Highlights of this issue

Outcomes in first-episode psychosis

Three papers in the Journal this month consider issues related to outcomes and treatment in first-episode psychosis. Crossley et al (pp. ) reviewed 15 randomised controlled trials of antipsychotic treatment in first-episode psychosis and found no evidence for differences in discontinuation rates or effect on symptoms between atypical and typical antipsychotics. Clear differences in side-effect profile were found, with atypicals causing more weight gain and typicals causing more extrapyramidal side-effects. In a longitudinal study based in India, Saravanan et al (pp. ) found that although baseline levels of illness insight and psychopathology did not predict outcome at 1 year, early improvement in either did predict a good outcome. The authors also found duration of untreated psychosis to be a predictor of relapse and functional impairment. Schennach-Wolff et al (pp. ) also found that early improvement in symptoms predicted outcome in their sample of individuals identified from centres throughout Germany. The authors concluded that an improvement of at least 30% in Positive and Negative Syndrome Scale total score at week 2 may be necessary to achieve response and remission.

Suicide risk in China and internet addition in Hong Kong

Tong & Phillips (pp. ) found that suicide risk among those with a mental disorder varied by mental disorder diagnosis – with mood disorders conferring the greatest risk – and by demographic cohort – with urban residence being associated with greater suicide risk than rural living. The authors recommend consideration of non-mental-health-based interventions for prevention of suicide in rural communities within low- and middle-income countries. Fu et al (pp. ) utilised data from a two-wave panel household survey of adolescents in Hong Kong to examine the concept of internet addiction. The authors found a prevalence rate of 6.7% based on a threshold of reporting at least five symptoms and, in addition, found an association between the number of internet addiction symptoms reported and 1-year changes in suicidal ideation and depressive symptoms.

Cigarettes and depression, cannabis and psychosis

Many studies have found an association between cigarette smoking and depression but whether the association reflects an underlying causal link remains uncertain. Utilising data from a longitudinal study based in Christchurch, New Zealand, Boden et al (pp. ) found that the association between nicotine dependence and depression persisted after accounting for non-observed fixed and observed time-dependent sources of confounding. In addition, using structural equation modelling, the authors were able to report that the best-fitting model was one in which the direction of causality was from nicotine dependence symptoms to depressive symptoms. Using a structured time-sampling technique, Henquet et al (pp. ) found that daily-life cannabis use predicted increases in positive affect for both patients with psychosis and healthy controls, while decreases in negative affect and increases in hallucinatory experiences were found only in the patient group.

Understanding the therapeutic effects of ECT

Yatham et al (pp. ) found that treatment with electroconvulsive therapy (ECT) led to a down-regulation in serotonin receptors when positron emission tomography scans were compared at baseline and following ECT in a group of individuals with antidepressant-refractory depression. The authors also found some evidence of a correlation between receptor down-regulation and improvement in depressive symptoms.

Carer distress and abusive behaviour

In a longitudinal study of abusive behaviour by carers towards individuals with dementia, Cooper et al (pp. ) found that the prevalence of self-reported abusive behaviour increased over time. An increase in abuse score was predicted by an increase in anxiety and depression and by less domiciliary care at baseline.