September BJPsych Highlights

A Room of One's Own

Most psychiatrists will be familiar with the fact that eating disorders have the highest mortality rate of any psychiatric condition, though I'm not sure that the profundity of that statement really registers as deeply as it should. Most data are based solely on anorexia nervosa and on limited population samples. Iwajomo et al (pp. 487–493) explored this more broadly in a retrospective Canadian cohort study of almost 20 000 individuals (90% women) who had ever received in-patient care for an eating disorder. All-cause mortality was five times higher than that of the general population, and we are reminded that death often comes at so young an age; the number of potential years of life lost was six times that of the wider community. Anna Paspala writes more in this month's Mental Elf blog at https://elfi.sh/bjp-me31.

Anorexia nervosa can profoundly affect many physiological systems, and the large majority of individuals with anorexia nervosa will have at least some cardiovascular complications. Smythe et al (pp. 477–486) systematically reviewed cardiac abnormalities that were identified through echocardiography. Twenty-three eligible studies covered almost 1000 individuals, with a mean age of 17 years and body mass index of 15.2 kg/m². Significantly reduced left ventricular mass and cardiac output, increased E/A ratio (velocities of early and late diastolic filling) and incidence of pericardial effusions were found. Positively, improvements in these were seen as weight was restored. The authors argue that we need to use such findings to help stratify severity and management needs.

Orlando

Most psychiatrists will be familiar with the challenge of when to delineate ‘pathological’ from ‘normal’ grief. Deep loss can be very profound and cover a range of presentations. The issue is not helped by the fact that there are five diagnostic sets for pathological grief. Each of these will inevitably produce different prevalence data, which can vary quite widely from 10% to 30% depending on the evaluated cohort. Problematically, researchers do not always report on the diagnostic tools used, making cross-comparison of research in the field highly challenging. In an interesting analysis piece, Lenferink et al (pp. 473–476) call for harmonisation to help move the field forward. It is helpful to be able to make a diagnosis in the right circumstances: those who suffer pathological grief can benefit from an appropriate grief-focused intervention. Consensus on criteria would be a helpful start.

Diagnostic systems have an even worse historical track record when it comes to issues of sexual and gender identity and ‘pathologising’ people. An editorial by Perlson et al (pp. 471–472) tackles the challenges specifically for transgender and gender-diverse (TGD) people. We are reminded of the stigma TGD individuals can face, including gender identity conversion efforts of others, and how this can affect individuals’ mental health. The authors welcome the DSM-5's change from ‘gender identity disorder’ to ‘gender dysphoria’ as an affirming one but say that complete depathologisation and uncoupling of gender diversity from diagnostic systems is a necessary next step. We need to do better in mental health.

To the Lighthouse

Most psychiatrists will be familiar with the challenge of the perinatal period to the mental health of women. Fewer of us will be as confident with the figures on physical health during this time. Two papers in this month's BJPsych help to rectify this gap. Magnus et al (pp. 501–506) explored the risk of miscarriage in women in a history of psychiatric disorders, utilising a Norwegian registry of over half a million pregnancies. All evaluated mental health conditions came with significantly higher odds ratios of miscarriage, though this varied considerably among conditions: bipolar affective disorder had the highest odds ratio at 1.35 and intellectual disabilities the lowest at 1.07. Perhaps surprisingly, given the direct impact on physiology, the odds ratio for eating disorders was near the lowest at 1.14, and the relative risk in psychoses was similarly ‘low’. Easter et al (pp. 494–500) report on a historical cohort study to identify obstetric ‘near misses’. The rate was 884/100 000 maternities in women who had a history of secondary mental healthcare preceding their delivery, compared with 575/100 000 in those who had not. The greatest risks were for acute renal failure, cardiac arrest, failure or infarction, and obstetric embolism. These papers remind us that perinatal risks in individuals with mental illnesses are not only psychiatric in nature.

Low birth weight is associated with a range of poorer later-life health and social outcomes. The ‘developmental origins of health and disease’ hypothesis posits that early life adversity may have long-lasting effects, but it has been difficult to determine a causal relationship, with many confounders such as environmental and socioeconomic factors. Orri et al (pp. 507–514) tackle this, using 48 independent single-nucleotide polymorphisms measured in a genome-wide association study of just over a quarter of a million people. Their analysis supported a role for low birth weight causally contributing to the risk of attention-deficit hyperactivity disorder, post-traumatic stress disorder and suicide attempts. It was also found to contribute to a range of adverse socioeconomic outcomes including poorer educational attainment and social deprivation. Hope et al (pp. 515–522) ask what the healthcare impact of maternal mental illness (MMI) is on children and adolescents. Using a large primary care register of almost half a million children, the 23% of these who were exposed to MMI averaged 2.21 more healthcare contacts per year. This occurred across a range of settings, including out-patients and in-patients, the emergency department and primary care. The estimated additional cost to the National Health Service was £656 million per annum. Of course, none of this should be felt to ‘blame’ the mothers but to highlight the additional needs of both them and their children. Indeed, it suggests we might better target interventions earlier in this more vulnerable group.