Author Posting. (c) Taylor & Francis, 2007.
This is the author's version of the work. It is posted here by permission of
Taylor & Francis for personal use, not for redistribution.
The definitive version was published in Child Care in Practice, Volume 13
Issue 4, October 2007.
doi:10.1080/13575270701504802
(http://dx.doi.org/10.1080/13575270701504802)
Suicide and Young People: the case of Northern Ireland
Mike Tomlinson.
School of Sociology, Social Policy and Social Work
Queen’s University
Belfast
BT7 1NN
1
Abstract
Suicides in Northern Ireland are examined in the context of what is known about global and
regional trends with respect to gender and age, and change over time. For Northern Ireland,
suicide numbers and rates are plotted for 10-24 year olds from 1967 to 2005. Questions are
raised about the validity of officially registered suicides in the light of inverse correlations with
accidents and the importance of car crashes as a cause of death for this age group. The
increase in suicides during the transition to peace over the last decade is considered in terms
of research on mental and the conflict, and assumptions about the legacies of violence.
Among a range of professionals there is a lack of recognition of conflict-related issues and
how these impact on children and young people. The article concludes with an agenda for
research into self-harm, suicide contagion and the role of popular culture (including the
internet) in mediating ideas about suicide.
Key words: suicide, accidents, conflict, Northern Ireland, children, self-harm
2
Introduction
This article is based on some of the findings and recommendations of a systematic literature
review carried out as part of Northern Ireland’s suicide prevention strategy (Tomlinson, 2007).
The review’s aim was to evaluate any research that might shed light on how the Northern
Ireland (NI) conflict has affected mental health in general and suicide in particular. In addition
to the literature review, a detailed analysis of suicides and accidental deaths was carried out.
Before considering the evidence from NI, the review looked at the global picture with respect
to both mental health and suicide, with a particular focus on the consequences of armed
conflicts for both combatants and civilians.
Age is a relatively minor theme in the review but children and young people are considered at
various points, notably in the literature debating the impact of violence and sectarian conflict
on the young (Gallagher, 2004). In this article the focus is on suicide trends and the
significant gaps in knowledge that surround self-harm among younger people. It is timely to
consider these issues. There is growing concern in local communities and among health and
social care professionals over a suicide ‘epidemic’ and a clustering of suicides within certain
areas and friendship groups.
Suicide patterns and trends
The World Health Organization (no date) maintains a global suicide data set which gives
suicide rates by age and gender for most countries, though the availability and quality of the
data varies enormously: only four African countries are covered. (Where otherwise indicated,
all the country-based figures quoted below come from the World Health Organization.) Rates
are negligible for most middle eastern countries. They are low throughout Latin America with
the exception of Cuba. The wealthy countries in the pacific region have relatively high suicide
rates. Japan recorded a rate of 25.5 suicides per 100,000 people (the standard measure) for
2003, twice the rates in Australia and New Zealand where rates peaked in the 1990s (New
Zealand Ministry of Health, 2006). Japan’s rate is also twice that in the United States and
Canada.
3
The picture in Europe is one of marked contrasts. The highest suicide rates are found in the
countries of the former Soviet Union and Eastern Bloc. In the mid-1990s Lithuania’s rate
reached 46, almost double the rate in the other Baltic states. The Russian Federation follows
the Ukraine pattern quite closely only at a higher level with a rate of 34 in 2004. Not all former
Eastern-bloc countries have high suicide rates, however. Romania has a rate of 12.5 (2004)
and the Czech Republic, 15.5. Poland stands out for having a steady upward trend from 1955
onwards: the rate for men rose 2.5 times by 2003.
A recent report on the state of mental health in the European Union discussed differences in
suicide rates across what were then the 15 countries of the EU plus Norway (European
Commission, 2004). For 1997, the report shows that rates vary from 3 in Greece to 24 in
Finland. The United Kingdom (UK) was third lowest with a rate of 7 and Ireland was 8 th
highest at 13.
