Quite an experience to live in fear, isn't it?

In-patient suicides are particularly devastating for relatives, other patients and staff. Inevitably, our wards contain the most vulnerable individuals, but it is that word ‘contain’ to which we turn – a purportedly safer and monitored space during such periods. It is therefore an especially troubling concern, not infrequently raised, that this hospital environment might precipitate such acts during admission or immediately following discharge. Professors Large and Kapur debate correlation and causation (pp. 269–273), providing thoughtful counterpoints and frequent areas of agreement; the lack of good evidence on putative hospital harms is striking, although the practical and ethical challenges of such work are equally considerable. It would seem simultaneously true that in-patient units save lives and can be highly distressing places for some; I recommend the piece to stimulate ward multidisciplinary team reflective practice. Ribeiro et al take a different tack (pp. 279–286), meta-analysing the magnitude and utility of depression and hopelessness as suicide risk factors. They evaluated longitudinal studies, as – unlike cross-sectional work – these can stratify risk factors, not just identify correlates. Significant methodological issues were noted in this literature, and although depression and hopelessness increased risk for suicidal acts and death, their overall predictive values were weaker than anticipated.

Failure to respond to first- or second-line (or third- or fourth-line) interventions in major depressive disorders (MDD) is common – indeed almost inevitable in the cohorts referred to secondary care. However, while guidelines tick off various options for refractory care, they can feel like somewhat arbitrary lists. Every clinician is very familiar with the challenge of whether to do ‘more of the same’ with a different drug from a familiar class, or to test something perhaps more exotic but less evidenced. McAllister-Williams et al (pp. 274–278) propose an expert consensus framework for what they label ‘multi-therapy-resistant MDD’. It feels a very useful aid for supporting clinicians’ decision-making in these difficult but common scenarios.

I've seen things you people wouldn't believe

Akena and colleagues (pp. 301–307) note the challenges of diagnosing depression in low-literacy individuals, an educational status that affects up to one-third of sub-Saharan Africans. They report on a novel visual scale aimed at overcoming the problems of standard written batteries, validating it against a structured diagnostic interview and demonstrating it to be an appropriate screen for depression. The figures used, illustrated in this month's journal, provide a remarkable visual representation of symptoms that we usually describe verbally or test through text.

Continuing the theme of similar-but-different, some evidence shows affective disorders to be more common among individuals with intellectual disabilities than in the general population, but good data have been lacking, especially about mania. Cooper and colleagues update us (pp. 295–300), reporting on a prospective cohort of more than 600 people with mild to profound intellectual disability. The incidence of depression was similar to that in the wider population, but the higher prevalence implies that it is either under-treated or a more problematic condition when it occurs. Curiously, despite the fact that 22% of this entire cohort were on mood stabilisers – mainly for epilepsy – the incidence of mania was significantly greater than expected, something not previously reported in the literature. Clinician awareness of these data and accurate diagnoses delineating affective disorders from ‘problem behaviours’ are called for. Kathryn Mitchell and Stephen Moore from Belfast and Western Trusts expand on the issues in this month's Mental Elf blog at https://elfi.sh/bjp-me13.

‘Connectedness’ is a debated theme in a digital world of perhaps excessive social media, but we recognise the utility of social capital in supporting recovery from periods of mental ill-health. Sweet et al mapped out (pp. 308–317) what they label ‘personal well-being networks’ (PWNs) in 150 individuals with serious mental illness, linking social ties, places, and activities, and examining how they alter well-being. Three ‘types’ were described – formal and sparse, family and stable, and diverse and active – with well-being and social capital differing between them. PWNs are argued to help individualise and contextualise people's lives and resources.

All those moments will be lost in time, like tears in rain

Copy number variants (CNVs) are segments of DNA present in increased or reduced numbers (known as duplications or deletions, respectively), which are risk factors for neurodevelopmental disorders. However, despite commonly having transdiagnostic effects, they have typically been studied in relation to a single condition, and their nature in adults with intellectual disability and psychiatric co-morbidities has been less understood. Thygsen et al (pp. 287–294) determined high pathogenic CNV yields – occurring in about 13% of almost 600 studied individuals from three European intellectual disability sites. Phenotypic presentation was highly variable, confirming a broad role for the relevant genes, and the challenges of potential CNV screening in idiopathic ID are discussed.

Autoimmune encephalitis is postulated to underlie some psychotic illnesses. Accurate and early detection would open up the possibility of treatment with immunosuppressants, but the condition is not always clinically evident, and serum investigations appear to be less reliable than those utilising cerebrospinal fluid (CSF). Oviedo-Salcedo et al tested the utility of CSF analysis for different neuronal autoantibodies (pp. 318–320) in 124 individuals with psychotic illnesses. No positive CSF titres were identified, although low-level serum autoantibody titres were found in three participants. The authors note the lack of neurological symptoms in their cohort and counsel caution in interpretation of their findings, stating that they do not invalidate the need for targeted CSF sampling.

CNVs and CSF: all small-print stuff or an exciting part of neuroscience futures? Barnaby Nelson and colleagues (pp. 262–264) give us their thoughts on psychiatry research trends, including prediction research and ‘big N’ data.

Finally, Kaleidoscope (pp. 326–327) discusses schizotypy and psychopathy, nudging and bullshit, with some words on conspiracy thinking. And – finally finally – if you are, or know of, an enthusiastic psychiatry trainee who would like to blog for us, let us know: https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/information/write-a-blog-for-the-bjpsych.