De Leo, D .uk



Factors associated with suicide in four age groups: a population based studyAbstractBackground; Life events and circumstances leading to death change throughout the life course. Aims; Four age groups within those who have died by suicide are compared in terms method of suicide, sex, occupation, mental disorders, prior suicide attempts and life events prior to death. Methods; Analyses were based on a database of deaths by suicide and undetermined intent based on data in the NI coronial files from 2005-2011 (N=1667). Results; Hanging is very prominent as a method of death within the under 20 age group. Women who die by suicide are more likely to have a known mental disorder than men, and the proportions increase with age group. Relationship difficulties are associated with many of these deaths and particularly for males aged under 40 years. Physical health and life events were more relevant than mental illness per se, in males aged over 61 yearsLimitations: For a sizeable proportion of the cases included in the current database there was no information on the life events prior to deathConclusions; Understanding the factors associated with suicide across age groups is essential to informing suicide prevention strategy and programmes and the development of more nuanced and effective interventions. Suicide in NI: a mixed methods analysis of four age groupsSuicide is a leading cause of death globally with the world health organisation estimating that a person dies by suicide every 40 seconds (WHO, 2014). In addition to the distress and loss of life, it is estimated that each suicide costs the economy around 1.4 million (Kennelly, 2007). At a regional level, Northern Ireland (NI) is a suitable place in which to study this phenomenon. NI currently has the highest rate of suicide in the UK and rates have doubled in the past 10 years (NISRA, 2014, Snowcroft, 2013, Tomlinson, 2012). The increase in suicide rates has come at a time when the NI population are emerging from a 30 year period of civil conflict and violence known as the “Troubles”.It is believed that the conflict may have in fact protected the population from suicide as a consequence of a shared sense of struggle and increased connectedness among disadvantaged groups (Tomlinson, 2012). However the post conflict context has seen high suicide rates among young people who were not exposed to the violence, which has given rise to concerns about the transgenerational transmission of traumatic experiences and an intergenerational impact of the conflict on mental illness (O’Neill et al., 2015). In addition, the group who were most exposed to the violence continue to have high rates of suicidal behaviour as they grow older, in the current climate of comparative stability (NISRA, 2014, O’Neill et al., 2014). Mental illness is widely regarded as the strongest risk factor for suicide (O’Connor & Nock, 2014), and in NI the rates are among the highest in the western world, with findings from the world mental health surveys showing that 39% of the NI population met the criteria for a lifetime mental disorder, including mild disorders (Bunting et al., 2012). These high levels of mental disorders are of concern in relation to suicidal behaviour. However, most people with a mental disorder to not die by suicide and we must of course look beyond mental illness in order to understand this behaviour fully (Nock et al., 2008, O’Connor & Nock, 2014).Contemporary theories of suicide (e.g. O’Connor, 2011, Joiner, 2005, Klonsky & May, 2014) highlight the complexity of suicide, the multiple factors involved, and the importance of social context and life stressors in influencing suicidal behaviour. International research on suicide points to the importance of age group and gender differences in understanding patterns of deaths (De Leo et al. 2013, Koo et al., 2016). We know that life events and circumstances leading to death change throughout the life course and this, coupled with the shifts in age group patterns in deaths by suicide, point to differences in age cohorts of suicides in NI. Additionally, NI has different patterns of exposure to risk factors. Research shows that NI has higher levels of trauma related mental illnesses in those who were alive at the time of the Troubles, and Troubles-related trauma exposure is associated with suicidal behaviour (O’Neill et al., 2014a). Knowing someone affected by the Troubles was also associated with self-harm in young people (O’Connor et al., 2014). It is theref(ore important that we differentiate between sub groups, and examine the factors associated with suicide in NI in age groups so that we can enhance our ability to predict suicide and develop more nuanced and effective interventions. MethodsMethods are outlined detail in O’Neill et al. (2014b). Approval for