 to 59 increased significantly between 2007 and 2011 (22.2 deaths per 100,000 population in 2011)

• Female suicide rates were highest in 45 to 59-year-olds in 2011 (7.3 deaths per 100,000 population)

This bulletin presents the latest suicide statistics for the UK, England (including figures for regions) and Wales. New figures are presented for 2011 with a back series of data covering 1981 to 2010 to allow comparison. The potential impact of the use of narrative verdicts by coroners on suicide rates at regional level in England, and in Wales, is also presented. A brief discussion of registration delays in relation to suicide statistics is also included.

In 2011 there were 6,045 suicides in the UK. The suicide rate was significantly higher in 2011, compared with 2010 (11.8 and 11.1 deaths per 100,000 population respectively), and was the highest rate since 2004.

Of the total number of suicides, 4,552 were males and 1,493 were females. The age-standardised suicide rates in 2011 were 18.2 and 5.6 deaths per 100,000 population for males and females respectively. Suicides rates have been consistently lower in females than in males over the past three decades.

The suicide rate was highest in middle-aged men (those in the 30 to 44-year-old and 45 to 59-year-old age groups). In particular, the suicide rate in males aged 45 to 59 has increased significantly in the last five years to 22.2 deaths per 100,000 population.

Males

In 1981 there were 4,129 male suicides (an age-standardised mortality rate of 19.8 deaths per 100,000 population). In 2011 the number of suicides was slightly higher at 4,552, but the suicide rate had fallen significantly to 18.2 deaths per 100,000 population. However, the suicide rate did not decline consistently over this period, instead fluctuating from year to year.

Male suicide rates increased in the 1980s, and peaked at 21.9 deaths per 100,000 population in 1988. Suicide rates tended to decrease between 1988 and 2010, though there were some annual rises, (for example, higher rates were seen in 1998 and 1999). After more than a decade of falling suicide rates in males (with the exception of 2008), the rate increased significantly between 2010 and 2011 (from 17.0 to 18.2 deaths per 100,000 population). This is the highest male suicide rate since 2002.

Females

In 1981 there were 2,466 female suicides (an age-standardised mortality rate of 10.4 deaths per 100,000 population). This was the highest rate seen in the 31-year period covered by this bulletin. By 2011 the number of female suicides had fallen to 1,493 (5.6 deaths per 100,000 population). However, as with males, there have not been consistent year-on-year decreases in female suicides.

The female suicide rate declined steadily between 1981 and 1994 from 10.4 to 6.1 deaths per 100,000 population, and then remained relatively stable between 1994 and 2004. The rate then declined for three years in a row, to reach an all-time low of 5.0 suicides per 100,000 population in 2007. Since then, female suicides have tended to increase (with the exception of 2009 when there was a small decrease), and the suicide rate of 5.6 deaths per 100,000 population in 2011 was significantly higher than in 2007.

Several changes were introduced in 2011 that could have affected suicide statistics.

Narrative verdicts

Narrative verdicts are a factual record of how, and in what circumstances, the death occurred. They are sometimes returned by coroners where the cause of death does not easily fit any of the standard ‘short-form’ verdicts (more details can be found in the ‘Narrative verdicts in England and Wales’ section below). Additional guidance was provided to the ONS coding team in order to improve the coding of narrative verdicts in England and Wales. Common phrases used by coroners in narrative verdicts were identified and added to terms allowed for the classification of intentional self-harm in the existing coding frame. Prior to 2011 some deaths with a narrative verdict were coded as accidents. This is because ONS applies the ICD rules for coding cause of death such that where no indication of intent has been given by the certifier, deaths from injury or poisoning must be coded as accidents.

Improvements were also made to allow better identification of narrative verdicts, as prior to 2011 ONS could only identify hard-to-code narrative verdicts (see below for more information on hard-to-code narratives).

In October 2011 an advice note was issued to coroners in England and Wales that provided guidance on the information that should be included in a narrative verdict to help ONS code cause of death using the International Classification of Diseases (ICD). In potential self-harm cases, coroners were advised that the description of the circumstances should make clear the intention of the action that led to death (for example, whether it was deliberate self-harm rather than an accident). In addition, the note advised that where the circumstances of death met the legal criteria for a suicide verdict, then a suitable short form verdict should be used.