At the global level the average suicide rate recorded for men has almost doubled over a fifty
year period to the year 2000 (from 16 to 28 per 100,000). For women, the increase was from
5 to 8. This reflects the fact that in most parts of the world, suicide rates for men are higher
than for women – markedly so in some regions: the gender differential is six times in the
Russian Federation and Poland. But there are some notable exceptions.
China stands out because suicide rates for women either match or exceed those for men. For
‘selected rural areas’, the rates were as high as 32.3 for women, 23.2 for men in the late
1980s, and a similar gap remained in the late 1990s. Another exception is the Tamil Nadu
region of Southern India where over a ten year period the average suicide rate among young
women (15-19) was found to be 152 – more than double the rate for men (69) in the same
age group. This is 70 times the rate for young women in the UK (Aaron et al., 2004).
The typical age-related pattern throughout the world is that suicide rates increase with age.
But in countries with high and/or rising suicide trends, the highest rates tend to be recorded
4
for the middle age bands: men aged 45-54 have a suicide rate of 129 in Lithuania, for
example. There is a clear gender effect at work here. In European Union countries, there was
a strong decrease in female suicide rates across all countries (except Spain and
Luxembourg) over a twenty year period up to 1999. For men, the trends were more
contradictory. Rates increased strongly in the Republic of Ireland, less so in Spain and
Luxembourg, and marginally in Norway and Greece. The UK rate declined very slightly, but
for some other countries the fall was sharper: Sweden, Denmark, Austria, Finland and
Portugal. So where suicide trends are upwards, this is principally because more men aged
35-54 are taking their own lives. But it is a trend that is bolstered by more suicides among
younger men: in Estonia men aged 25-34 have a suicide rate of 57.
Britain and Ireland
Lester et al. (1997) compared suicide trends for age and gender in Britain and Ireland using
four fixed points over a thirty year period up to 1990. In 1960 NI’s suicide rates across the age
bands matched the profile of the Republic of Ireland, though at a higher level. Scotland and
England and Wales were also closely matched. By 1980, rates in the Republic had more
than tripled for the 35-54 groups and the shape of the profile is again most similar to NI. For
the last date in the series (1990) the peak rate for both the Republic and NI is for the 55-64
age group.
Bunting and Kelly (1998) compared suicide trends across the UK between 1982 and 1996, for
men and women, and for the two age-groups 15-44 and 45+. Suicide rates for all categories
went down except for men aged 15-44. The English trend for this category rose steadily to the
early 1990s but declined thereafter, unlike in the other countries where the rates continued to
climb. NI overtook the English rate by 1996.
As Brock et al. (2006) show, Scotland has by far the highest suicide rates for men and
women, reaching 33 for men in 1998/2000. NI’s rate for men was the lowest (18.1) compared
to all the UK countries and all the English regions at the start of the 1990s. But it overtook the
5
English rate around 1998 and is now in line with the UK average for 2002/04 (18.3). The
extent of the suicide problem in Scotland is further emphasized by the analysis of suicide
rates in over 400 local areas throughout the UK for two periods falling either side of the 1998
Agreement (1991-1997 and 1998-2004). In both periods well over half the top 20 areas for
suicide are in Scotland. Between the two periods, Belfast West climbed from 259th to 13th
place with a rate of 34.2 for men; Belfast North was ranked 319th in the pre-1998 period and
11th thereafter with a rate of 35.4. Most of these changes are accounted for by increased
suicide rates for men in the younger age groups, notably among the under 25s and 25-34
year olds.
The most recent figures for registered suicides suggest that NI is on a sharply rising curve.
There were 138 suicides in 1997 rising to 291 in 2006. As Figure 1 indicates, the suicide rate
for younger men (15-24) doubled between 1995 and 2000 (to over 30 per 100,000). From a
much lower base, there was an eight-fold increase in the rate for younger women from 1994
to 2000 but the rate has halved since then.
[FIGURE 1 ABOUT HERE]
Figure 1: Registered suicides among under 25 year olds
Source: Compiled from Annual Reports of the Registrar General for Northern Ireland; and
Samaritans, 2006.