the research was obtained from the Ulster University ethical committee. Cases were recorded by year of death and deaths by suicide and undetermined intent were generated by staff from the NI Coroner’s Service (CSNI). This subsequently directed file selection. The requirements for a coronial verdict of suicide were stricter than those required for classification as a probable suicide and inclusion in the database. Undetermined deaths, which were probable suicides, were classified by both the senior Coroner and also, following an analysis of the file, the Research Associate. Undetermined deaths were classified as suicide where the means of death was that of a common means of suicide and it was likely that the individual took direct action that led to their death. Case validation was undertaken with the assistance of NI Statistics and Research Agency (NISRA) personnel to ensure that the cases in the database were those included in the official NISRA statistics on deaths by suicide. For each case data was extracted from physical files stored in the CSNI archives and electronically recorded in a database. Data on established risk factors including prior suicidal behaviour, diagnosed physical and mental health conditions, pharmacological profiles, demographics, substance use and prior adverse events were extracted. Health disorders and service use was assessed using the deceased person’s medical notes (where available), police reports and next of kin statements. Socioeconomic indicators were identified through the same sources as well as information included in pathology reports regarding occupation and geographical position (coordinates). Information on adverse events prior to death was coded by both the Research Associate and another investigator independently. There was a high level of concordance between the two and any discrepancies were resolved prior to statistical analysis. The main source of discrepancy was where particular life events could be coded in more than one category, and it was agreed that these events were recorded in all categories that were relevant. Operational definitions of variables are as follows: age and marital status refers to status at time of death. Previous suicidal behaviour includes hospital and non-hospital treated events; due to the potential variation in the recording of these events we did not differentiate between suicide attempt and non-suicidal self-injury. In terms of service use, primary care refers to care under the General Practitioner; secondary care refers to outpatient mental health treatments; tertiary care refers to psychiatric inpatient care. Mental disorders refer to both mental and substance disorders. Chi squared analyses were used to examine associations between (a) age group and method of suicide, (b) age group, sex, mental disorders and prior suicide attempt/ behaviour, (c) age groups and life events prior to death, and (d) age group and occupation. Analyses were implemented using the IBM SPSS package. ResultsThe four groups are significantly different in relation to the use of hanging as a method, with 84.1% of the under 20s dying by hanging, compared with 65.7% of the 21-40 age group, 41.5% of the 41-60 age group and 60.6% of the over 60s (χ2(df=9)=132.104, p>.000). Almost one in five of the 21-40 age group died by overdose. There was a large difference in the use of overdose in the 41-60 age group with 40.5% of women and 15.5% of men in this age group using this method. There were no gender differences in the methods used in the under 20s. In the other groups males were more likely to use hanging than females with the highest differences in the 41-60 age group (38% compared with 60%) and 61+ age groups (25.5% compared with 47.2%) (χ2(df=3)=17.954, p<0.000).[INSERT TABLE 1 ABOUT HERE]The proportions who were known to have a mental disorder increase with age group and the difference between the age groups is statistically significant ( χ2(df=3)=38.744, p<0.000). For women the proportions with a disorder continue to rise, whereas in males the proportion in the 61+ age group is lower than the 41-60 age group; with the proportions known to have a disorder peaking at 65% in the 41-60 age group. The proportions who have had recorded previous suicide attempts or suicidal behaviour are similar across the age groups for women (χ2(df=3)=2.433, p= 0.488) but for men the oldest age group had a significantly fewer proportion with a prior attempt (χ2(df=3)=28.028, p>0.000). [INSERT TABLE 2 ABOUT HERE]Statistical tests were not undertaken for the life events variable due to the levels of missing data. The proportions of males and females who died following relationship breakups or difficulties is highest in the under 20s; however whilst a lower proportion of females in the 21-40 age group had relationship