This work may not be copied, distributed, displayed, published, reproduced, transmitted, modified, posted, sold, licensed, or used for commercial purposes. By downloading this file, you are agreeing to the publisher’s Terms & Conditions.


Suicidal Behavior: Strategies for Prevention at the Individual and Population Levels

Maria A. Oquendo, MD

Published: August 26, 2014




For more information about

see page 845

Suicidal Behavior: Strategies for Prevention at the Individual and Population Levels

This month’s inaugural Focus on Suicide section includes articles that address both predictors for and protectors against suicide risk.

Two of the works report on the use of population or surveillance databases to identify clinical risk factors that may aid in the recognition of patients at risk for suicide attempts. Bhaskaran and colleagues address a question often confronted by clinicians evaluating suicide attempters: Does the fact that a patient expresses disappointment or even anger at having survived an attempt have clinical significance? Using the Suicide Assessment Form in Emergency Psychiatry (SAFE) Database, the authors evaluated 922 suicide attempters’ reactions to surviving the attempt and the effect of the attempt method on risk of reattempt in the next 6 months. Those who reattempted (8.8%) were more likely to report that they were ambivalent about surviving or continued to wish they were dead. This association remained even after accounting for age, sex, depression, substance abuse, and method of the initial attempt. Of note, analyses did not take into consideration the challenges of follow-up wherein some observations may have been censored, so results must be viewed with caution. Moreover, it is not known if the quality of the reaction (anger versus disappointment) has an effect on future risk. Finally, because only 6 months of follow-up time was evaluated, the time period over which risk remains elevated is still unknown. Nonetheless, this critical clinical parameter deserves further study.

In their national cohort study, Webb et al examined risk factors for suicidal behavior in individuals with bipolar disorder. Using Swedish national registers covering a 36-year period, 15,337 persons with bipolar disorder were matched by age and sex to 20 individuals per case from the general population. 14,677 unaffected siblings of those with bipolar disorder were similarly matched to a second general population sample. Outcomes included suicide, suicide attempts evaluated in hospitals, and violent and nonviolent criminal behavior. Bipolar disorder was associated with a nearly 19-fold risk for suicide death and a 14-fold risk for suicide attempt. The risk of violent and nonviolent crime was also increased, but to a lesser degree (5-fold and 3-fold, respectively). Not surprisingly, risk was less pronounced in the unaffected siblings, but still present. Three parameters predicted all 4 outcomes: prior suicide attempts, a documented alcohol/drug disorder, and hospital admission for the first 2 episodes of bipolar disorder, with the latter possibly functioning as a proxy for clinical concern about suicide risk. Thus, both articles suggest that prior attempts and their characteristics contribute to our understanding of individual risk for future suicidal behavior.

Three works provide support for a variety of primary suicide prevention strategies. MacGregor et al studied depressed parents and their offspring longitudinally and found that parents with avoidant attachment tended to be more likely to have offspring who made suicide attempts; further, when offspring did attempt suicide, they had more numerous attempts and severe intent. On the other hand, parents with anxious attachment were not more likely to have offspring with suicide attempts, but when their offspring attempted suicide, they did so with greater intent and suffered more medical damage. Thus, both types of insecure parental attachment traits were associated with more severe suicidal behavior in offspring attempters. Whether interventions to address insecure parental attachment can prevent transmission of risk for suicidal behavior or reduce its severity requires investigation as a potential primary prevention strategy.

Another primary prevention strategy is education of primary care providers, thus far tested mostly in civilian populations. Hochman et al studied 170 active-duty military suicides. Using a retrospective nested case-control design, they noted that although only 27.6% of suicides sought care from a mental health specialist during their entire service time, 38.3% used primary care services the month before death, suggesting that primary care provider education has utility for military suicide prevention.

Finally, Till and Niederkrotenthaler examined websites retrieved in the United States and Austria using a structured content analysis based on current media guidelines for suicide reporting. Utilizing the search term suicide, method-related search terms, and help-seeking-related terms, they found that protective website characteristics were about twice as frequent as harmful ones. As expected, sites retrieved using method-related terms had more harmful than protective characteristics compared to those identified using the term suicide. The opposite was true when help-seeking-related terms were used. Interestingly, US search engines displayed more sites with a preponderance of protective features than Austrian engines. Mounting concern about the negative effects of information posted on the Internet requires active monitoring and intervention. Only through systematic evaluation of content can policy makers and investigators provide recommendations regarding responsible posting. Such evaluation of website content and other primary prevention strategies together can ensure that help and referrals for those in need are easily available before the first manifestation of suicidal behavior.

Related Articles

Volume: 75

Quick Links: