The Scottish Government

« Previous | Contents | Next »

Listen

1. INTRODUCTION

1.1. Background

The epidemiology ⁹ of suicide has changed markedly over the past 30 years, in Scotland as in other (post-) industrialised countries. In particular, the sex gap has continued to grow, with men in Scotland now over three times more likely to kill themselves than women (men: 24.3 per 100 000 population aged 15+ years; women: 7.0/100 000 in 2002) . The age profile is also considerably altered: suicide risk, rather than tending to increase with age, is now highest in the 15-44 age groups, thereafter declining with age.

With respect to the influence of socio-economic status, there is evidence to suggest that lower socio-economic groups have been, and continue to be, more at risk of suicide than higher socio-economic groups. In 1991-93 there was a clear social gradient in suicide mortality between occupational social classes, with a four-fold difference in mortality between social classes I and V for the United Kingdom as a whole. However, this was mainly associated with excess mortality in social class V, as the gradient was less apparent when the experiences of social class V were excluded. Scotland had the worst level of suicide mortality, with higher rates ¹⁰ in each social class than all other countries and regions (with one exception) . There was considerable geographic variation in suicide mortality for social classes I and V. For social class I, suicide rates in Scotland were 3.5 times those in the North East (the region with the lowest mortality in this social class) . For social class V, rates in Scotland were 2.7 times higher than in London (where rates for this class were significantly below most other areas in the UK) ¹¹.

Within Scotland, geographical variations in suicide have also been identified. We have known for some time that rates are higher than expected in the Highlands, but there is increasing evidence of a stark polarisation between the most and least deprived areas. Thus, a recent study (Boyle et al 2005) showed that, while suicide rates declined significantly in all deprivation quintiles for older adults (45+) , the ratio between the most and least deprived quintiles widened slightly from 1.51 to 1.81. However, the gap widened much more (from 2.98 to 4.02 ) among young adults (15-44) , and especially among young women (from 2.96 to 5.77; compared to an increase from 2.99 to 3.67 among young men) .

As relevant background for planning the implementation of Choose Life, the suicide prevention strategy in Scotland, detailed information about the epidemiology of suicide at regional and local levels is required. The General Register Office (Scotland) ( GRO(S) ) has collected and collated a considerable amount of information on each suicide (or possible suicide) death, but relatively little has been published. A comprehensive dataset could be constructed, which, together with appropriate general population data (also available from GRO(S) ) , would help to inform planners and practitioners about the suicide situation in Scotland. Many specific issues could be addressed. As a first step, we suggest examining variation in suicide incidence ¹² by sex, age and socio-economic status across the whole of Scotland and separately for each local area. Given the possible unreliability of social class information derived from death certificates, it would be important to develop and apply area-level measures of economic status based on residential postcodes ('deprivation category') . The extent to which the 'suicide gap' (inequalities in suicide risk) varies according to the prevailing rate of suicide and the level of socio-economic deprivation in the locality also deserves further examination.

The aim and objectives of this study are set out below.

1.2 Aim and objectives

The broad aim of the project is to support the implementation of Choose Life by providing detailed information on the epidemiology of suicide at national and local levels.

The more specific objectives of the project were defined as:

1. To establish the incidence of suicide in Scotland over the period 1989-2004 ¹³, including time (temporal) trends ¹⁴, at national and local levels
2. To describe variation in suicide rates by sex, age and social class ¹⁵, over the period 1989-2002 ², at national and local levels
3. To examine the relationship between the suicide rate, on the one hand, and the gradient in suicide rates by age and social class ¹⁵, on the other, over the period 1989-2002 ², at local level
4. To examine the relationship between the level of socio-economic deprivation ¹⁵, on the one hand, and the gradient in suicide rates by age and social class ¹⁵, on the other, over the period 1989-2002 ², at local level.
5. To compare individual-level estimates of the relationship between suicide and social class ¹⁵ with area-level estimates of the relationship between suicide and socio-economic deprivation ¹⁵.

In fact, we were unable to fulfil objectives 3 and 4 because the number of suicide deaths at local level was too small to permit meaningful or interpretable analysis. However, we have considerably extended the study by examining in some detail:

• the allocation of deaths to intentional self harm or undetermined categories, including variation by method and geography ( section 3.1)
• methods of suicide, at national and local levels ( section 3.6)
• suicide rates by area deprivation, at national and local levels, including analyses by gender and age group, and calculation of changes in the social gradient ('suicide gap') ( section 3.9) .

These additional analyses take advantage of a greater range of available data and add value by: highlighting the consequences of adopting a broader definition of suicide; and identifying trends in methods of suicide and the influence of area-level deprivation, thereby contributing to the development of appropriate suicide prevention policy and practice responses.

The main focus of this report is the 14 year period leading up to and including 2002, the year in which Choose Life was launched ¹⁶. This is consistent with our intention to provide a detailed picture of suicide in Scotland prior to the implementation of the national suicide prevention strategy and action plan. In relation to the examination of suicide trends over time (at both local and national levels) (objective 1) , the analysis is extended to 2004, in order to provide some indication of stability or change in the suicide rate (key outcome measure) during the first two years of Choose Life implementation.

1.3 Structure of the report

The methods used in this study are set out in section 2 (starting on page 10) , the findings in section 3 (starting on page 15) and the conclusions in section 4 (starting on page 53) . Section 5 (page 56) is a glossary and the references can be found in section 6 (page 57) . A technical appendix starts on page 58. An extensive set of annexes (starting on page 64) provides data on each local authority and health board in relation to suicide trends over time (annexes 1 and 2) , male suicide rates by social class (annexes 3 and 4) , suicide rates by method and sex (annexes 5 and 6) , suicide rates by age group and sex (annexes 7 and 8) and the association between socio-economic deprivation and suicide (annexes 9 and 10) .

« Previous | Contents | Next »