Figure 1 also shows suicide trends in NI from 1967 for those aged 10-24. Suicides for this
age group show a broad upward trend both in terms of numbers and share of all suicides.
6
Taking all the deaths for this age group, ‘transport accidents’ typically account for a third or
more of deaths annually, and suicides about one fifth. But this has not always been the case.
Going back to pre-conflict 1967, suicides account for just under 5 per cent of all deaths
among 10-24 year olds.
The youngest age group represented in the raw data is 10-14. While the numbers involved
are very small, there is a discernible increase in recent years. In the two years 1967 and
1968, three children aged 14 or younger killed themselves, but from 1969 to 1980 only one
suicide occurred in this age group. Taking ten year bands from 1967, the annual average
number of suicides remains at 0.4 for twenty years (less than one death every two years).
That figure doubles for the period 1987-96 and, for the remaining nine years for which figures
are available (1997-2005), the average is now 1.1 a year.
Unfortunately, further age breakdown of 15-24 year old suicides is not possible. The way the
figures are presented does not facilitate meaningful social analysis, either in terms of various
legal definitions applying to children and young people, or in terms of establishing at which
ages suicides are concentrated, how this varies by gender and if there is movement over
time.
Conflict, mental health and suicide
Evidence that experience of the NI conflict is associated with poorer mental health is strong
(Miller et al., 2003; O’Reilly and Stevenson, 2003; Muldoon et al., 2005). Population-based
surveys show several things. First, those who experienced most violence have significantly
higher rates of depression than those with little or no experience. People whose areas had
been heavily affected by violence had very high rates of depression. So the relationship
between conflict experience and poor mental health, and between lack of conflict experience
and good mental health, is well established.
7
How important this is to suicide – especially among younger people – is less clear. Most
people who take their own lives are known to be suffering from depression. But only a
minority of suicides were being treated for a recognized disorder or illness (National
Confidential Inquiry, 2006). Women suicides are much more likely than men to have been
diagnosed with depression, though women are significantly less likely than men to kill
themselves. Understanding suicide as the outcome of psychological sickness, however,
provides only part of the picture.
There are two basic issues that need to be tackled when studying suicide sociologically. The
first is the dominance of suicide statistics in framing discussions of the nature and extent of
the problem, and the second concerns the social processes and circumstances which shape
suicidal ideas and action.
The classic sociological text, Durkheim’s Suicide (2002), remains at the heart of debates
within sociology because of its treatment of official statistics as ‘social facts’ and debates
around ‘scientific’ approaches to understanding social phenomena. As Atkinson points out,
the data is key:
[W]hether one’s interest is in the relationship between sociological and psychological
modes of explanation… important decisions have to be made about the status and
adequacy of the data chosen for analysis. (1978, pp.31-2)
As with murder, definitions of suicide (‘self-murder’) are still anchored in the idea of intention
and the assumption that this can be accurately determined by medical and legal authorities.
This remains the case, notwithstanding the recent practice of including deaths of
‘undetermined intent’ in headline suicide figures, or the Luce Report (2003) criticism that
current processes of death certification are insensitive to the needs of the bereaved.
Judgements made by families, doctors, coroners, registrars and others are affected by a
range of factors (Prior, 1989) and conditions of war, political unrest and social conflict bring
special factors into play. In this respect it seems unsurprising that there was a sudden drop in
the registration of suicides in NI in the early 1970s. Similarly, the decriminalization of suicide
in the Republic of Ireland (which took place as late as 1993) and changes in the coding of
8
coroners’ findings, had a considerable impact on the official figures (Lester et al.,1997;
Kelleher and Daly, 1990; Walsh et al., 1990).
In NI as elsewhere, there is a considerable discrepancy between the number of suicides that
occur in a given year and those that come to be registered that year: only 72 of the 291
suicides registered in 2006 actually occurred in 2006 (Northern Ireland Statistics and
Research Agency, 2007). This delay means the official figures are a poor guide to changes in
trends.