This additional guidance could have resulted in an increased number of narrative verdicts coded as intentional self-harm in 2011, when previously they may have been coded as accidents because of a lack of information about intent. This could be contributing to the apparent increase in the suicide rate in 2011.

ICD-10 rule changes

In 2011 a new version of the ICD-10 software (version 2010) was introduced in the UK, which included a rule change that affected deaths coded as an event of undetermined intent. In the previous versions of ICD-10 (software version 2001.2 in England and Wales, version 2006 in Scotland, and version 2003 in Northern Ireland) deaths mentioning both acute poisoning and drug dependence would be assigned an underlying cause of a mental and behavioural disorder due to psychoactive substance use (an ICD 10 F code), so would not be included in suicide statistics. However, in software v2010 this causal sequence is no longer valid, so the acute poisoning is selected as the underlying cause of death. The acute poisoning may be accidental (ICD-10 codes X40–X49) or a poisoning of undetermined intent (Y10–Y19), depending on the intent. As events of undetermined intent are included in the National Statistics suicide definition, this ICD rule change could have potentially increased the number of deaths included in ONS suicide statistics.

To understand the impact of the introduction of ICD-10 v2010 on mortality statistics in England and Wales, ONS carried out a bridge coding study in which a sample of deaths that had previously been coded using v2001.2 were then independently recoded using the new version of ICD-10 (Office for National Statistics, 2011).

The impact of the new version of ICD-10 on suicide figures was not reported in the bridge coding study, but analysis presented here shows that the new version of ICD-10 caused no change in the number of deaths being coded as intentional self-harm (this is to be expected as the rule change did not affect intentional self-poisoning deaths) and a 2% increase in the number of deaths coded as an event of undetermined intent. However, not all of the information provided by coroners at registration was available to use when recoding deaths, so the bridge coding study results for suicides should be treated with caution.

The ICD-10 changes do appear to have caused a large increase in the number of deaths coded as an event of undetermined intent in Scotland (245 estimated deaths under the old ICD-10 rules and 362 deaths under the new rules). The increase in Scottish figures will obviously also have resulted in an increase in the overall number of UK suicides. The differing impact in England and Wales and Scotland may be due to differences in the death registration systems used in each country. More information can be found on the National Records for Scotland website.

Males

Between 2010 and 2011, the male suicide rate increased in every age group except those aged 75 and over, although very different trends have been seen across the age groups since 2001.

Between 2001 and 2011 the suicide rate was consistently highest among males aged 30 to 44 compared with all other age groups. In 2001 the suicide rate in this age group was 23.4 deaths per 100,000 population, which was significantly higher than the rate in any other age group. The suicide rate in this age group has not changed significantly in the last 11 years, and the rate was 23.5 deaths per 100,000 population in 2011.

The suicide rate in males aged 45 to 59 has increased significantly since 2007, and by 2011 was almost as high as the rate in the 30 to 44-year-old age group (22.2 deaths per 100,000 population). This is the highest suicide rate in 45 to 59-year-olds since 1986.

The latest suicide prevention strategy for England (Department of Health, 2012) identified middle-aged men as one of the high-risk groups who were a priority for suicide prevention. A recent report by the Samaritans suggested that middle-aged men, especially those from poorer socio-economic backgrounds are particularly at risk of suicide due to a combination of factors. These include social and cultural changes (for example, rising female employment and greater solo living) that have particularly impacted on the lives of the cohort of men who are now in mid-life (Samaritans, 2012).

In 2011 there were 12.7 suicides per 100,000 population in males aged 60 to 74, and this rate has not changed significantly since 2001. This is relatively low compared with males aged 30 to 59.

The suicide rate in older men (those aged 75 and over) has shown the opposite trend to middle-aged men, as the rate declined significantly between 2004 and 2011. Moreover, those aged 75 and over were the only age group where the suicide rate fell slightly in 2011 (from 14.8 suicides per 100,000 population in 2010 down to 13.8 in 2011). The suicide rate in males aged 75 and over is now low compared with historical levels (see Reference Table 5 (587 Kb Excel sheet)) and with other age groups.