problems prior to death (33.7%), the proportion of males in this group rose from 47.9% in the under 20 age group to 55% in the 21-40 age group. The death or illness of another person was reported for a lower proportion of males than females in the 21-40 age group and 41-60 age group; whilst the proportions for males and females in the over 60s and under 20s were similar. The proportions for whom the death or illness of another was reported peaked, at just over one in five, in the 41-60 age group. There were also age group differences in the Fears for own health category with this issue recorded for a much higher proportion of women aged 21-40, than men in this age group (12% compared with 3.9%). In the 61+ age group over three times the proportion of men than women were believed to have had fears for their own health. Higher proportions of men than women had recorded financial or employment concerns, with the gender differences being higher in the under 20s (0% of females, compared with 16% of males) and the 41-60 age group (10.8% of females and 18% of males). Finally the proportions with substance difficulties were similar in all the under 60 group and were much less common in the over 60s. For males and females in the under 20s the rates were similar; however more males in the 21-40 age group had substance disorders (10.4% compared with 6.9%), whilst the pattern was reversed for the 41-60 age group (15.4% of females and 5.4% of males).[INSERT TABLE 3 ABOUT HERE]There were statistically significant differences across all four age groups with regards to occupational the category of the deceased (χ2(df=12)=1243.357, p<0.000) and the occupational category of the deceased also differed between males and females (χ2(df=4)=13.745, p<0.05). In the under 20s similar proportions of males and females were employed, twice the proportion of males as females were unemployed and a higher proportion of females (almost half, 47.5%) were students. In the 21-40 group again similar proportions of males and females were employed (the highest proportion of all the age groups, 45.2% and 43.6%) and higher proportions of males than females were unemployed (a third of women and 40.8% of men). The proportions of students in this age group were low at 3.2% for males and 1.4% for females. The proportions with unknown occupational status start to increase at this age, with 14% of males in this age group coded “unknown”. In the 41-60 age group there is the only gender difference in the proportions employed, with just over a third of women (35.6%) and 44.3% of men in this category. Similar proportions of men and women in this age group are unemployed. A slightly higher proportion of males than females are retired (5.2% compared with 3.8%) and employment is categorised as unknown for a fifth of women and 13% of men. None of the deceased in the 41-60 age group or the 61+ group were categorised as being a student at the time of death. In the oldest age group, those aged over 61 years, similar proportions of men and women were employed almost twice the proportion of females as males were unemployed and 58.8% of women and 68.3% of men were retired. A slightly higher proportion of females than males were categorised as unknown in this age group (17.6% of women, compared with 14.8% of men). [INSERT TABLE 4 ABOUT HERE]DiscussionPrevious reports from the NI suicide database have illustrated the main gender, occupational, service use characteristics of the deceased as a whole (O’Neill et al., 2014b, O’Neill et al., 2016). This analyses examines the data in relation to four age groups in order to inform the suicide prevention strategy and programmes. The following limitations need to be taken into account when interpreting the study findings. One of the key difficulties with the research is that the data regarding the circumstances surrounding the deaths and life events prior to death were based on coronial files; data were not systematically recorded for each death. Life events prior to death were recorded in only 61% of cases. There are many reasons why information may have been withheld or not recorded, including the stigma associated with certain events, so we cannot assume that the missing data was random missing data. Due to variation in recording systems the “prior suicidal behaviour” variable may include both non-suicidal self-injury and suicide attempt. Furthermore the broad age groups used in the analyses may mask more subtle age effects; for example the circumstances surrounding suicides of individuals in their 20s and 30s may be quite different. Recent studies of suicide risk have emphasised the differences between the factors that lead to suicidal ideation, which is more common, and those that are associated with death by suicide. The data used in this study does not permit differentiation between these two aspects of suicidal behaviour (Nock et al., 2010). The under 20 group appear to be quite different to the other age groups in terms of the methods used. Hanging remains the most common method overall (O’Neill et al., 2016), and hanging is an even more common method for this group. The rates of hanging decreased with age group. There is little can be done to prevent access to means of hanging since numerous everyday items may be used as ligatures. However other studies of suicide in this age group point to the importance of the internet as a source of information on suicide prior to death (National Confidential Inquiry, 2016). It is therefore possible that online information on this method has fuelled the use of hanging among this age group. Reducing access to online information, or promoting suicide prevention organisations and helpseeking behaviour online remain important means of preventing this behaviour (National Confidential Inquiry, 2016, O’Connor et al., 2014). Self-harm is one of the biggest predictors of death by suicide and is prevalent (O’Connor & Nock, 2014), with around one in 10 young people in NI reporting this behaviour (O’Connor et al., 2014). Interestingly, the most common methods are cutting, and poisoning (O’Connor et al., 2014), whilst the primary cause of suicide death in this age group is hanging. This suggests a reduced level of intent among those who self-harm, supporting theories of self-harm as a socially or physiologically reinforced stress modifying behaviour, for this age group (Nock and Prinstein, 2004; O’Connor et al., 2014).Whilst overdose is the cause of a minority of suicides overall, it is a very common method in women in all but the first age group. It led to the deaths of 40.5% of women in the 41-60 age group and almost a third of the women in the 21-40 age group and 61+ group. This finding supports suicide prevention strategies that restrict access to medication that can be used for overdose, particularly among women (WHO, 2014). Women who die by suicide are more likely to have a known mental disorder than men, and the proportions increase with age group. The rates of known disorders are higher among women in all the age groups again confirming earlier findings that women are more likely to report symptoms (Bunting et al., 2012, Bunting et al., 2013) and seek treatment (Andrade et al., 2014, O’Neill et al., 2014b). In men there is also an increase in proportions known to have a mental illness from age 20-60 years. However there is a decrease in the oldest age group reflecting perhaps a reduced willingness to disclose symptoms or increased prominence of life events in suicide risk for this group. These results illustrate the importance of mental health anti-stigma programmes in suicide prevention (WHO, 2014). There is evidence that large proportions of those individuals who die by suicide have a mental disorder (Foster et al., 1997, Nock et al., 2008, O’Neill et al., 2016). The current findings showed lower rates of diagnosed disorders in the younger age groups. This may reflect a lower prevalence of disorders in these age groups, or indeed, the underdiagnosis of disorders in this group. To address the underdiagnosis of disorders the initial signs of mental illness and suicidal behaviour need to be promptly assessed and treated (WHO, 2014). A wealth of studies demonstrate that women are more likely to attempt suicide and survive, than men (O’Neill et al., 2014b, O’Neill et al., 2016). In this study, the incidence of any suicide attempt is similar for both men and women (and consistently around 15% across all the age groups). This may in part reflect the fact that women tend to make many more attempts than men (O’Neill et al., 2014b) and that in this analysis we examined any attempt rather than considering the number of attempts. It is worth noting that for a sizeable proportion of the cases included in the current database there was no information on the life events prior to death (O’Neill et al., 2016). Notwithstanding this, it appears that relationship difficulties are associated with many of these deaths and particularly for males aged under 40 years (O’Neill et al., 2016). Whilst it is difficult to prevent relationship difficulties, it is important that support is available for individuals in a relationship crisis and that efforts to increase emotional intelligence and self-management across the general population are encouraged (WHO, 2014). The theme of relationship loss is also evidenced in the proportions who die following the death or illness of another (O’Neill et al., 2016). Both these themes illustrate the importance of improving social connections (Joiner, 2005) and acknowledging the social context in suicide prevention treatments (WHO, 2014). They also emphasise the need