Much has been made of ‘concealment’ – the capacity of families and others to hide evidence
around cause of death (or injury). There may be social and material reasons for covering up
suicide and these change over time and between social groups. Douglas (1971) argues that
biases in official statistics reflect class, occupation and other aspects of social structure. In
situations where suicide is stigmatized – for example through law or religious teaching –
‘there will be both differential tendencies on the parts of members of different (official)
categories to have any “suspicious” deaths within their families categorized as something
other than “suicide” and differential degrees of success in these attempts’ (Douglas,1971,
p.129). The main alternative to ‘suicide’ is for a death to be classified as an accident.
In the 1960s, 80 per cent of all suicides in NI were from poisoning or drowning, ‘soft’ methods
that are regarded as particularly prone to interpretation and variable registration. When
‘accidents’ and ‘suicide’ deaths from poisoning are examined, highly significant inverse
correlations are found for the period 1967-72 and for the longer period, 1967-82 (Tomlinson,
2007, pp.88-90), suggesting that some suicides were hidden in ‘accidents’. This is a
particularly important point with respect to young men and adolescents (aged 10-24) for
whom suicide and car crashes dominate as causes of death. Indeed, there is a branch of
suicide research that investigates single occupancy car crashes, arguing that a proportion of
these (5-10 per cent) could reasonably be classified as suicides (Phillips, 1979; Peck and
Warner, 1995). Car crashes become less important in older groups, accounting for less than
9
a quarter of all male deaths among 25-34 year olds and little over a tenth of male deaths for
25-44 year olds.
While there are social processes that contribute to the concealment of suicides, there are
others which may lead to greater public recognition of the problem. In recent years, local
communities have engaged in protests around suicide, highlighting in particular the suicides
of younger people and the lack of readily accessible mental health services. The period of
transition to peace appears to have led to a greater willingness within communities most
affected by conflict, to acknowledge mental health problems and to see these as a legacy of
the conflict (Shirlow et al., 2005). Such protests drew a hostile response initially, with one
Minister suggesting that the growth in suicides among young males in North and West Belfast
was attributable to punishment beatings by paramilitary organizations. This idea has been
part of the official response to the suicide problem for at least ten years and has only changed
very recently (Tomlinson, 2007, pp.8-11).
The lack of recognition of conflict-related issues, especially as they impact on younger
people, is widespread among teachers and health and social care professionals, according to
some researchers (Burrows and Keenan, 2004; Kilkelly et al., 2004). This observation reflects
a broad debate running through the literature between those who see the conflict as being
marginal to mental well-being (Curran and Miller, 2001) and those who see it as fundamental
(Hayes and Campbell, 2005; Hamber, 2004). For the latter perspective it is essential to
understand the ‘intergenerational transmission of trauma’ (Bar-On, 1996; Dulmus and
Wodarski, 2000) in order to appreciate that negative impacts on children and young people
will endure unless they are openly addressed. Burrows and Keenan refer specifically to youth
suicide as an example of how traumatic events are consciously or subconsciously re-enacted.
This type of complex psycho-social interaction does not necessarily lead in the direction of
depression and suicide, but may emerge in physical ill-health (Kapur and Campbell, 2004;
Wilkinson, 1996).
10
It has been argued that the higher rates of suicide during the transition to peace are a
reflection of a loss of social and political integration said to be characteristic of ‘wartime’
(McGowan et al., 2005). This is not well theorized or researched, however, especially in
relation to children and young people. Some peoples and communities were relatively
untouched by the conflict, while others lived in neighbourhoods on a quasi war footing for
decades. Divisions within the communities may be as important as divisions between them,
so any theorizing around polarization and social integration needs to take account of wide
variations in experience.
Challenges for research
It is clear from this brief article that there are major gaps in the available knowledge and
research on how children and young people are positioned in relation to self-harm and suicide
in NI. As has been seen, the crude age categories in published suicide statistics prevent
meaningful social analysis. It is only quite recently that attempts have been made to estimate
the prevalence of self-harming, with the focus on younger people. One approach is to monitor
relevant hospital attendances for parasuicide, less than ten per cent of which result in a death
registered as a suicide. Yet the vast majority of self-harming by young people does not come
to the attention of health and social care professionals (Brophy, 2006). This is a known issue
and progress is needed as a matter of priority so that the ways in which anxiety, distress and
depression are being experienced and expressed by younger people can be better
understood and supported.
