Diabetes and mental illness...

Highlights of this issue

Kimberlie Dean

Diabetes and mental illness

Two papers in the Journal this month examine the association between diabetes and mental illness: one considers survival in those dually diagnosed and the other investigates the impact of antipsychotic treatment on diabetes incidence. Vinogradova et al (pp. 272–277) utilised data from the QRESEARCH database of over 9 million patients drawn from general practices across the UK and found that those diagnosed with schizophrenia or bipolar disorder and diabetes had higher mortality rates than those with diabetes alone. The authors highlight the importance of improving diabetes care for those with severe mental illnesses. Using the Danish population registers, Kessing et al (pp. 266–271) found that those purchasing antipsychotic medication, either first- or second-generation, experienced an elevated incidence of diabetes. They found substantial differences between individual drugs and an increase in risk associated with both duration of treatment and polypharmacy

Near-lethal suicide attempts

In a matched case–control study based in 19 UK prisons, Rivlin et al (pp. 313–319) found that all those who had made a near-lethal suicide attempt had a current psychiatric disorder compared with 62% of the matched controls. The strongest associations were found for major depression and psychosis; alcohol misuse was not found to be associated with suicide attempt and the elevated risk of antisocial personality disorder did not reach statistical significance. In a qualitative study by Biddle et al (pp. 320–325), the main reasons for using hanging as a method of suicide attempt were found to be the nature of the death anticipated (certain, rapid and painless) and the accessibility of materials. The authors discuss the potential for preventive strategies aimed at countering perceptions of hanging.

Cost-effectiveness of treatments for depression

Siskind et al (pp. 291–296) evaluated the comparative costs and benefits of usual care and improved stepped care compared with no treatment at baseline, in a study of treatment for depression among women in Chile. Using the per-capita GDP for Chile as a threshold, the improved stepped-care programme appeared very cost-effective and usual care was found to provide value for money compared with no treatment. In a randomised controlled trial involving a sample of individuals identified in primary care as having a new episode of depression, therapist-delivered online cognitive–behavioural therapy (CBT) was found to be more expensive than usual care but outcomes were improved in the CBT group. Hollinghurst et al (pp. 297–304) concluded that if £20 000 per QALY were considered an acceptable minimum cost threshold, then online CBT delivered by a therapist is likely to be cost-effective.

Mental healthcare transition and cannabis-induced memory impairment

In a study conducted across six mental health trusts in England, Singh et al (pp. 305–312) found that, for most service users, transition from child to adult mental healthcare is poorly planned, poorly executed and leads to a poor experience for those involved. A significant proportion of young people who needed referral for transition were not referred and less than 4% of those accepted by adult services experienced optimal transition. Using a novel methodology in which the level of cannabinoids in the cannabis actually smoked by each study participant was analysed, Morgan et al (pp. 285–290) found that those who smoked cannabis low in cannabidiol demonstrated marked impairment in prose recall. Cannabidiol content did not affect psychotomimetic symptoms, which were found to be elevated in all those assessed when intoxicated.

Disability in those at high-risk of psychosis

Velthorst et al (pp. 278–284) compared baseline differences in disability between individuals at clinical high risk of psychosis who made a transition to psychosis compared with those who did not. From a range of social disability domains, difficulties in getting along with others was found to contribute to the prediction of transition to psychosis. The authors comment that in addition to improving psychosis prediction in such samples, an assessment of disability is likely to be important in identifying unmet needs for both the individual and their family.