for appropriate supports services for the bereaved and those who are facing divorce or separation (WHO, 2014). As reported previously (O’Neill et al., 2016), fears for own health was a factor for a higher proportion of women than men generally. However, when age differences were considered, this was not the case in the over 60s where this issue was recorded for a much higher proportion of men than women. This again supports the theory that physical health and life events were more relevant than mental illness per se, in males aged over 61 years, and supports a studies by Fassberg et al. (2016) and Koo et al., (2016) who found physical health and functional disability to be related to suicidal behaviour in older people. Substance disorders are a prominent factor in many of the deaths, particularly in the younger age groups. However the age and gender group differences illustrate key stages at which we need to exercise vigilance. For men, the rates decline after age 40, however for women the rates remain at the same level until age 60. Substance use remains an important element of any suicide prevention strategy, and the fact that in NI 56% of those who die by suicide have taken alcohol illustrates the role of alcohol in these deaths (O’Neill et al., 2016). In terms of life events, the rates of recorded financial and employment crises were similar in both men and women in the 21-40 and 41-60 age groups. In the other lower and upper age groups, the issue was a more common problem in men than women overall. This is known to be a risk factor for suicide and there is emerging evidence that the economic crisis and subsequent austerity measures have led to increases in deaths by suicide (Corcoran et al., 2015, Martin-Carrasco et al., 2016, O’Neill et al., 2016). Whilst unemployment must be considered as a significant risk factor due to economic associations and also psychological associations such as sense of purpose, so too must other economic issues such as indebtedness or precarious working conditions (Martin-Carrasco et al., 2016). The current findings suggest that this may have differential impacts upon different ages and genders. The findings with regards to employment categories highlight some interesting patterns. In terms of occupational status of those who died by suicide, within those who were unemployed, there was a predominance of males within the under 40s. As expected, high proportions of individuals in the over 60 group were retired, and only half the proportion of men than women were categorised as unemployed in this age group. The proportions in the “unknown” occupational category, around 15% of males in all the over 21 groups and 20% and 17% of women in the over 41 groups, may represent individuals who had caring responsibilities or worked in informal settings that proved difficult to categorise. Almost half of females who were under 21 were students, this is higher than the males for this age group. Student suicides, particularly those among women, are seen as being linked to social perfectionism and negative social comparison; and again this is reflected in suicide theories (O’Connor, 2011). The existence of this sub group may also be seen to support the findings from Rodway et al (in press) in relation to exam stress as a suicide risk factor in this age group. There is a need to ensure that schools and colleges adopt suicide prevention initiatives and that appropriate support is available for students who struggle with mental health difficulties and self-harm, because both are risk factors in this group. These results give us insights into the various subpopulations of suicides in NI, and highlight age and gender group variations. The findings reinforce the need to guard against over simplification of suicide risk factors and the importance of viewing suicides not as one complete population, but as a number of subpopulations, each with unique patterns and contributing factors. Strategies with the goal of preventing suicides not only need to address mental illness and prior suicidal behaviour, but also the social and economic factors that lead to the life events that precede these deaths. They should inform suicide prevention strategies and signal opportunities for intervention, across the population, in a range of settings where vulnerable individuals may be reached. They support the need to strengthen mental health services, but indicate that stigma reduction campaigns may actually be best targeted towards the younger age groups, as this is the group who are least likely to have a known mental health problem. ReferencesAndrade, L.H., Alonso, J., Mneimneh, Z., et al. (2014). Barriers to mental health treatment: results from the WHO World Mental Health surveys. Psychological Medicine, 44(6), 1303-1317. Bunting, B.P., Murphy, S.D., O’Neill, S., et