There is a lack of qualitative research exploring how families and local communities cope with
depression, self-harming and suicide, and how knowledge of, and attitudes towards potential
sources of help vary. An important aspect of this is ‘suicide contagion’, or imitation, which
researchers claim may account for up to ten per cent of suicides, and possibly more amongst
younger people (Gould et al., 1989; McKenzie et al., 2005). Very little is known about the
processes that lead to suicide clustering: apparently linked chains of suicide within families,
friendship groups and local communities. Similarly, it is only recently that researchers have
11
begun to ask questions about the role of popular culture, new communications and the
internet in mediating ideas about suicide (Gill, 2007).
While a start has been made in understanding how the NI conflict has affected the registration
of suicides, the recognition of the suicide problem and the speed and nature of the responses
to it, there is a long way to go before children and young people are brought fully into the
picture.
12
References
Aaron, R., Joseph, A., Abraham, S., Muliyil, J., George, K., Prasad, J., Minz, S., Abraham, J.
V., and Bose, A. (2004). Suicides in young people in rural southern India. The Lancet, 363,
9415, pp.1117-1118.
Atkinson, J. M. (1978). Discovering Suicide: Studies in the Social Organisation of Sudden
Death. London: Macmillan.
Bar-On, D. (1996). Attempting to overcome the intergenerational transmission of trauma. In
R. J. Apfel and B. Simon (eds) Minefields in Their Hearts: The Mental Health of Children in
War and Communal Violence. New Haven, Conn.: Yale. pp. 165-188.
Brock, A., Baker, A., Griffiths, C., Jackson, G, Fegan, G. and Marshall, D. (2006). Suicide
trends and geographical variations in the UK, 1991-2004. Health Statistics Quarterly, 31.
Brophy, M. (2006). Truth Hurts: Report of the National Inquiry into Self-Harm Among Young
People. London: Mental Health Foundation.
Bunting, J. and Kelly, S. (1998). Geographic variations in suicide mortality, 1982-96.
Population Trends, 93.
Burrows, R. and Keenan, B. (2004). Bearing witness: supporting parents and children in the
transition to peace. Child Care in Practice, 10, 2, pp.107–25.
Curran, P. S. and Miller, P. W. (2001). Psychiatric implications of chronic civilian strife or war:
Northern Ireland. Advances in Psychiatric Treatment, 7, pp.73–80.
Douglas, J. D. (1971). The Sociological Analysis of Social Meanings of Suicide. In A. Giddens
(ed) The Sociology of Suicide. London: Frank Cass. pp.121-151.
Dulmus, C. and Wodarski, J. (2000). Trauma-related symptomatology among children of
parents victimized by urban community violence. American Journal of Orthopsychiatry, 70,
pp.272-277.
Durkheim, E. (2002). Suicide. London: Routledge. (First published 1897.)
European Commission, Health & Consumer Protection Directorate General (2004). The State
of Mental Health in the European Union.
13
Gallagher, A. (2004). After the War Comes Peace? An Examination of the Impact of the
Northern Ireland Conflict on Young People. Journal of Social Issues, 60, 3, pp.629-642.
Gill, A. (2007). Suicide Surfers. Sunday Times Colour Supplement, 4 May, pp.24-33.
Gould, M. S., Wallenstein, S. and Davidson, L. (1989). Suicide clusters: a critical review.
Suicide and Life-Threatening Behavior, 19, pp.17 –29.
Hamber, B. (2004). The Impact of Trauma: A psychosocial approach.
http://www.brandonhamber.com (last accessed, 12 March 2007).
Hayes, P. and Campbell, J. (2005). Bloody Sunday, Trauma, Pain and Politics. London: Pluto
Press.
Kapur, R. and Campbell, J. (2004). The Troubled Mind of Northern Ireland: An Analysis of the
Emotional Effects of the Troubles. London: Karnac.
