An integrated framework for suicide risk prediction

Suicide is a major concern in society. Despite of great attention paid by the community with very substantive medico-legal implications, there has been no satisfying method that can reliably predict the future attempted or completed suicide. We present an integrated machine learning framework to tackle this challenge. Our proposed framework consists of a novel feature extraction scheme, an embedded feature selection process, a set of risk classifiers and finally, a risk calibration procedure. For temporal feature extraction, we cast the patient’s clinical history into a temporal image to which a bank of one-side filters are applied. The responses are then partly transformed into mid-level features and then selected in 1-norm framework under the extreme value theory. A set of probabilistic ordinal risk classifiers are then applied to compute the risk probabilities and further re-rank the features. Finally, the predicted risks are calibrated. Together with our Australian partner, we perform comprehensive study on data collected for the mental health cohort, and the experiments validate that our proposed framework outperforms risk assessment instruments by medical practitioners.

Suicide among male road and rail drivers in Australia: A retrospective mortality study

Objectives: This paper aims to describe the epidemiology of suicide among males employed in driving occupations (road and rail) compared to other male occupations in Australia. Methods: Suicide cases among road and rail drivers were extracted from a national dataset of occupationally coded suicide cases for the period 2001 to 2010. Suicide rates per 100 000 were calculated and standardised using the Australian standard population (2001). Incidence rate ratios (IRR) with 95% confidence intervals were calculated using Mantell Haenszel rates and compared to all employed suicide cases. Results: The majority of suicides in this occupational category occurred in truck drivers, followed by road and rail drivers. 98% of these suicides were among males; hence only males were included in further analyses. The age-standardised rate of male suicide among Road and Rail drivers over the period 2001 to 2010 was 22.6 per 100 000 (95% CI 19.2 to 25.9). The IRR of suicide in this occupational group compared to other male occupations was 1.42 (95% CI 1.26 to1.60). Conclusions: Suicide among Road and Rail drivers is higher than in the other male occupations. Suicide prevention initiatives addressing these risk factors, while also providing access to treatment for those at risk, are clearly needed.

The management of adolescents at risk of suicide by Australian general practitioners: issues in practice and confidence

This study involved an anonymous survey of 41Victorian GPs regarding their diagnostic and treatment practices with adolescent patients with depression and/or suicide ideation. The results indicated that the majority of respondents correctly diagnosed the level of depression and the risk of suicide in a case scenario. Although they commonly asked some of the questions related to an assessment of suicide risk, they rarely conducted a comprehensive risk assessment and the level of referral to telephone and internet crisis services was poor. Most GPs indicated a lack of confidence in their ability to detect and manage depression and suicide in this population and strongly emphasized a need for more training.

Euthanasia and assisted suicide : a liberal approach versus the traditional moral view

Within the context of the debate over the recent suspended sentence given to John Stuart Godfrey by Underwood J in the Supreme Court of Tasmania for assisting his elderly mother with her suicide, this article examines some of the more popular arguments for and against the moral acceptability of euthanasia and assisted suicide. This article considers the arguments put forward on the “difference principle” by Rachels and Nesbitt before critically examining the liberal approach to the euthanasia issue as proposed by Kuhse. It is argued that whilst Kuhse is correct to reject the difference principle, she does so for the wrong reasons. The penultimate section of the article provides an overview of the traditional moral view against killing. The final part assesses whether the arguments put forward by proponents of the liberal approach are capable of overcoming this view.

Impact of an empowerment-based parent education program on the reduction of youth suicide risk factors

Purpose: To evaluate the impact of parent education groups on youth suicide risk factors. The potential for informal transmission of intervention impacts within school communities was assessed.

Methods: Parent education groups were offered to volunteers from 14 high schools that were closely matched to 14 comparison schools. The professionally led groups aimed to empower parents to assist one another to improve communication skills and relationships with adolescents. Australian 8th-grade students (aged 14 years) responded to classroom surveys repeated at baseline and after 3 months. Logistic regression was used to test for intervention impacts on adolescent substance use, deliquency, self-harm behavior, and depression. There were no differences between the intervention (n = 305) and comparison (n = 272) samples at baseline on the measures of depression, health behavior, or family relationships.

Results: Students in the intervention schools demonstrated increased maternal care (adjusted odds ratio [AOR] 1.9), reductions in conflict with parents (AOR .5), reduced substance use (AOR .5 to .6), and less delinquency (AOR .2). Parent education group participants were more likely to be sole parents and their children reported higher rates of substance use at baseline. Intervention impacts revealed a dose-response with the largest impacts associated with directly participating parents, but significant impacts were also evident for others in the intervention schools. Where best friend dyads were identified, the best friend’s positive family relationships reduced subsequent substance use among respondents. This and other social contagion processes were posited to explain the transfer of positive impacts beyond the minority of directly participating families.

Conclusions: A whole-school parent education intervention demonstrated promising impacts on a range of risk behaviors and protective factors relevant to youth self-harm and suicide.

