A comparison of specific positive future expectancies and global hopelessness as predictors of suicidal ideation in a prospective study of repeat self-harmers

Abstract

Background

Hopelessness and the lack of positive future expectancies have been related to suicidality. This is the first study to compare the power of positive future expectancies and global hopelessness in the prediction of suicidal ideation. In short, are specific positive expectancies or global hopelessness attitudes more closely related to suicidality?

Method

One hundred and forty four adults hospitalized following a suicidal self-harm episode completed a range of clinical and psychological measures in hospital and were followed up approximately 2.5 months after discharge. All participants reported at least one other self-harm episode in addition to the index episode.

Results

Hierarchical regression analyses confirmed that specific positive future expectancies were better predictors of Time 2 suicidal ideation than global hopelessness. In addition, as hypothesized, negative future thinking was not independently associated with suicidal ideation.

Limitations

Short-term follow-up.

Conclusions

Specific, idiographic expectancies for positive events (i.e., positive future thinking) are more important predictors of suicidal ideation than global attitudes of hopelessness. Unlike global hopelessness, they provide more options for intervention (e.g., identifying life goals and plans). These findings are particularly noteworthy given the widespread use of measures of global hopelessness. The theoretical and clinical implications are discussed.

Introduction

Approximately one million people kill themselves globally per annum (World Health Organization, 2000) and a history of self-harm is the best predictor of completed suicide (Maris, 1991). Moreover, as many as two thirds of those who complete suicide have attempted suicide at least once previously (e.g., Appleby et al., 1999, Isometsa and Lonnqvist, 1998). Consequently, research in suicidology, including the present study, often focuses on non-fatal suicide attempters to improve our understanding of the etiology of suicidal ideation and behaviour.

Studies focused on global hopelessness, defined as global attitudes of pessimism for the future, have been common and fruitful in suicide research for many decades. Indeed, the link between hopelessness and suicide is powerful and well established, such that it is a key risk factor for suicidality (Beck et al., 1990, Kazdin et al., 1983, O'Connor and Sheehy, 2000, O'Connor and Sheehy, 2001). Moreover, global hopelessness mediates the relationship between depression and suicidality (Salter and Platt, 1990), as well as predicting suicidal repetition (Petrie et al., 1988) and completed suicide (Beck et al., 1989). However, up until the 1990s (MacLeod et al., 1993) it was not clear which components of pessimism for the future were most pernicious. Specifically, is the absence of positive expectancies for the future functionally equivalent to the presence of the negative expectancies for the future – and indeed are these two conceptualizations of the future differentially associated with suicide risk?

To investigate this issue, MacLeod et al. (1997) devised the Future Thinking Task which requires participants to generate (within time constraints) events that they are looking forward to (positive future expectancies) and those which they are worried about (negative future expectancies). In a series of studies, MacLeod and others thereafter found that positive and negative future expectations are not functionally equivalent and that the lack of positive future thinking is particularly associated with suicide risk, while negative future expectations, for the most part, are not independently associated with suicidality (MacLeod et al., 1998, MacLeod et al., 1997, MacLeod et al., 1993, Hunter and O'Connor, 2003, O'Connor et al., 2007, O'Connor et al., 2004, O'Connor et al., 2000). In short, suicidal individuals are impaired at generating positive future expectancies. What is more, the specific and deleterious effects of the lack of positive future thinking are independent of verbal fluency, depression and negative attributional style (MacLeod et al., 1997, O'Connor et al., 2000, O'Connor and Cassidy, 2007).

The conceptualization of suicidal behaviour as an inability to generate positive future expectations fits with a predominant model of suicidal behaviour: The Cry of Pain model (Williams, 2001, Williams et al., 2005a, Williams et al., 2005b; see also O'Connor, 2003). Williams' model builds upon Baumeister's Suicide as Escape from Self (1990) account of suicide by viewing suicidal risk as a response to entrapment rather than escape per se. In short, the Cry of Pain model argues that suicidal behaviour is reactive, a response to a stressful situation that has three components: defeat, no escape and no rescue. Accordingly, if you have fewer positive future expectancies (low potential rescue), this may increase your suicide risk because it increases the likelihood that you perceive yourself to be in a state of entrapment which is inescapable. Specifically, we would argue that enjoyable and meaningful things (i.e., positive future thoughts), if present, would rescue people from their misery, despair and unbearable psychological pain – thereby reducing thoughts of entrapment. In addition, one's capacity to generate and pursue future expectations is akin to the identification, pursuit and attainment of goals as described by Carver and Scheier in their model of self-regulation (e.g., Carver and Scheier, 1998).

Despite the growing corpus of cross-sectional, case-control and prospective studies drawn from both clinical and non-clinical populations which demonstrate this positive future thinking effect, it has yet to be determined whether positive future thinking is indeed a proximal, prospective predictor of suicidality beyond the effects of hopelessness – an established suicide risk factor. Consequently, the central aim of this study is to compare the relative importance of global, self-report hopelessness (as assessed by the Beck Hopelessness Scale; Beck et al., 1974a, Beck et al., 1974b) and specific, idiographic expectancies for positive events (i.e., positive future thinking) in the prediction of suicidal ideation.

In the present investigation, we recruited repeat self-harm patients who completed a range of clinical and psychological measures within 24 h of a self-harm episode and then followed them up again, on average, 2.5 months later. We limited our inclusion to repeaters given that a history of self-harm is a stronger predictor of repetitive self-harm and completed suicide than no self-harm history (e.g., Appleby et al., 1999). In addition, given that the suicidal intent of a self-harm episode rather than the seriousness is often a better predictor of repeat suicidal behaviour and completed suicide, we focussed on those self-harmers who reported suicidal intent (Hawton, 2000, Skegg, 2005). As the main aim of the present study was to test empirical research questions rather than to determine clinical outcome per se, we chose a relatively short follow-up period (i.e., 2.5 months) to minimize participant attrition but at the same time allowing for a significant change in our outcome variable (i.e., suicidal ideation) between Time 1 and Time 2 (similar to Spirito et al., 2003, O'Connor et al., 2007).

In short, we addressed one central research question. Are specific positive future expectancies more important than global attitudes of hopelessness in the prediction of suicidal ideation? In addition, we hypothesised that negative future expectancies would not be independent predictors of suicidal ideation.