The suicide rate in younger males (those aged 15 to 29) decreased significantly between 2001 and 2006, and despite some annual fluctuations the rate has not changed significantly since then (13.3 deaths per 100,000 population in 2011). Moreover, the suicide rate in the youngest age group is still relatively low compared with middle-aged men.

Females

In 2011 the highest female suicide rate was in 45 to 59-year-olds, and the lowest suicide rate was in 15 to 29-year-olds. Between 2010 and 2011 the rate increased slightly in every age group, except for women aged 60 to 74, where there was a slight decrease.

Historically, the suicide rate in middle-aged and older women (those aged 45 and over) was much higher than in younger women. Reference Table 6 (587 Kb Excel sheet), (available to download from the ONS website) shows that suicide rates in these middle-aged and older women declined sharply in the 1980s and early 1990s. Suicide rates in women aged 60 to 74 continued to fall, but rates in other age groups tended to level off in the late 1990s.

The suicide rate in females aged 75 and over remained relatively stable between 2001 and 2004 (though there were some annual variations). The rate then declined significantly between 2004 and 2007, but has remained relatively stable since then (4.8 suicides per 100,000 population in 2011).

The suicide rate in women aged 60 to 74 tended to fall between 2001 and 2011, and was the only female age group to show a slight decline in the suicide rate between 2010 and 2011 (from 4.7 to 4.6 deaths per 100,000 population).

Since 2003 the 45 to 59-year-old age group has had the highest female suicide rate, compared with other age groups. The suicide rate in this age group increased slightly between 2001 and 2005 and then decreased slightly between 2005 and 2009. However, in the last two years the rate increased slightly again, reaching 7.3 suicides per 100,000 population in 2011 – the highest rate since 2006.

In 2011 the second highest female suicide rate was in 30 to 44-year-olds. The rate in this age group reached an all-time low of 5.3 suicides per 100,000 population in 2006, but increased to 6.4 suicides per 100,000 by 2011.

Females aged 15 to 29 had the lowest suicide rates of any age group. However, the rate in this age group has increased significantly from a record low of 2.9 deaths per 100,000 in 2007 to 4.2 deaths per 100,000 in 2011.

England

There were 4,509 suicides among people aged 15 and over in England. The age-standardised suicide rate increased by 6% between 2010 and 2011, from 9.8 to 10.4 deaths per 100,000 population.

Hospital admissions for intentional self-harm or an event of undetermined intent

In 2010/11 and 2011/12 there were more than 114,000 inpatient hospital admission episodes each year in England for intentional self-harm or an event of undetermined intent (The Health and Social Care Information Centre, 2011 and 2012). This represents a 7% increase compared with 2009/10, when there were just over 107,000 similar admissions (The Health and Social Care Information Centre, 2010). This is broadly in line with the increased suicide rate in England in 2011.

Wales

In Wales there were 341 suicides in those aged 15 years and over in 2011 (270 males and 71 females). In the last two years the suicide rate has increased by 30% from a low of 10.7 deaths per 100,000 population in 2009 to 13.9 per 100,000 in 2011. This is the highest suicide rate seen in Wales since 2004, and the increase was seen in both males and females.

Between 1981 and 1990 the male suicide rate was fairly similar in England and Wales, but from 1991 onwards the rate has been higher in Wales (28% higher in 2011). However, a different picture is seen for females: between 1981 and 1996, the female suicide rate tended to be lower in Wales than in England, but from 1997 onwards there has been no consistent pattern.

English regions

In 2011 the suicide rate was highest in the North East at 12.9 deaths per 100,000 population and lowest in London at 8.9 per 100,000 (see Table 1).

The suicide rate fell in two regions in between 2010 and 2011 (West Midlands and London), and rose in seven regions (South West, South East, North West, East of England, East Midlands, North East and Yorkshire and the Humber). The largest increase was in Yorkshire and the Humber, where the suicide rate increased by 21% in 2011.

Reference Table 11 (587 Kb Excel sheet) shows that over the last 10 years (2002 to 2011) male suicide rates have tended to be highest in the North East and the North West and lowest in London and the East of England. Females suicide rates have tended to be highest in the North West and the South West, and lowest in Yorkshire and the Humber and the West Midlands.

Results for Scotland and Northern Ireland

Suicide figures for Scotland are produced by National Records of Scotland (formerly the General Register Office for Scotland) and can be found on their website.