Suicide among psychiatric patients : a case-control study

Objective: To examine patient- and treatment-based differences between psychiatric patients who do and do not die by suicide. Method: By linking databases of deaths and psychiatric service use in Victoria, we compared 597 cases who suicided over 5 years with individually matched controls. Results: Cases and controls could not be distinguished on the majority of patient- or treatment- based characteristics. The exceptions were that cases were more likely to be male, less likely to be outside the labour force, more likely to have recent contact with inpatient and community services, and more likely to have a registration as their last contact. Conclusions: Patients who suicide 'look' similar to those who do not, suggesting prevention approaches should ensure that all psychiatric patients receive optimal care, including appropriate detection, diagnosis, assessment and treatment of mental health problems, and careful, individualised assessment of suicide risk.

Inter-annual rainfall variations and suicide in New South Wales, Australia, 1964-2001

The suicide rate in New South Wales is shown to be related to annual precipitation, supporting a widespread and long-held assumption that drought in Australia increases the likelihood of suicide. The relationship, although statistically significant, is not especially strong and is confounded by strong, long-term variations in the suicide rate not related to precipitation variations. A decrease in precipitation of about 300 mm would lead to an increase in the suicide rate of approximately 8% of the long-term mean suicide rate.

Prevention of suicide : police powers, parliamentary intent and judicial interpretation

While in most countries suicide is no longer a crime, it is also acknowledged that the state has an interest in the preservation of human life, prevention of suicide, and protection of vulnerable persons from harming themselves. In a civil, secular and democratic society, however, the public law principle of state protective powers has to be balanced against the private law principle of personal autonomy (personal self-determination). Under the doctrine of autonomy, competent adults of sound mind can make legally binding voluntary choices, including the so-called ‘death-choice’ (refusal of life-sustaining or life-prolonging treatment as well as suicide). To add to the complexity, whereas the powers of the state in relation to suicide and its prevention have been codified, the concepts of personal autonomy and personal liberty are grounded in common law. Stuart v Kirkland–Veenstra [2008] VSCA 32, which is at present being considered by the High Court of Australia, exemplifies tensions that arise in the suicide-prevention area of jurisprudence. This article explores powers and duties of police officers in relation to suicide prevention and the notion of mental illness by reference to the Kirkland–Veenstra case, the relevant statutory framework and the common law.

The role of loneliness in prison suicide prevention and management

Suicide in prisons has been seen as a public health concern reaching crisis proportions around the world. In this study, data from 60 inmates in a South Australian remand facility were used to examine the relationship between loneliness and known predictors of self-harm, such as depression and hopelessness. As predicted, the results suggested that prisoners who scored higher on a measure of loneliness reported higher levels of depression, hopelessness and indicators of suicidal behaviour. The implications of these findings for suicide management are discussed. It is concluded that whilst specialist mental health services are needed to treat problems such as depression, loneliness is something that may be managed by correctional staff with no professional experience or training.

Assessing the risk of suicide at triage

Mental health related presentations to Australian emergency departments are steadily increasing. There is a growing incidence of depression, substance abuse, and other mental illnesses in the Australian population. Mental health problems will contribute 15% of the total world disease burden by 2020. Triage nurses are pivotal to the early detection and management of mental health problems.

The rapid assessment of mental health presentations at triage requires skill, knowledge, experience and confidence. One of the more complex aspects of triage is suicide risk assessment.

Legal bans on pro-suicide web sites : an early retrospective from Australia

There are worldwide concerns that pro-suicide web sites may trigger suicidal behaviors among vulnerable individuals. In 2006, Australia became the first country to criminalize such sites, sparking heated debate. Concerns were expressed that the law casts the criminal net too widely; inappropriately interferes with the autonomy of those who wish to die; and has jurisdictional limitations, with off-shore web sites remaining largely immune. Conversely, proponents point out that the law may limit access to domestic pro-suicide web sites, raise awareness of Internet-related suicide, mobilize community efforts to combat it, and serve as a powerful expression of societal norms about the promotion of suicidal behavior.

Dancing with death : the grey area between suicide related behavior, indifference and risk behaviors of heroin users

Suicide-related behavior (SRB) among heroin users is a complex and multifaceted continuum, including such fringe areas as indifference and "risky" behavior. The article investigates the nuances and intersections of SRB, using qualitative semi-structured interviews with 60 regular heroin users recruited primarily from syringe programs in Geelong, Australia. Twenty-eight percent of interviewees reported a previous suicide attempt and 45% reported serious consideration of it. Types of SRB reported included: Suicide attempts, instrumental suicide-related behaviors, suicidal ideation, indifference and risk-taking thoughts and behaviors. Heroin users engage in much behavior which inhabits a grey area of SRB. The use of a nomenclature which addresses the elements of lethality and intent improves the ability of research to properly define and categorize SRB in drug-using populations. But the categories should not be overinclusive; indifferent attitudes towards death and risk-taking behaviors can sometimes be a functional response to the risk environment of heroin users.