Kelleher, M. J., and Daly, M. (1990). Suicide in Cork and Ireland. British Journal of Psychiatry,
157, pp.533-8.
Kilkelly, U., Kilpatrick, R., Lundy, L., Moore, L., Scraton, P., Davey, C., Dwyer, C. and
McAlister, S. (2004). Children’s Rights in Northern Ireland: research commissioned by the
Northern Ireland Commissioner for Children and Young People, Belfast: NICCY.
Lester, D., Cantor, C. H. and Leenaars, A. (1997). Suicide in the United Kingdom and
Ireland,.European Psychiatry, 12, pp.300-304.
Luce Report (2003). Death Certification and Investigation in England, Wales and Northern
Ireland. Report of an Independent Review of Coroner Services, Commissioned by the Home
Office and Chaired by Tom Luce. Royal College of General Practitioners. Summary Paper
2003/05.
McGowan, I., Hamilton, S., Miller, P. and Kernohan, G. (2005). Contrasting terrorist-related
deaths with suicide trends over 34 years. Journal of Mental Health, 14, 4, pp.399-405.
McKenzie, N. et al. (2005). Clustering of suicides among people with mental illness. The
British Journal of Psychiatry, 187, pp.476-480.
Miller, R. L., Devine, P. and Schubotz, D. (2003). Secondary Analysis of the 1997 and 2001
Northern Ireland Health and Social Wellbeing Surveys. Belfast: Queen’s University.
14
Muldoon, O., Schmid, K., Downes, C., Kremer, J. and Trew, K. (2005). The Legacy of the
Troubles: Experience of the Troubles, Mental Health and Social Attitudes. Belfast: School of
Psychology, Queen’s University.
National Confidential Inquiry (2006). Avoidable Deaths: five year report of the national
confidential inquiry into suicide and homicide by people with mental illness. Manchester: NCI,
University of Manchester.
New Zealand Ministry of Health (2006). Suicide Facts: Provisional 2003 All-Ages Statistics.
Monitoring Report No. 1. Wellington: Ministry of Health.
Northern Ireland Statistics and Research Agency (2007). Mortality Statistics for 2006. Press
Release, 22 March. Belfast: Northern Ireland Statistics and Research Agency.
O’Reilly, D and Stevenson, O. (2003). Mental Health in Northern Ireland: have “the Troubles”
made it worse? Journal of Epidemiology and Community Health, 57, pp.488-492.
Peck, D. L. and Warner, K. (1995). Accident or Suicide? Single-Vehicle Car Accidents and
the Intent Hypothesis. Adolescence, 30, 118, pp.463-72.
Phillips, D. P. (1979). Suicide, Motor Vehicle Fatalities, and the Mass Media: Evidence
Toward a Theory of Suggestion. The American Journal of Sociology, 84, 5, pp.1150-1174.
Prior, L. (1989). The Social Organisation of Death, Medical Discourse, and Social Practices in
Belfast. London: Macmillan.
Samaritans (2006). Information Resource Pack 2006, Ewell: The Samaritans.
Shirlow, P. Graham, B., McEvoy, K., Ó hAdhmaill, F. and Purvis, D. (2005). Politically
Motivated Former Prisoner Groups: Community Activism and Conflict Transformation. Belfast:
Northern Ireland Community Relations Council.
Tomlinson, M. (2007). The Trouble with Suicide. Mental Health, Suicide and the Northern
Ireland Conflict: A Review of the Evidence. Belfast: Department of Health, Social Services
and Public Safety (Northern Ireland).
Walsh, D., Cullen, A., Cullivan, A. and O’Donnell, B. (1990). Do statistics lie? Suicide in
Kildare – and in Ireland. Psychological Medicine 20, pp.867-891.
15
Wilkinson, R. (1996). Unhealthy Societies: The Afflictions of Inequality. London: Routledge.
World Health Organization (no date). Suicide: Country Reports and Charts.
http://www.who.int/mental_health/prevention/suicide/country_reports/en/index.html (last
accessed, 12 March 2007).
16