Suicide figures for Northern Ireland are produced by the Northern Ireland Statistics and Research Agency and can be found on their website.

Annually, there are around 30,000 coroner’s inquests held in England and Wales that conclude with a verdict. Around 90% of these inquests conclude with a 'short form' verdict such as accident, misadventure, natural causes, suicide or homicide. Narrative verdicts can be used by a coroner or jury instead of a short form verdict to express their conclusions as to the cause of death following an inquest.

A narrative verdict can be given in a range of different circumstances, and for a variety of causes of death. Table 2 shows that almost half of all narrative verdicts in England and Wales in 2011 were where the underlying cause of death was some type of disease, most commonly a circulatory disease or a type of cancer (a neoplasm). For example, a narrative verdict is sometimes given if the deceased died from mesothelioma (an industrial disease which is often associated with exposure to asbestos in the workplace), but the source of the asbestos exposure could not be determined. Other examples of narrative verdicts include instances where the deceased may have suffered an accidental fall as a result of an underlying health condition, or where the deceased died whilst having medical treatment for an underlying health condition.

Just over half of all narrative verdicts in England and Wales in 2011 resulted from injury or poisoning (that is, as an ‘external’ cause of death), rather than a disease. Some of these narrative verdicts clearly state the intent (for example, accidental) and mechanism (for example, hanging, poisoning) of death. However, in some cases the coroner does not indicate whether the fatal injury was accidental or if there was deliberate intent to self-harm. These latter deaths are defined by ONS as ‘hard-to-code’.

ONS applies the rules of coding cause of death such that where no indication of intent has been given by the certifier, deaths from injury or poisoning must be coded as accidents. Therefore, unsurprisingly, Table 2 shows that about 40% of narrative verdicts are assigned an accidental cause of death.

An examination of a sample of hard-to-code narrative verdicts that ONS had coded as accidents (based on ICD coding rules) in 2011, found that between 40 and 50% of the deaths were where people had fallen in their home or in a care setting. There was no suggestion that these deaths were anything other than an accident. The remaining deaths had a variety of causes including poisonings and hangings. While many of these other deaths are likely to be accidents, suicide researchers are concerned that cause coding rules force ONS to code some probable suicides as accidents (Gunnell et al, 2011), meaning that official suicide figures may be an underestimate. These concerns have grown in recent years, as the number of narrative verdicts has increased.

Between 2001 and 2010 there were large year-on-year increases in the number of narrative verdicts returned by coroners in England and Wales (Reference Table 13). In England, there were almost double the number of narrative verdicts in 2010 (3,170) than there were in 2006 (1,592). In Wales, the number increased almost three-fold over the same period from 52 in 2006 to 147 in 2010. This trend is supported by figures from the Ministry of Justice showing that the number of unclassified verdicts (including narrative verdicts) in England and Wales increased from 2,406 in 2006 to 4,180 in 2010.

Moreover, there is considerable variation in the use of narrative verdicts between coroners. Until 2011 the region with the highest proportion of narrative verdicts (out of all inquests that returned a verdict) was the West Midlands. In 2010 around 20% of inquests in the West Midlands resulted in a hard-to-code narrative verdict compared with only 8% in the following regions: South West, North East, South East, London and Wales. This regional variation led to concerns that if the use of narrative verdicts continued to increase, it could affect the reliability of ONS suicide statistics – especially at a sub-national level.

To help improve the accuracy of suicide statistics, in January 2011 additional guidance was given to the ONS coding team in order to improve the coding of narrative verdicts. ONS also provided advice to coroners on the types of narrative verdicts that are hard to code, elements of which the Local Government Committee of the Coroners Advisory Group incorporated into an advice note issued in October 2011. This advice note is unlikely to have had a big impact on suicide statistics in 2011, as most of the deaths would have been registered before the note was issued. ONS will continue to monitor the impact in the future.

