Big data and clinical epidemiology

The discernment of early indicators of illness feature in this month's issue. Specifically, self-harm in youth predicts later mental illnesses such as depression, anxiety and anorexia (Wilkinson et al, pp. 222–226), and school achievement appears to fall off in secondary school for those coping with self-harm (Rahman et al, pp. 215–221). A diagnosis of attention-deficit hyperactivity disorder is a predictor of later suicide attempts: Chen et al (pp. 234–238) offer helpful findings that methylphenidate and atomoxetine do not increase suicide risk; indeed, use of the former actually reduces suicide attempts. Early consumption of cannabis is strongly linked to the development of future psychosis, even allowing for family history, levels of prodromal symptoms and substance use (Mustonen et al, pp. 227–233; Collizi & Murray, pp. 195–196). These papers indicate when and where we should intervene, and show that there are important health benefits. It is less clear how we should prevent emotional and behavioural experiences following terrorist attacks conceived as political violence and conflict.Reference Dom, Schouler-Ocak, Bhui, Demunter, Kuey and Raballo⁵ The study by Vandentorren et al (pp. 207–214) following the Paris terrorist attacks shows that 11%, 18% and 31% of civilians developed symptoms of depression, post-traumatic stress disorder and anxiety, respectively, after 6 months; symptom levels were much lower among rescue workers, although longer-term follow-up may show greater health problems.Reference Smith, Cross, Poland, Clay, Brookes and Maidment^6–Reference Jacobson, Norman, Nguyen and Brackbill⁸

Language matters

Returning to the historical archives of the journal, readers are likely to be entertained by some (and horrified by others) of the preoccupations and language used to describe people with mental illnesses; the records show horrendous social and environmental conditions in which patients survived. Scientific advances have permitted a more fine-grained classification, however imperfect, of different types of illness–symptom–context–risk constellations. What might the plight have been of people with dementia, a condition now known to affect 50 million people globally at a cost of $1 trillion?Reference Pickett, Bird, Ballard, Banerjee, Brayne and Cowan⁹ More people are being diagnosed with dementia in the UK, and care standards are improving.Reference Donegan, Fox, Black, Livingston, Banerjee and Burns¹⁰ More research is now needed, not only on biological–cellular and lifestyle-focused prevention, but also on care processes and support for patients, with implications for all health professionals and teams in primary and secondary care.Reference Smith, Cross, Poland, Clay, Brookes and Maidment^6, Reference Ballard, Corbett, Orrell, Williams, Moniz-Cook and Romeo¹¹ Doctors appear to struggle with the delivery of a dementia diagnosis in memory clinics and vary from being sensitive-suggestive to abrupt, the latter surprisingly being more prominent with those with more significant impairments (Dooley et al, pp. 239–245). Prognosis was rarely discussed.

Pitman (pp. 197–198) explores another area concerning communication and explains how health professionals show insufficient awareness and care for the families of people taking their own lives. There are important preventive opportunities for advising and supporting those bereaved to cope with loss and to protect their mental health, and thus to also prevent further tragedies. How can it be that services for the bereaved are somehow left to evolve in the voluntary or charitable sectors? There is a difficult balance to strike: to be optimistic, promote recovery and help process emotional distress, trauma and loss, while also providing more intensive and compassionate care where it is needed, and not overlooking powerful preventive opportunities among the bereaved.,Reference Sveen and Walby^12, Reference Linde, Treml, Steinig, Nagl and Kersting¹³

Advances in scientific methods

Research methods are now much more powerful, but they must stay relevant to health service contexts and implementation challenges, including escalating costs of healthcare. Nobis et al (pp. 199–206) provide hope for people with depression and diabetes, giving evidence that a web-based intervention is cost-effective and affordable. An alternative conceptualisation of early intervention is to move back in time not only to youth, childhood, or neo- or antenatal periods, but to the early stages of cell differentiation and stem cell research. For example, Liu & Howard (pp. 193–194) outline how such approaches might target hippocampal neurogenesis to reduce the chances of, and impairments related to, neuropsychiatric illnesses, cognitive impairment and poor memory. The archives and the historical legacy of the Royal College of Psychiatry publications and commitment to science are substantial, yet we have much to remember and learn in order to sustain prevention and treatment efforts through the remarkable amount of knowledge and a number of research tools are emerging. What might we see in healthcare, and psychiatric practice, after the next 150 years? I would welcome some scientifically grounded speculation and foresight from those developing new technologies and approaches that we might scale up and use to transform the experience and outcome of mental illnesses.