Suicide – prediction, environmental context, trends and the role of personality

Several papers in the BJPsych this month are focused on suicide, taking perspectives ranging from a broad societal to a specific individual level. Chan et al (pp. ) undertook a review of cohort studies that examined risk of suicide following an episode of self-harm. Their meta-analysis of risk factor studies identified four key factors – a history of previous episodes of self-harm, suicidal intent, physical health problems and male gender – while the review of risk scales uncovered a lack of evidence to support the use of any individual scale. The authors comment on the limited clinical utility of considering four factors that are likely to be common in clinical populations and they call on clinicians to instead undertake comprehensive assessments focused on the risks and needs specific to the individual who has self-harmed. In a linked editorial, Mulder et al (pp. ) echo this call for a move away from risk categorisation to individual engagement with patients, considering their specific needs and circumstances. The authors also highlight the impossibility of predicting rare events such as suicide and note the widespread use of suicide risk scales in the absence of supportive evidence.

Two papers in the BJPsych this month consider the impact of societal factors on suicide rates – timing of suicide in relation to provision of mental health services and national economic conditions. Kapur et al (pp. ) explored the possibility of a ‘weekend effect’ on suicide rates such as has been reported in relation to deaths among those admitted to hospital at different times of the week. The authors found that the incidence of suicide was actually lower at the weekends for in-patients, those within 3 months of discharge and patients under the care of crisis resolution home treatment teams. The authors further found no increase in suicide rates in August, when junior doctors change rotations, compared with other months. In an editorial by Fountoulakis (pp. ) the assumed need to intervene at the general population level when economic conditions are unfavourable rather than targeting vulnerable subgroups such as those with mental illness is questioned. Arguments are made that this approach is not only ineffective but may have a negative impact on vulnerable mental health patients who should have received the benefit of a targeted approach.

Redmore et al (pp. ) examined trends in suicides and accidental deaths in adolescents in England and Wales, taking a long-term perspective. Over the period 1972 to 2011, adolescent suicide rates were stable for those aged 10–14 years while accidental deaths declined. For those aged 15–19 years, there was a peak in suicide rates in 2001 followed by a decline; associations were found between higher suicide rates and both male gender and level of deprivation. The last paper addressing suicide in the BJPsych this month looks at the role of axis II personality disorders on risk of both suicide and attempted suicide in a Chinese sample. Tong et al (pp. ) found that personality disorders conferred an increased risk of suicidal behaviour but their prevalence in both the community and among those with suicidal behaviour was low.

Observational and experimental approaches to understanding PTSD

Post-traumatic stress disorder (PTSD) is the focus of a number of papers in the BJPsych this month with a wide range of methodologies applied. In a cross-national comparison of PTSD prevalence, Dückers et al (pp. ) found that while the expected association between trauma exposure and PTSD prevalence was confirmed at the country level, greater country vulnerability was associated with a decreased occurrence of PTSD. The authors explore the possible explanations for this ‘vulnerability paradox’, including the possibility that deprived high-vulnerability countries might foster conditions that increase resilience to trauma. Moving from country level to the individual level, Heir et al (pp. ) analysed data from a survey conducted 10 months after the 2011 Oslo bombing to explore the role of perceived life threat on PTSD in two groups – those directly and those indirectly exposed to terror. The expected association between perceived threat and PTSD among those directly exposed to the trauma was confirmed and was also identified among those indirectly exposed. The authors discuss the implications of their findings for treatment approaches, including the potential benefits of attempting to moderate perceptions of threat among the latter group.

Two papers in the BJPsych this month evaluate treatments for PTSD. ter Heide et al (pp. ) tested eye movement desensitisation and reprocessing (EMDR) in an adult sample of refugees and asylum seekers with chronic PTSD and found no differences in safety or efficacy when compared with stabilisation as usual. The authors comment on the possibility that the efficacy of EMDR might have been improved by increasing the number of sessions, by targeting refugees referred to general rather than specialist mental healthcare or by combining treatments. They highlight the accumulating evidence refuting the notion that directly targeting traumatic memories in refugee populations may cause harm. In a study of trauma-focused treatment, either prolonged exposure or EMDR therapy, in PTSD patients with psychosis, van Minnen et al (pp. ) found that treatment benefit was not reduced in those with the newly introduced dissociative subtype of PTSD. The authors also highlight their finding that those with the dissociative subtype were not more likely to drop out of treatment either. Also addressing the diagnostic evolution of PTSD and the current DSM-5 conceptualisation, Rosen (pp. ) examines the impact of Robert Spitzer's proposals to improve the validity of PTSD in DSM-5. Rosen concludes that Spitzer's recommendations were largely ignored and that his goal of ‘saving PTSD from itself’ may need to be revisited post-DSM-5.