Prior to 2011, ONS could only identify hard-to-code narrative verdicts within our mortality database, so figures for the years 2001 to 2010 reported in Reference Table 13 (587 Kb Excel sheet) do not show the total number of narrative verdicts. In 2011 ONS made improvements to enable the identification of all narrative verdicts. Reference Table 13 (587 Kb Excel sheet) shows that in 2011 the total number of narrative verdicts fell by 12% in England (2,796 narrative verdicts), but increased by 5% in Wales (155 narrative verdicts). However, Ministry of Justice figures suggest that the number of unclassified verdicts (including narratives) increased by 5% in England and Wales between 2010 and 2011 (from 4,180 to 4,400). This is likely to be due to differences in the ONS definition of a narrative verdict and the MoJ definition of an unclassified verdict, as there is no fixed definition of a narrative verdict.

Between 2010 and 2011 the number of hard-to-code narrative verdicts decreased by 46% in England and by 49% in Wales. Every region in England showed a decrease in the number of hard-to-code narrative verdicts. The largest decrease was in the North East, which saw a 71% fall; and the smallest decrease was in Yorkshire and the Humber, which only saw a 5% fall.

Hence, the additional guidance given to ONS coders and coroners appears to have significantly reduced the number of hard-to-code narrative verdicts, thus improving the reliability of ONS suicide statistics.

An analysis to assess the impact of narrative verdicts on suicide rates in England and Wales was undertaken by ONS in 2011 (Hill and Cook, 2011). Simulated age-standardised suicide rates were calculated for the years 2001 to 2009 using two different assumptions:

Scenario 1: suicide rates were calculated assuming all deaths where a hard-to-code narrative verdict meant that the death been coded as an accidental hanging (ICD-10 codes W75–W76) or accidental poisoning (ICD-10 codes X40–X49) were intentional self-harm.

Scenario 2: suicide rates were calculated assuming that half of these deaths were intentional self-harm.

Hangings and poisonings were used as these are the two most common methods of intentional self-harm in England and Wales.

The results showed that between 2001 and 2009 there were no statistically significant differences between the published and simulated suicide rates at national level. ONS has now repeated the above analysis using the latest figures for regions of England, and for Wales (see Background Note 6).

Table 1 above shows the existing suicide rates for regions of England, and Wales, for 2011. Table 4 shows results of combining all accidental hangings and accidental poisonings from hard-to-code narrative verdicts with existing suicide rates (Scenario 1), for regions of England, and Wales, for 2011.

In England in 2011, when hard-to-code narrative verdict deaths with an underlying cause of accidental hanging or accidental poisoning were added to existing suicides, the male suicide rate increased from 16.1 to 16.7 deaths per 100,000 population (a 3% increase). The corresponding female suicide rate rose from 4.9 to 5.2 per 100,000 (a 7% increase). In Wales the suicide rate increased from 22.5 to 23.4 per 100,000 in males (a 4% rise); and from 5.6 to 6.0 per 100,000 in females (a 6% rise).

Within regions, the largest increases in suicide rates (as a result of this simulation) were in the East of England for males (a 7% increase) and the West Midlands for females (a 12% increase). For males there was no difference in the rank ordering of regions based on simulated suicide rates, compared with ranks based on the actual suicide rates. That is, London still had the lowest male suicide rate, and the North East had the highest. For females, the main change was in the ranking of the West Midlands, which was second lowest (out of 9 regions) based on the actual suicide rates, but fourth highest based on the simulated suicide rates (though the difference between some regions was very small).

Although there were apparent increases in simulated suicide rates, and small changes in the rank ordering of regions for females, it is important to note that none of simulated suicide rates were significantly higher than the actual suicide rates. Moreover, the simulated suicide rates presented above represent a ‘worst-case scenario’, as not all accidental hangings and poisonings are suicides. In addition, events of undetermined intent are also included in ONS suicide statistics, and not all of these deaths are caused by self-inflicted harm. Therefore, ONS suicide statistics still provide a reliable estimate of national suicide trends, despite the use of narrative verdicts.

Population estimates for mid-2002 to mid-2010 for England and Wales were revised in December 2012 to take account of the results of the 2011 Census to ensure a consistent time series over the decade. However, they were not available in time to be used in this suicide bulletin. Therefore suicide rates for 2002 to 2010 were calculated using mid-year population estimates for the reference year based on the 2001 Census.

