Abstract
In recent years the focus on risk, and risk assessments, within mental health services has increased markedly. This is largely the consequence of the emphasis being placed upon dealing with adverse events. While it is clearly important to identify contributing factors to such events with the aim of improving services, clear clinical predictors of such events remain elusive. Research studies are able to identify risks but translation of the research findings is problematic because the predictive value of these risks is low.
For example, suicide following discharge from a psychiatric unit is a particularly tragic event and one where predictive risk factors would have obvious utility. In this month's Australian and New Zealand Journal of Psychiatry (ANZJP) Large et al, [1] who performed a meta-analysis of studies examining suicide in the year following discharge from a psychiatric hospital, found that a risk index based on associations they identified with a history of self-harm, depressive symptoms, reports of suicidal ideation, recent social difficulties and an unplanned discharge, had low utility. Extrapolating further, they estimated that 3% of patients identified as “high risk” can be expected to suicide, but, of even greater concern 60% of those that suicide would have been categorised as “low risk”. Continuing with this theme, Mulder [2] provides a thoughtful editorial that addresses four myths surrounding suicide risk assessment; (i) that certain risk factors predict the likelihood of suicide in an individual patient; (ii) that people in high-risk groups are likely to die by suicide; (iii) focusing resources on high-risk groups reduces the number of suicides and that (iv) treating risk factors will result in a lower suicide rate. Ironically risk assessment tools for predicting aggressive behaviour also perform poorly, however Phillips et al [3] found that an unstructured clinical risk assessment based on clinical expertise had some utility. This finding emphasises the value and importance of clinical experience and judgement in assessing risk.
Another area of interest is child sexual abuse, which is unfortunately common and not surprisingly a recognised risk factor for a variety of psychiatric disorders. Its prevention is a clearly a priority. Martin et al [4] in a well designed longitudinal study, found that penetrative child sexual abuse was associated with features of social disadvantage; poor maternal education, maternal smoking and unmarried status. These features may be associated with other socio-environmental aspects of social disadvantage that were not measured in the study, such as heightened crime and antisocial activity that could also be associated with the risk of child sexual abuse. Public health measures to reduce the risk of child sexual abuse will need to address means of reducing social disadvantage with an emphasis on providing good education. Highlighting this by focusing research on this issue is to be commended.
At the other end of the age span, Djernes et al, [5] report increased mortality among the depressed frail elderly when compared with their frail non-depressed counterparts and a community cohort of elderly persons. The depressed group made less use of ‘somatic’ hospital services than the other groups of elderly persons, but made more use of psychiatric services. The authors point out that there needs to be more effective follow-up regimes and collaborative care between old age psychiatry services and primary care.
An area that is receiving some recognition is that of early psychosis intervention programs. More recently these have been in the news because of the Australian budget announcements. Investigating this further, Wong et al [6] found that an early psychosis program in Hong Kong was more cost effective in improving outcome than the pre-existing service. There were higher medication costs for the early psychosis program (higher use of second-generation antipsychotics) but lower hospital costs compared with the pre-existing program. A potentially novel way of testing whether a second-generation antipsychotic is having the desired effect is described in the study by Blessing et al [7]. This group followed up a clinical observation that patients with schizophrenia have increased cutaneous vasoconstriction (cold hands) and that in animal models second-generation antipsychotics seemingly reduced this. Using thermal imaging they showed that hand temperature increased the following a dose of a second-generation antipsychotics and that this change was correlated with the clinical effect.
The gold standard randomised controlled trial has become somewhat tarnished over the past few years, especially the generalizability of the findings to real world patients. There is a need for more effectiveness trials examining the outcomes in the real clinical settings. This is precisely what Khoo et al [8] have done and shown impressive and sustained positive outcomes for patients with PTSD treated at their six-week, group-based CBT program.
The letters section of this month's issue of the journal provides us with some interesting clinical observations ranging from epidemic Koro [9], adverse events of drugs [10] such as duromine-induced psychosis [11], and late onset neutropenia with clozapine [12] and the effects of new treatment modalities for example the effectiveness of rTMS in reducing auditory hallucinations [13]. These brief communications not only make fascinating reading but also make important contributions in advancing knowledge and exploring new ideas. As such I strongly encourage the submission of such reports to the ANZJP, especially by trainee psychiatrists and interested students.