Crude classifications and false generalisations

Careful classification is an important focus in this month's BJPsych. Philosophers and anthropologists consider the ability to define and categorise to be a fundamental human cognitive characteristic. It is also a fundamental tool for robust research. And, true to H. G. Wells’ words, crude classification can be the curse of an organised researcher.

The publication of ICD-11 brings new classifications for existing disorders. Adjustment disorder receives new specific diagnostic criteria – a conceptual departure from its place as a diagnosis of exclusion. Using data from three African countries, Levin et al (pp. 557–564) are the first to use a network analytic approach to identify central symptoms of adjustment disorder and see whether these align with the new ICD-11 criteria. Despite cultural variations in samples, they consistently found that ‘preoccupation’ symptoms are of relatively greater importance than ‘failure to adapt’ symptoms – providing a potential therapeutic target for future interventions.

Crude classifications even exist for well-established, essential treatments for mental health disorders. Castelpietra et al (pp. 532–537) state that ‘psychotherapy covers a very broad range of generally ill-defined activities … in the absence of any internationally agreed classification of relevant interventions’. They propose a new potential international classification system for psychotherapy to ensure we are talking about the same treatments in research and funding the right ones in practice too.

In an Analysis article, Alison Heru (pp. 565–568) is more concerned with the false generalisations made about expressed emotion in historic research. Heru argues that high expressed emotion is not a universally negative attribute, with different experiences in different cultural settings. For example, research in African American families found that high expressed emotion/high criticalness was associated with better outcomes for family members with schizophrenia. It would be interesting to see research about expressed emotion reincorporate the concept of ‘family warmth’, which was excluded from early research as an expressed emotion component because it was considered too hard to measure.

Putting good classification into practice

Accurate classification of psychotherapy – as per Castelpietra's proposal – could pave the way for clearer understanding of treatment effects. This is particularly true in the case of systematic reviews and meta-analyses – such as two of this month's research papers – that often pool results from heterogenous studies.

Breedvelt et al (pp. 538–545) use this methodology to explore whether psychological interventions can be viable alternatives or additions to antidepressants to prevent depressive relapse or recurrence. Contrary to current guidelines, they suggest that psychological interventions could be used either while tapering antidepressants or in addition to long-term antidepressants in the prevention of relapse. A research paper by Nord et al (pp. 546–550) may provide answers about the underlying neurobiological mechanisms behind these results. Meta-analytic data show that the neural changes associated with antidepressant and psychotherapy treatment occur in significantly different brain regions, but both have effects on the affect network. In future, clearer classification of psychotherapies might even allow us to delineate neural changes associated with different modalities if they exist.

The digital divide

Two articles shed light on the complex interaction between the digital space and mental health.

Bu et al (pp. 551–556) examine the impact that a number of different lockdown-related daily activities had on the risk of depression and anxiety early in the pandemic. Surprisingly, one of their findings was that communicating with family and friends via video calls, phone or messaging was associated with higher levels of depression. It is possible that the pandemic put additional pressure on already strained relationships, but the authors argue that this finding could also be related to the experience of digital communication. Apparently, communication over the telephone (or other digital means) can increase loneliness – perhaps because such interactions are less emotionally rewarding experiences than face-to-face communication.

Nonetheless, digital skills and resources have been a lifeline for many in the past 18 months – providing access to vital healthcare, social care and communities during the pandemic. Unfortunately, though, the digital space is another area where people with severe mental illness face exclusion. In their powerful editorial, Spanakis et al (pp. 529–531) warn about a lack of digital access for people with severe mental illness. This digital exclusion compounds the multiple and severe health inequalities that people with mental illness already face. It is an important caution to all mental health services that might be considering a more permanent transition to digitally focused delivery models.

Finally, as we head into the winter months and towards the festive period, Kaleidoscope (pp. 573–574) asks how to measure sunshine and value compassion in people.