Revised population estimates are due to be published for other geographical areas as follows:

• March- April 2013: Revised population estimates for subnational areas in England and Wales, mid-2002 to mid-2010

• Autumn 2013: Revised population estimates for the UK, mid-2002 to mid-2011

Suicide rates for 2002 to 2010 were calculated using mid-year population estimates for the reference year based on the 2001 Census, and will therefore be revised in future releases. Rates in England and Wales for 2011 were calculated using mid-2011 population estimates based on the 2011 Census, and will not be revised. Provisional UK population estimates for 2011 were produced by combining 2011 census-based England and Wales population estimates, with 2001 census-based population estimates for Scotland and Northern Ireland. UK suicide rates for 2011 will therefore also be revised, but as Scotland and Northern Ireland make up a smaller proportion of the UK, we do not expect these revisions to be large. Consequently, any comparisons between rates for 2002 to 2010, and 2011, should be treated with caution, as suicide rates across the two time periods are not directly comparable due to the planned revisions.

Differences between the 2011 Census and 2011 population estimates based on the 2001 Census which will impact on 2002 to 2010 revisions

A high level reconciliation report explaining national level differences between 2011 Census estimates and population estimates for March 2011 rolled forward from the 2001 Census, was published alongside the first release of 2011 Census data in July 2012. The rolled forward estimates were 476,000 lower than the 2011 Census estimate for England and Wales which splits down into 144,000 males and 332,000 females. The report discusses initial research carried out into the potential causes of this difference.

The most substantial positive differences for males are in the 10 to 19 and 30 to 39 age ranges. The 2011 Census has more males at ages 10 to 19 and 30 to 39 than the rolled forward estimates. However, the opposite is true for males aged 20 to 29, where Census estimates are substantially below the rolled forward estimates. For females, the pattern is slightly different. The Census has considerably more females at ages 10 to 19 and 30 to 39. However, there are also more females at ages 20 to 29. For other ages (0 to 9 and ages 40 and above) the Census estimates and the rolled forward estimates are relatively close for both males and females.

In England and Wales all suicides are certified by a coroner following an inquest. The death cannot be registered until the inquest is completed, which can take many months or even years, and ONS is not notified that a death has occurred until it is registered. The only exception to this is when someone will be charged in relation to the death. In this instance the coroner must adjourn the inquest, and they may carry out an accelerated registration. The full details of these deaths are not recorded until the inquest is completed, but the majority are eventually coded as assaults.

The death registration system in Northern Ireland is similar to that used in England and Wales, in that all suspected suicides are referred to the coroner. The family of the deceased may ask for an inquest and if one has been held, the registrar will register the death on receipt of the coroner's report. If there is no inquest the General Register Office (GRO) will write to the deceased’s family (or other informant) to ask them to register the death. However, if the death is not registered within a year of its occurrence, GRO are able to register the death on the authority of the Registrar General.

In Scotland a death must be registered within eight days. The Procurator Fiscal has a duty to investigate all sudden, suspicious, accidental, unexpected or unexplained deaths and any death occurring in circumstances that give rise to serious public concern, and a Fatal Accident Inquiry may follow. If the results of toxicological tests or a post mortem are not yet known, the cause of death can be given as "unascertained, pending investigations", and the actual cause of death will be entered at a later date. Therefore National Records of Scotland (NRS) receive notification of deaths more quickly than ONS and the Northern Ireland Statistics and Research Agency (NISRA). However, although NRS may know what caused the death (for example, hanging, poisoning), they may not be told whether it was due to an accident, assault or intentional self-harm until after the statistical database has been ‘frozen’ for the year. So NRS may have to code the death as an event of undetermined intent, which would be counted as a probable suicide. Consequently, Scotland has proportionally more deaths coded as being due to events of undetermined intent (and hence as probable suicides), compared with England, Wales and Northern Ireland.

In common with most other UK mortality statistics, suicide figures are presented for deaths registered in a particular calendar year, which enables figures to be published in a timely manner. The alternative would be to publish statistics based on the year in which the death occurred. However, if ONS were to do this the publication would be delayed in order to allow enough time for the majority of the deaths that occurred in a given year to be registered. Moreover, it would be inconsistent with other ONS suicide figures (for example, those published annually in ‘Deaths registered in England and Wales’).