al. (2012). Lifetime prevalence of mental health disorders and delay in treatment following initial onset: evidence from the Northern Ireland study of health and stress. Psychological Medicine, 42(8), 1727-1739. Bunting, B.P., Murphy, S.D., O’Neill, S., et al. (2013). Prevalence and treatment of 12 month DMS-IV disorders in the Northern Ireland study of health and stress. Social Psychiatry and Psychiatric Epidemiology, 48(1), 81-93. Corcocran, P., Griffin, E., Arensman, E., et al. (2015). Impact of the economic recession and subsequent austerity on suicide and self-harm in Ireland: an interrupted time series analysis. International Journal of Epidemiology, 44(3), 969-977. De Leo, D., Draper, B.M., Snowdon, J. et al. (2013). Suicides in older adults: a case-control psychological autopsy study in Australia. Journal of Psychiatric Research, 47(7), 980-988. Fassberg, M.M., Cheung, G., Canetto, S.S., et al. (2016). A systematic review of physical illness, functional disability, and suicidal behaviour among older adults. Aging and Mental Health, 20(2), 166-194 Foster, T., Gillespie, K., McClelland, R. (1997). Mental disorders and suicide in Northern Ireland. British Journal of Psychiatry, 170, 447–452.Joiner, T.E., 2005. Why People Die by Suicide. Harvard University Press, Cambridge, MA. Kennelly, B. (2007). The economic cost of suicide in Ireland. Crisis, 28, 89-94Klonsky, E.D., & May, A.M., (2014). Differentiating suicide attempters from suicide ideators: a critical frontier for suicidology research. Suicide and Life-Threatening Behaviour, 1, 1–5. Koo, Y.W., K?lves, K.., De Leo, D. (2016). Suicide in older adults: a comparison with middle-aged adults using the Queensland Suicide Register. International Psychogeriatrics., 17, 1-12.Martin-Carrasco, M., Evans-Lacko, S., Dom, G., et al. (2016). EPA guidance on mental health and economic crises in Europe. European Archives of Psychiatry and Clinical Neuroscience, 266(2), 89-124. National Confidential Inquiry (2016). National confidential inquiry into suicide and homicide by people with mental illness. Retrieved from: Northern Ireland Statistics and Research Agency Suicide Deaths (2014). Retrieved from: ?. .uk/demography/default.asp31.htm?. Nock, M.K., Borges, G., Bromet, E.J., et al. (2008). Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. British Journal of Psychiatry, 192(2), 98-105. Nock, M.K., Hwang, I., Sampson, N.A. et al. (2010). Mental disorders, comorbidity and suicidal behavior: results from the National Comorbidity Survey Replication. Molecular Psychiatry, 15(8), 868-876. Nock, M. K. and Prinstein, M.J. (2004). A functional approach to the assessment of self-mutilative behaviour. Journal of Consulting and Clinical Psychology, 114(1), 140-146.O’Connor, R.C. (2011). The integrated motivational-volitional model of suicidal behaviour. Crisis, 32(6), 295–298. O’Connor, R.C., Rasmussen, S., & Hawton, K. (2014). Adolescent self-harm: a school-based study in Northern Ireland. Journal of Affective Disorders, 159, 46-52. O’Connor, R.C., & Nock, M.K. (2014). The psychology of suicidal behaviour. Lancet Psychiatry, 1, 73-85. O’Neill, S., Ferry, F., Murphy, S., et al. (2014a). Patterns of behaviour in Northern Ireland and associations with conflict related trauma. PlosOne, 9(3), e91532. O’Neill, S., Corry, C.V., Murphy, S., et al., (2014b). Characteristics of death by suicide in Northern Ireland from 2005 to 2011 and use of health services prior to death. Journal of Affective Disorders, 168, 466-471. O’Neill, S., Armour, C., Bolton, D., et al. (2015). Towards A Better Future: The Trans-generational Impact of the Troubles on Mental Health. Commission for Victims and Survivors; Belfast, N. Ireland. O’Neill, S., Corry, C., McFeeters, D., et al. (2016). Suicide in Northern Ireland: an analysis of gender differences in demographic, psychological and contextual factors. Crisis-The Journal of Crisis Intervention and Suicide, 37(1), 13-20. Rodway, C., Tham, S.G., Ibrahim, S., Turnbull, P., Windfuhr, W., Shaw, J., Kapur, N., & Appleby, L. (inpress). Suicide in children and young people in England: a consecutive case series. The Lancet Psychiatry, in press. Snowcroft, E. (2013).Samaritans Suicide Statistics Report 2013. Data for 2009–2011. Samaritans. Retrieved from: ? Samaritans%20Suicide%20Statistics%20Report%202013.pdf? (accessed21.04.14).Stanley, I.H., Hom, M.A., Rogers, M.L., et al. (2016). Understanding suicide among older adults: a review of psychological and sociological theories of suicide. Aging and Mental Health, 20(2), 113-122. Tomlinson, M. (2012). War, peace and suicide: the case of Northern Ireland. International Sociology, 27, 464. World Health Organisation (WHO) (2014). Preventing suicide; a global imperative. Retrieved from: ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download