Unintended consequences

The award-winning scientific documentary maker and director Ann Druyan said ‘the greatest thing that science teaches you is the law of unintended consequences’.Reference Mellor¹ Considering and explaining to patients the risk of unexpected outcomes from proposed treatments is part of the daily challenge of our clinical practice. How does one consent to or inform people of risks that are unknown and unforeseen? Unintended consequences are an important feature in three of this month's BJPsych articles.

The national STOMP campaign has proven successful in reducing inappropriate antipsychotic use for challenging behaviour in people with intellectual disability. However, in their important analysis article this month, Branford et al (pp. 488–493) use prescribing data from NHS Digital to argue that this campaign may have unexpectedly led to an increase in antidepressant prescribing in this group instead. They are concerned that this could be resulting in unintended harms as antidepressants are prescribed without indication, for indeterminate lengths of time, with little evidence for prolonged use in people with intellectual disability. Their words are an important reminder to me of the frequent lack of prescribing evidence for people in this vulnerable patient group.

A short report by Herzog et al (pp. 485–487) uses ecological momentary assessment – involving repeated, real-time monitoring using smart phones – to demonstrate that non-suicidal self injury reduces suicidal ideation in the hours following a self-harm event. The authors surmise that this reduction in short-term suicidality may be part of people's motivation to self-harm. However, they believe that there could be unintended consequences: this short-term relief may lead to repeated self-harm and a resulting desensitisation towards the associated pain and fear, ultimately increasing the longer term risk of completed suicide.

One well-established unintended consequence of admission to psychiatric in-patient beds is a substantial short-term increase in suicide risk following discharge. A comprehensive cohort study by Musgrove et al (pp. 468–475) using data from the Clinical Practice Research Datalink over nearly two decades confirms and strengthens this finding in the UK and shows that the suicide risk is elevated for both working age and older adults, particularly in the first 3 months post-discharge.

Maintaining a biopsychosocial approach to depression

A combination of this month's papers offers a biopsychosocial approach to the treatment of depression, particularly in hard-to-treat cases. From a ‘bio’ perspective, Furukawa et al (pp. 440–447) use systematic review and meta-analytic methodology to show that aripiprazole augmentation for refractory depression may achieve most of its efficacy in the lower range of its licensed dose (2–5 mg) and may offer no additional benefits above 10 mg. This is despite many guidelines recommending a wide treatment range between 2 and 15 mg. Enneking et al (pp. 476–484) used a functional magnetic resonance imaging study to show that patients with either remitting or relapsing depression showed different levels of brain activity within different emotion processing regions of the brain.

Addressing the ‘psycho’ and ‘social’ aspects, Angelakis et al (pp. 459–467) review the effectiveness of varying complexities of cognitive–behavioural therapy (CBT) in reducing depression in adults, including complex and ultra-complex CBT with a number of different added ‘psycho-social’ therapeutic components, such as psychoeducation or training in social or relaxation skills. They found that people under 59 years old, with comorbid conditions were most likely to benefit from ultra-complex or complex CBT – interestingly and perhaps counterintuitively best delivered without the help of mental health specialists.

A more abstract area of research looks at the use of prognostic models to determine the likelihood of relapse or recurrence in people with remitted depression (Moriarty et al (pp. 448–458)). Unfortunately, the systematic review found a lack of robust research, with most studies showing medium or high risk of bias, suggesting that it is too soon to think we can predict individual depression outcomes.

Developments in schizophrenia genetic research

This sense of being too soon to know exactly what the research is telling us to do clinically is an important theme in Professor Curtis's editorial (pp. 437–439) about the recent discovery of further schizophrenia risk genes in the SCHEMA study. He takes a measured and fascinating look at what these findings might mean for the way we think about and diagnose schizophrenia. The Editorial is pragmatic about what it might mean practically for people's treatment in the future, acknowledging that at the present the prospect of using genetic information to drive personalised medicine in mental health is unlikely.

Although genetic abnormalities only account for a small proportion of cases of schizophrenia, these results importantly and unequivocally demonstrate a link between mental illness and physical processes; however, they do not negate the importance of psychological and social aspects: ‘the importance of these findings is to insist that the “bio” not be excluded completely from a biopsychosocial approach to psychiatry, not that it should be considered to explain away other relevant factors’.