Due to the length of time it takes to hold an inquest, figures presented in this bulletin for England and Wales and Northern Ireland are based on deaths that may have occurred months or even years before 2011. Figure 4 shows that in 2011 the average registration delay for suicides in England and Wales was 158 days. So out of the 4,871 suicides registered in 2011, just under half occurred before 2011. In England and Wales the average registration delay has gradually increased over time; in 2001 the average delay was 103 days, increasing by 53% to 158 days in 2011. However, registration delays have increased more slowly since 2008, suggesting the upward trend is levelling off.

Suicide statistics in Northern Ireland also suffer from significant registration delays, although the trend over time has been more variable than in England and Wales. In Northern Ireland average registration delays for suicides peaked in 2005 at 339 days, but had decreased sharply to 157 days by 2011. The average registration delay for suicides is now virtually the same in England and Wales and Northern Ireland – both just over five months. So of the 286 suicides registered in Northern Ireland in 2011, almost 60% occurred before 2011. Prior to 2004 there were seven coroner’s districts in Northern Ireland. Following a review of the coroner’s service, the separate districts were amalgamated into one centralised coroner’s service. This reorganisation of the coroner’s service may explain the reduction in registration delays in Northern Ireland since 2005.

In 2011 the average registration delay in Scotland was just seven days. Although the registration delay has increased slightly since 2001, 97% of suicides registered in Scotland in 2011 also occurred in 2011.

The differences in the death registration systems in England and Wales, Scotland and Northern Ireland have implications for the comparability of mortality statistics across the UK. That is, the UK suicide figures for deaths registered in 2011 will comprise deaths occurring in different time periods for different countries of the UK. However, as suicide trends tend to change relatively slowly over time, this is unlikely to have a great impact on the usability of UK suicide statistics.

Additional information on registration delays for suicides, including separate figures for males and females, and an indication of the range of registration delays (the lower and upper quartile) can be found in Reference Table 15 (587 Kb Excel sheet). Information on registration delays for a range of causes in England and Wales in 2011 can be found on the ONS website.

Suicide figures for the UK, England and Wales, England, Wales, and regions of England, and results from the analysis of the impact of hard-to-code narrative verdicts returned by coroners in England and Wales, can be found in a Microsoft Excel workbook on the Office for National Statistics website (587 Kb Excel sheet) .

The workbook contains:

• age-specific suicide rates per 100,000 population (with 95% confidence limits) and numbers of suicides: by sex and five-year age group, for the UK, England and Wales, England, and Wales, 1981 to 2011

• age-specific suicide rate for broad age groups (with 95% confidence intervals): for males and females, United Kingdom, 1981 to 2011

• age-standardised suicide rates per 100,000 population (with 95% confidence limits) and numbers of suicides: by sex, for the UK, England and Wales, England, regions of England and Wales, 1981 to 2011

• number of narrative verdicts: by underlying cause of death, England and Wales, 2011

• number of narrative verdicts: by sex, for England and Wales and regions of England, 2001 to 2011

• simulated age-standardised suicide rates per 100,000 population (with 95% confidence limits) from Scenario 1 (see above): by sex, for England and Wales and regions of England, 2001 to 2011

• median registration delays (and the lower and upper quartiles) in England and Wales, Scotland and Northern Ireland, 2001 to 2011

Use of the statistics

Suicide statistics provide an indicator of mental health and are important for monitoring trends in deaths resulting from intentional (and probable) self-harm. The statistics are widely used to inform policy, planning and research in both the public and private sector and they enable policy makers and support services to target their resources most effectively. Key users include the Department of Health and devolved health administrations, public health observatories, local and health authorities, academics, and charity organisations.

Policy context

Each constituent country of the UK has a suicide prevention strategy in place which aims to identify risk factors, take action via cross-sector organisations, and ultimately reduce suicide rates.

In September 2012 the Department of Health launched ‘Preventing Suicide in England: a cross-government outcomes strategy to save lives’. This strategy aims to reduce the suicide rate and improve support for those affected by suicide and was informed by an earlier consultation on preventing suicide in England. The new strategy outlines six areas for action including: reducing the risk of suicide in key high-risk groups (for example, people in the care of mental health services, people with a history of self-harm, people in contact with the criminal justice system, and men aged under 50); reducing access to the means of suicide; and supporting research, data collection and monitoring.

