There are many different definitions of suicide, and it is important for us to define from the outset what we mean by suicide and attempted suicide. Durkheim (1951), in his important contribution to the taxonomy of suicide, “Le Suicide,” defined suicide as follows:

It is this definition that we use in this book. In particular, we should note that when we talk about “attempted” suicide, we are talking about failed suicide. In other words, the young people in this book who make up the categories involving attempted suicide were drawn from hospital populations. Their suicidal act had resulted in damage to themselves to the extent that they required medical treatment, even though this damage fell short of actual death. Since it is not possible to use young people who have completed suicide in research of this type, those who have failed in the suicide attempt are the closest we can come to a group who have not survived the tragedy of suicide.

The definition proposed by Durkheim has proved useful for purposes of communication among workers in the field and for gathering statistics. It forms the basis of the national statistics available from most countries. However, it has its limitations.

So far, the literature reviewed suggests that depressed adolescents and risk-taking adolescents differ from normal adolescents in their problem-solving, risk-assessment, and decision-making capacity. The S/RT model described in Figure 8.2 suggests that early risk factors are precursors to the development of psychopathology such as depression and problem behaviors (conduct disorder and substance abuse problems). Alcohol, stress, and peer pressure may exacerbate these deficits, while protective factors may offer stability to young people in this situation.

This chapter and the next two describe research that shows empirical support for the S/RT model in three distinct ways. This chapter checks that depressive symptomatology and problem behaviors relate to the mediating constructs: problem solving, hopelessness, protective factors, and decision making. Chapter 10 checks that these mediating constructs relate to suicidal behavior and risk-taking behavior. Chapter 11 tests whether these constructs do mediate between depression and problem behaviors on one side and suicidal behavior and risk-taking behavior on the other.

The S/RT model aims to demonstrate links between cognitive constructs and behavioral outcomes. In previous research and in the literature, adolescents are often grouped by terms such as “depressed,” “problem behaviour,” “risk takers,” and “suicidal.” The design used to test the links in the S/RT model uses these groups. Adolescents can also be measured according to their depressive symptoms, problem behaviors, risk-taking behaviors, and suicidal behaviors. Broadly speaking, we note that depressive symptomatology and problem behaviors are like independent variables; suicidal and risk-taking behaviors are like dependent variables.

Epidemiology

Epidemiological studies indicate that suicide is a leading cause of death among adolescents throughout the industrialized world. The rate of suicide among adolescents increased dramatically over the last part of the 20th century, and this increase is of the greatest concern to the community. In the United States of America, the overall rate of suicide has remained steady but the mean age of those committing suicide has decreased. The rate at which young people commit suicide has risen dramatically (see Table 1). This trend is mirrored in Australia, where the overall rate of suicide for 15- to 24-year-olds is 16.4 per 100,000, among the highest in the industrialized world (UNICEF, 1993).

Age and Gender

The suicide rate among male adolescents, aged 15–19 years, is particularly alarming. In Australia, the Australian Bureau of Statistics (1995) reported a fourfold increase between 1960 and 1995 in suicides for young males. In 1960, 50 males between the ages of 15 and 24 committed suicide, representing a rate of 6.8 per 100,000 of the general population. By 1994 this had increased to 384, or a rate of 27 per 100,000. Female suicides are significantly less common than male suicides (approximately 1:5; Leslie, Stein, & Rotheram-Borus, 2002).

These trends are seen across the world. Figures 1.1 through 1.5 illustrate both the gender discrepancy in suicide and the increase in adolescent rates in a selection of countries. The statistics are the most recent available and were obtained from the World Health Organization, Geneva, 2000.

The studies outlined later in the book refer to adolescents who are described as exhibiting problem behaviors or conduct disorder. Adolescents with these problems invariably engage in higher than normal levels of risk-taking. We examine the factors that predispose these adolescents to risk-taking.

Environmental Variables

FAMILY DYSFUNCTION

If adolescents feel supported by their families, they are less likely to engage in risk-taking behavior. A study that looked at perceived quality of support found that adolescents with either emotional or behavioral problems were four times more likely to report a lack of family support than students without these problems (Garnefski & Diekstra, 1997). Those adolescents with both emotional and behavioral problems were eight times as likely to feel unsupported by their families. Students who indicated a lack of support from more than one system (family, school, or peers) were also likely to have more problems. These results suggest that similar background problems and family risk factors are involved in the development of both emotional and behavioral problems. These factors may include family dysfunction and real or perceived life stressors.

The development of problem behavior has been linked with the same background factors as suicidal behavior (see Chapter 3). A study into family risk factors for adolescent suicide concluded that family dysfunction is likely to be a nonspecific risk factor for the development of psychopathology in children (Wagner, 1997).

The literature now almost universally accepts that individuals who complete or attempt suicide are likely to be suffering from a mental disorder (Mazza & Reynolds, 2001). Lewinsohn et al. (1996) examined adolescents who had attempted suicide and found that 100% of the boys and 94% of the girls had a diagnosable mental disorder. This was a large prospective study using data collected from 1,709 high school students who were assessed on several occasions across a 10-year period from 1985. The researchers were, therefore, able to monitor psychiatric problems associated with suicide before the suicide attempt was made. In other studies, suicidal adolescents have not been prediagnosed and it is more difficult to be certain of the relationship between psychiatric disorders and suicide. The evidence that is available, however, is compelling.

Of 53 adolescent suicides in Finland, all but 3 were diagnosed with a psychiatric disorder (Marttunen, Aro, & Lonnqvist, 1992). Fifty-one percent were diagnosed with depressive disorders, 26% with alcohol abuse or dependence, 21% with adjustment disorder, 17% with conduct disorder or antisocial personality disorder, and 6% (three cases) with schizophrenia. Antisocial behavior during the year preceding the suicide was reported for 45% of the boys and 33% of the girls. In 78% of these cases, a diagnosis of alcohol abuse or depression was also present.

Introduction

This integrative chapter aims to present the Suicide and Risk-Taking (S/RT) model developed from our review of the extant literature and to provide the theoretical basis for that model. The chapter begins with a review of how the model was derived, an explanation of how it differs from a similar model proposed in the literature, and an outline of the component parts of the model. We then briefly revisit the literature concerned with the major factors associated with suicide and risk-taking included in the model in order to provide a rationale for the structure of the model. The chapter proceeds to outline a theoretical framework to explain the way in which the proposed mediating variables, problem solving, hopelessness, protective factors, and decision making are hypothesized to work in suicidal and risk-taking behaviors.

An Earlier Model

In a review article, Yang and Clum (1996) suggested that early environmental factors are linked to suicidal behavior by means of the mediating influence of cognitive factors. In order to reach their conclusion, Yang and Clum examined evidence from other studies to show links between (a) cognitive deficits and suicidal behavior, (b) early environmental factors and cognitive deficits, and (c) early environmental factors and suicidal behavior. They proposed the model of suicidal behavior shown as Figure 8.1.

This model proposes that early negative life events form a background to suicidal behavior, but the exhibition of this behavior is determined by several cognitive factors acting as mediators.

In this chapter we turn our focus to the outcome behaviors – suicidal behaviors and serious risk-taking. This research tests whether these behaviors are related to the mediating constructs in our S/RT model (Figure 8.2). One of our aims was to find whether the cognitive mediating constructs might be useful as potential indicators of a trajectory toward suicidal behavior. If we can identify common deficits in the cognition of suicidal young people, it may be possible to target preventative programs more effectively in order to remediate these deficits. The same reasoning applies to serious risk-taking behaviors. The strategy is to compare suicidal and risk-taking adolescents with each other and with a comparison group of adolescents who do not show these behaviors.

Method

PARTICIPANTS AND GROUP ALLOCATION

All participation was voluntary and all participants received 20 dollars for their time. There were no indigenous Australians in these groups. Participants in the groups were recruited from a variety of sources: through consultant referral from a hospital; through referral from social workers in Juvenile Justice; through referral from psychologists and psychiatrists in private practice; and from a shelter for homeless youth.

Adolescents were interviewed individually, and they generally completed the questionnaires in a session of 1½ hours. In the course of data collection, it became apparent that several of the adolescents were unable to read and write and could not understand the questions even when they were read to them.

The research reported in Chapters 9 and 10 aimed at demonstrating the relationship between the cognitive constructs in the S/RT model and depressive symptoms and problem behaviors (Chapter 9), and the cognitive constructs and suicide and risk-taking behaviors (Chapter 10). The research used between-group comparisons to demonstrate the relationships. The data analyses reported in this chapter directly test whether the nine cognitive variables mediate between depressive symptomatology and problem behaviors on the one hand and suicide and risk-taking behavior on the other. It uses path analysis to do this.

An advantage of the group comparison approach is that, by looking at the mean scores of the groups across the nine cognitive variables and the screening variables, we can make a cross-study comparison. There were three groups in the first set of analyses (Chapter 9) and four groups in the second set (Chapter 10). There was a comparison group in each study. The results from these two comparison groups can be pooled since a discriminant analysis found they were not significantly different from each other.

Path analysis (see Figure 11.1) is an extension of multiple regression that allows one to simultaneously assess the relative importance of two or more independent variables while testing for the influence of a mediating variable.

All the paths, direct and indirect, can be tested for statistical significance. Structural equation model programs such as EQS (Chou & Bentler, 1995) perform this analysis and test the paths for significance.

Although we can identify adolescents in the community who, through a combination of environmental factors and individual factors, are at risk of suicide when placed in a situation of stress, we still do not have a complete picture of the process leading to the act of self-destruction. It is necessary to understand the relationship between the emotional and cognitive processes that allow adolescents to reach the belief that death is the best solution available to them. The following review looks at the cognitive factors most commonly associated with suicide risk in adolescence. They include the concepts of hopelessness, problem solving, coping, and protective factors.

Hopelessness

The role of hopelessness in adolescent suicide is based on the suggestion that how a person thinks about his or her future when depressed is more likely to influence outcomes toward suicidal behavior than feeling states (Beck, 1963, 1967). Although there was some support for this theory in research conducted with adult suicide attempters, the research using adolescents appears to be much less consistent. Rotheram-Borus and Trautman (1988) found that in a group of adolescent Hispanic and black girls, neither hopelessness nor depression predicted suicidal intent. The authors suggested that hopelessness might be a symptom of depression in adolescents rather than an independent cognitive factor predicting suicidal behavior. They also pointed out that hopelessness might not be a good predictor of suicidal intent among minority adolescents.

Cole (1989) found that mood was a more important predictor of suicide than hopelessness, especially for boys.