Highlights of this issue

Psychosis – decision-making capacity, psychological treatment and disengagement

Decision-making capacity, being specific to the decision in question, may differ for treatment decisions compared with research participation decisions. In a cross-sectional study of psychiatric in-patients with psychosis, Spencer et al (pp. 484–489) found that, despite severe illness and a lack of decision-making capacity for treatment, decision-making capacity for research was often preserved. Symptoms associated most strongly with a lack of capacity to make research participation decisions were related to disorganised thinking and memory impairments, whereas lack of insight had the largest effect on decision-making capacity for treatment. The authors conclude that in-patient psychiatric units should not be regarded as off-limits for research but rather, where appropriate, recruitment for research should occur in these settings. Another paper in the Journal this month focuses on treatment of psychosis in acute in-patient settings – Jacobsen et al (pp. 490–497) have reviewed the evidence for the use of psychological therapies in this context. Following a systematic scoping review of the topic, the authors identified a clear need to improve the rigour of research in future studies and a need to overcome current difficulties in drawing any conclusions from study findings given the wide range of therapy types, outcomes measured and modes of delivery examined.

Early intervention for psychosis services developed with the aim of improving outcomes following a first episode and/or identified high-risk states but one of the barriers to achieving this aim has been rates of disengagement from services; disengagement rates documented to be up to almost one-third of patients. Solmi et al (pp. 477–483) found a range of factors to be associated with early disengagement from early intervention services in an East Anglia (UK) sample, including having severe hallucinations, milder negative symptoms, polysubstance use, not receiving a first-episode psychosis diagnosis and being employed. The authors conclude that disengagement is a multidimensional construct that requires the development of instruments, intended to assess the variety of associated factors, for use in future research.

Comorbid depression – in schizophrenia and dementia

The presence of comorbid depression in those with other disorders, including schizophrenia and dementia, may increase risks of adverse outcomes, but the comorbidity is often unrecognised. In a large French sample of community-dwelling out-patients with schizophrenia, Fond et al (pp. 464–470) found that current major depressive disorder was present in 28% of patients and was associated with a range of active symptoms including paranoid delusions, avolition, blunted affect and benzodiazepine consumption. Those on antidepressant treatment had lower levels of depressive symptoms but many treated patients remained depressed. Non-remitted patients had higher levels of paranoid delusions and alcohol misuse. In a UK cohort of patients with dementia, identified through an electronic health records database, comorbid depression (identified in 7%) was not found to be associated with mortality (Lewis et al, pp. 471–476). Mortality rates were, however, associated with being single and being of White British rather than Asian ethnicity. The authors comment on the established association between depression and mortality in the general community and hypothesise that many people with depression may have died before the peak onset period for dementia.

In another study of comorbidity, mental–physical multimorbidity, Camacho et al (pp. 456–463) report on a cluster-randomised trial of collaborative care versus usual care. Collaborative care or integrated mental and physical healthcare, was found to be associated with reduced levels of depression at 24 months. Collaborative care also appeared to be cost-effective, with the cost per quality-adjusted life-year gained being within internationally accepted willingness-to-pay thresholds.

The challenges of testing interventions in mental health

In an editorial in the Journal this month, Duncan et al (pp. 451–455) question the dominance of the randomised controlled trial in the context of the need to evaluate complex interventions in mental health. The authors support a realist approach focused on examining causal mechanisms and understanding interactions between interventions, patients and contexts. The authors discuss the potential for positivist and realist approaches to be reconciled.

Interestingly, even in circumstances in which the randomised controlled trial might be considered a highly feasible and effective means of evaluating an intervention in a mental health context, interpreting the results of such trials may be far from simple. Parker (pp. 454–455) highlights the conflicting conclusions drawn by different meta-analyses of antidepressant trials. The author argues that one of the key issues is the non-specific nature of diagnoses of major depression typically included in trials, potentially obscuring differential outcomes for subgroups within the larger and heterogeneous depression groups studied.