Following a public consultation in 2009, the Welsh Assembly Government published ‘Talk to Me: The National Action Plan to Reduce Suicide and Self Harm in Wales, 2009–2014’.

This is based on a strategic aim to:‘deliver co-ordinated action across all sectors of society for improving the mental health and well-being of the population of Wales, promoting resilience within individuals and communities, delivering timely and effective services to those people identified as being at risk and thereby reducing the rate of suicide and self-harm in Wales’.

The aim is underpinned by seven objectives, which include promoting mental health and well-being, delivering early intervention, improving information on suicide and suicide prevention, and restricting access to the means of suicide. The Action Plan also highlights a suicide prevention health gain target that has been in place since 2002 to reduce the European age-standardised rate by 10% by 2012. Progress towards this target was most recently reported in the Chief Medical Officer for Wales Annual Report 2011.

In Scotland, a 10-year ‘Choose Life’ suicide prevention strategy and action plan was launched in 2002 with the overarching aim to reduce suicide in Scotland by 20% by 2013 (Scottish Executive, 2002). A summary of progress to date and recommended objectives (which are similar to those in England and Wales) for the final phase of the strategy are reported in ‘Refreshing the National Strategy and Action Plan to Prevent Suicide in Scotland’, published by the Scottish Government in 2010.

In 2006, the Department of Health, Social Services and Public Safety in Northern Ireland (DHSSPS) published ‘Protect Life: A Shared Vision – The Northern Ireland Suicide Prevention Strategy and Action Plan, 2006–2011’. The strategy includes two targets: i. to obtain a 10% reduction in the overall suicide rate by 2008 ii. to reduce the overall suicide rate by a further 5% by 2011 The aim, objectives and approach are similar to those in other UK countries and specific actions focussing on both the general population and the target population are also highlighted. In 2012 the strategy was refreshed to cover the period 2011 to 2013 and the DHSSPS published an evaluation of the original ‘Protect Life’ strategy

People with mental illness have a higher suicide risk then the general population (Windfur and Kapur, 2011). A National Confidential Inquiry into Suicide and Homicide by People with Mental Illness was set up to help reduce this risk. The recommendations of this project could assist health professionals and policymakers improve patient safety and reduce the suicide risk of individuals who are in contact with mental health services. The most recent annual report from the Confidential Inquiry was published in July 2012.

Comparison with other countries

The suicide statistics presented in this bulletin are not always comparable with those produced by other countries because definitions and suicide registration methods vary. For example, deaths from injuries and poisonings of undetermined intent are included in UK suicide figures, (as well as deaths from intentional self-harm). This is because in the UK we assume that these deaths were self-inflicted, but there was insufficient evidence to prove that the deceased deliberately intended to kill themselves (Adelstein and Mardon, 1975). As this cannot be assumed in children, UK suicide figures routinely only include persons aged 15 years and over (although data for children aged ten and over are available on request).

Suicide figures published by Eurostat for European countries are based on a broadly comparable definition of deaths from intentional self-harm only. These are available for all ages and rates for males and females are age-standardised to the European Standard Population. Age-specific (or ‘crude’) rates for particular age groups are also available.

Suicide figures published by the World Health Organization (WHO) use official figures made available to WHO by its member states. These are based on actual death certificates signed by legally authorised personnel, usually doctors and, to a lesser extent, police officers. Although they are not all directly comparable or timely, the suicide figures published by the WHO give an overall perspective of the extent of suicide deaths around the world.

The National Statistics definition of suicide includes deaths given an underlying cause of intentional self-harm or an injury/poisoning of undetermined intent. In England and Wales it has been customary to assume that most injuries and poisonings of undetermined intent are cases where the harm was self-inflicted, but there was insufficient evidence to prove that the deceased deliberately intended to kill themselves (Adelstein and Mardon, 1975). This convention has been adopted across the UK. However, it cannot be applied to children due to the possibility that these deaths were caused by unverifiable accidents, neglect or abuse. Therefore, only persons aged 15 years and over are included in the UK suicide figures.

In the UK, deaths are currently coded using the International Classification of Diseases, Tenth Revision (ICD-10 World Health Organisation, 2010). The codes used to define the suicide figures presented in this bulletin are shown below: