And in the end, the love you take, is equal to the love you make

Three very challenging pieces in this month's BJPsych explore diverse and taxing issues from early-life trauma risking later-life child abuse potential, through the mental health of Islamic State child soldiers, to assessing capacity in those requesting assisted suicide. Borderline personality disorder and depression are associated with early-life maltreatment, although of course we are reminded how this is neither necessary nor sufficient for the development of either. Dittrich et al (pp. 412–418) take on the formidable task of trying to disentangle how these might modify the potential of mothers subsequently abusing their own children. They find what others have more anecdotally proposed: early-life maltreatment, borderline personality disorder and major depressive disorders are associated with intergenerational abuse potential, with emotional regulation a key modulator. The dangers of the findings being misinterpreted and the fact that the large majority with these problems will not commit such acts are clear, but the authors say we have a target – emotional regulation – that programmes should try to redress in relevant groups.

Kizilhan & Noll-Hussong (pp. 425–429) take on a very different form of early-life maltreatment – that experienced by child soldiers in the Islamic State. In a moving and highly thought-provoking piece, they report on interviews with 81 Yazidi children – aged 8 to 14 – who had fought for Islamic State for at least 6 months in northern Iraq. Compared with local peer comparators who also lived through war, they displayed very high rates of a range of mental health problems. It is a cohort liable to attract opprobrium and dismissal from some quarters, but these children are victims; the paper reminds us how today an estimated 300 000 children are fighting in wars across over 50 countries. Look out for our forthcoming podcast with the authors.

Many of us followed the recent case of 104-year-old botanist and ecologist David Goodall who travelled from his home in Australia to end his life in a euthanasia clinic in Switzerland. It was hard not to be moved by his words ‘I'm not happy. I want to die. It's not sad particularly, what is sad is if one is prevented from dying’. David Shaw and colleagues (pp. 393–395) discuss the assessment of decision-making capacity of those who are requesting assisted suicide. Notably, in some countries, such as Switzerland and the Netherlands, assisted suicide is independent of life expectancy, and thus not reserved for those with terminal illness; the presence of a chronic and refractory mental illness is potentially sufficient. However, studies have shown that a majority of psychiatrists feel that having a major mental illness of itself precludes such capacity. Conversely the authors of this piece argue that we, as a profession, have been too paternalistic, and it is wrong to assume that those with mental health issues cannot consent to end their lives. Three papers where it is easy to take a simplistic or polar viewpoint, but harder to work through with the necessary nuance they demand.

Now they know how many holes it takes to fill the Albert Hall

Low numbers to long follow-ups. Marwick et al systematically reviewed (pp. 398–403) n-of-1 trials in schizophrenia to see what we can learn. An n-of-1 trial is a prospective crossover of an individual sequentially placed on different treatments – it is that thing you do in clinic or ward rounds every day, although more systematically measured and masked to reduce biases. The identified literature was sparse, with only six studies meeting their inclusion criteria, and those were fairly variable in rigour. The authors are nevertheless optimistic that this underutilised methodology is a good fit for evaluating chronic and heterogeneous conditions such as psychoses. We are moving, if not quite to a post-randomised controlled trial world, then to one where increasingly different study methodologies’ strengths are valued in different circumstances. Last month's Kaleidoscope talked through the pros and cons of enormous national registry studies, and there is increasing focus (and hyperbole) about noisy ‘big data’ electronic records. I would recommend a debate on current evidence and utility of these varying non-randomised controlled trial methodologies for trainees’ teaching sessions – do you agree with Marwick that n-of-1 represent ‘the future of evidence-based care’ and ‘the epitome of patient-centred care’?

Something a bit more tried and trusted – a good old fashioned longitudinal birth cohort study. Archer and colleagues report (pp. 419–424) on a 53-year follow-up of adolescent affective symptoms and subsequent mortality in a UK population of almost 4000. It is interesting that although such symptoms in youth have established negative outcomes such as educational attainment and adult mental ill health, data on long-term morbidity are almost non-existent. Taking account of the confounders of parental social class, childhood cognition and illness and youth externalising behaviour, severe – but not mild or moderate – adolescent affective symptoms were associated with a significant increase in premature mortality up to half a century later. The findings reiterate the key need for early treatment.

I'd like to be under the sea, in an octopus's garden in the shade

Sounds kinda nice, and most of us value green (and blue) space. Indeed, it has been a painful loss to see some of the old psychiatric hospitals sell off their grounds to stay financially afloat, losing an oft-undervalued therapy environment at the same time. Or I am just romanticising, as I am wont to do? Well Stigsdotter et al tested (pp. 404–411) nature-based therapy (NBT), comparing it with a specialised cognitive–behavioural therapy programme for those with stress-based problems. NBT involves therapeutic conversations, awareness exercises such as mindful garden walking, gardening, reflection and relaxation time and a homework component. There were no differences between the interventions, with NBT as effective at improving psychological well-being and reducing burnout. With pun intended, more work in the field is needed, but something pleasing to reflect upon as we hit July. Danielle Rhydderch and Ian Collings from the ABMU Rehabilitation and Recovery service in Wales give more detail in our latest Mental Elf blog at: https://elfi.sh/bjp-me14.

From therapeutic gardens to hard-core neuroimaging and genetics; because that is just how the BJPsych rolls. Functional brain changes have previously been reported in obsessive–compulsive disorder (OCD), but any linking of these to genetic determinants had not been well explored. Hibar et al (pp. 430–436) report on the first such genome-wide investigation and found significant association between risk genes and changes in nucleus accumbens and putamen volumes, something they argue is consistent with contemporary models of OCD cortico-striato-thalamic circuitry. Via et al (pp. 437–443) explore the association between worry and ventromedial prefrontal cortical activity – a brain region known to be linked to processing learned safety and threat assessment. Compared with healthy controls, those with generalised anxiety disorder showed reduced activity in this region in response to safety signals, something the authors suggest may characterise generalised anxiety disorder.

Finally, talking through the conundrum of why antidepressants evoke feelings other medications never do, Kaleidoscope (pp. 445–446) teaches us our phrase of the month, and one I think cannot be repeated often enough: pharmacological Calvinism.