Attempted and completed suicide: Not what we expected?

Abstract

Background

Suicide attempters and suicide completers are two overlapping but distinct suicide populations. This study aims to present a more accurate characterization by comparing populations of suicide attempters and completers from the same geographical area.

Methods

Samples and procedure: All cases of attempted suicide treated at the emergency room of the Corporacio Sanitària i Universitària Tauli Parc de Sabadell in 2008 (n=312) were compared with all completed suicides recorded in the same geographical area from 2008 to 2011 (n=86). Hospital and primary care records were reviewed for sociodemographic and clinical variables. Statistical analysis: Chi-square, ANOVA, and Mann-Whitney U tests were used to identify characteristics related to suicide completion.

Results

Compared to suicide attempters, suicide completers were more likely to be male (73.3% vs. 37.8%; p<0.001), pensioners (73.7% vs. 23.4%; p<0.001), and people living alone (31.8% vs. 11.4%; p=0.006). Suicide completers more frequently presented somatic problems (71.7 vs. 15.7; p<0.001), Major Depressive Disorder (54.7% vs. 27.9%; p<0.001), and made use of more lethal methods (74.1 vs. 1.9; p<0.001). Suicide completers were more likely to have been followed by a primary care provider (50.0% vs. 16.0%; p<0.001). 92.3% of the suicides committed were completed during the first or second attempt.

Limitations

Suicide completers were not evaluated using the psychological autopsy method.

Conclusions

Despite presenting a profile of greater social and clinical severity, suicide completers are less likely to be followed by Mental Health Services than suicide attempters. Current prevention programs should be tailored to the specific profile of suicide completers.

Introduction

It is often assumed that suicide attempters and suicide completers represent the same group of distressed individuals evaluated at different stages on the path toward suicide (Blasco-Fontecilla et al., 2010). However, there is compelling evidence that in fact they are two distinct populations that share certain characteristics (DeJong et al., 2010, Fushimi et al., 2006). In fact, the few comparative studies of these suicide populations performed to date have ignored a number of important aspects. For example, some studies report that more than 60% of suicide completers had consulted their primary care physician in the previous month to suicide (Luoma et al., 2002); however, no comparative data on prior control of suicide attempters and completers at primary care or mental health services has been published. This means that preventive strategies focused on subjects with a high frequency of suicide attempts may not be sufficient to prevent suicide completion. Indeed, a better characterization of the distinct suicide subpopulations is required for effective preventive policies that specifically target each of these subtypes of suicidal behavior (Blasco-Fontecilla et al., 2010).

Suicide attempts are between 10 and 40 times more common than completed suicides (Jimenez-Trevino et al., 2012, Kovess-Masfety et al., 2011, Schmidtke et al., 1996). Among the few studies that have compared different subtypes of suicidal behaviors, some suggest that there are certain variables that differentiate the suicide attempters and suicide completers from each other (DeJong et al., 2010, Fushimi et al., 2006, Michel, 1987). Beautrais (2004) proposed that subjects who made medically serious suicide attempts bear many similarities to suicide completers. Fushimi et al. (2006) suggested that age, sex, method used and history of previous attempts are factors that distinguish between the two populations. Finally, DeJong et al. (2010) indicated that the use of alcohol prior to suicidal behavior, as well as work-related and financial problems, could also be differential risk factors.

The aim of this study was to compare the sociodemographic and clinical characteristics of suicide attempters and completers from the same geographical area, specifically focusing on certain factors that have not been fully studied to date (i.e. prior control of suicide attempters and completers at primary care or mental health services; presence of medical pathology). It is our hope that our results will lead to further characterization of these two populations, as well as provide useful data for prevention programs to help improve effectiveness in the future.