Separation anxiety disorder: Lumping or splitting?

To the Editor

Starcevic (2013) and Marnane and Silove (2013) present two perspectives on separation anxiety and the value of categorising it as a discreet disorder. Like so many of the disputes in our field, the issue is less about facts than about interpretation and purpose. Clearly there are adults who show separation anxiety. Are those individuals better served by having their distress and dysfunction labelled in a specific way? Is there any cost to them, the community, the profession, the scientific world by adopting such an approach? How do we evaluate those pros and cons?

Perhaps the greatest loss that accompanied the introduction of the Diagnostic and Statistical Manual of Mental Disorders (DSM) was the acknowledgement that much of our work involves maladaptive coping, either pervasively or in specific situations: what used to be called neurosis. While there are clearly differences between these presentations, they have much in common, as do the interventions that help. By contrast, a diagnostic approach that emphasises difference would potentially generate an infinitely expanding range of specific styles of maladaptation to new and interesting triggers.

It used to be accepted wisdom that most personality disorders (PDs) had their roots in adverse childhood experiences of various sorts. That, combined with an individual’s genetic biases, fostered a learned style of coping in a specific context that proved maladaptive in the wider society. Regrettably, the PD label acquired pejorative overtones, like neurosis before it. When I came to child psychiatry after 20 years with adults, I found my new colleagues very resistant to its use in teenagers, although the field is now moving steadily in that direction. Approached from the perspective of utilitarianism, it is a good label if it assists the youngsters concerned, by enabling us to see them as stuck in a predicament with a particular style of coping to survive there. This seems more sensible than a depression illness label, but only if it outweighs the negatives the PD label generates. They are young and flexible and perhaps we can steer them out of it. It is hardly a question of science.

There is a similar tension with another DSM fifth edition (DSM-5) creation, temper dysregulation disorder with dysthymia, a questionable entity whose main raison d’etre seems to be to avoid the over-diagnosis of paediatric bipolar: again, hardly science. Does complex post-traumatic stress disorder better serve the patients and the research community than borderline PD?

Reactive attachment disorder (RAD) presents similar dilemmas and is set for serious re-examination in DSM-5. We see lots of attachment-related psychopathology in child and adolescent practice, but existing labels address the symptoms not the postulated cause. We do not see much pure RAD, so it is intriguing to see that label now appearing in adults that I would have categorised as having personality problems. Where is the logic?

We struggle without a unifying model. Some things seem worth splitting off, for the rest I remain a lumper.