Separation anxiety disorder in adults: Is the neglect real?

When a certain psychiatric diagnosis is not often used, what does it mean? It may indicate that the condition in question is not particularly common, at least in the clinical settings. But it may also be interpreted as suggesting that the disorder is poorly recognised, especially if some data indicate that the condition is common in the general population. The assumption of poor recognition may then lead to a notion that diagnostic criteria should be changed to allow better detection. This approach to some disorders has been adopted by the architects of DSM-5, most notably generalized anxiety disorder (Starcevic et al., 2012). Within this framework, the diagnostic concept is not questioned and there is little or no consideration as to how it can be improved. The proposed solution is to lower the diagnostic threshold with the aim of improving recognition. If it then becomes easier to make the diagnosis, its prevalence may increase, albeit at the expense of possibly including many false positive cases. It now appears that this ‘formula’ will also be applied to adult separation anxiety disorder (SAD) in DSM-5.

Adult separation anxiety disorder and DSM-5

Adult SAD is not a new diagnosis and the official name in DSM-5 will still be ‘separation anxiety disorder’. The diagnostic criteria will be very similar to the DSM-IV criteria, but they will pertain to a wider range of behaviours to also encompass those that are more likely to be seen in adults. The total number of criteria and the minimum number of criteria required for the diagnosis will not change. The main change will be an exclusion of the DSM-IV criterion about the onset of SAD before the age of 18. Therefore, DSM-5 will make possible a diagnosis of SAD in an adult without a history of SAD in childhood or adolescence. This will presumably allow more adults with SAD to be recognised and diagnosed adequately.

Has adult separation anxiety disorder been neglected?

Adult SAD appears to be relatively rarely diagnosed in routine clinical practice. For example, a study of the first 1000 patients of an anxiety clinic in suburban Sydney where there is a high awareness of adult SAD reported no person with this diagnosis (Wagner et al., 2005).

Only when adult SAD was specifically looked for and the DSM-IV criterion about the juvenile onset was ignored, its rates were reported to be quite high. Thus, its lifetime prevalence rate in the general United States population was 6.6% (Shear et al., 2006), the frequency in one Italian clinical sample was 42.4% (Pini et al., 2010) and adult SAD comprised 23% of all diagnoses made in (the same?) suburban Sydney anxiety clinic (Silove et al., 2010). These figures highlight inconsistencies in the way in which the diagnosis of adult SAD is made. In one study, this diagnosis was based on an ‘unvalidated’ instrument administered by lay interviewers (Shear et al., 2006), while in the other it was based on a self-report questionnaire (Silove et al., 2010). Although the third study relied on a validated semistructured interview specifically developed to assess symptoms of SAD in children and adults, all its participants had major depression, bipolar disorder or one of the other anxiety disorders as a ‘principal diagnosis’ (Pini et al., 2010). It is difficult to ascertain the impact of adult SAD when it is an additional or secondary diagnosis (Pini et al., 2010) or whose status in this regard is uncertain (Silove et al., 2010), despite findings suggesting that it may be associated with substantial impairment even when the presence of co-occurring conditions is controlled for (Pini et al., 2010; Shear et al., 2006).

In summary, the prevalence of adult SAD is unclear and requires further study. If it is indeed a common condition, claims about its poor recognition and neglect would be justified; if it is not, adult SAD should not be considered a neglected disorder but rather as a condition that may be relatively unimportant. Indeed, the latter is the impression one has when opening almost any textbook of anxiety disorders (e.g., Simpson et al., 2010; Stein et al., 2010) and finding that there are no chapters or sections devoted to adult SAD.

Should pathological separation anxiety be the key feature of any adult disorder?

Separation anxiety is considered one of the basic fears in humans, and it is readily observable in children. With development and maturation, separation anxiety certainly does not go away, but prominent separation anxiety in adults is less ‘visible’ as a problem on its own and often occurs along with other psychopathology or as a part of it. Thus, manifestations of pathological separation anxiety such as uncontrollable worries about losing important attachment figures, intense fears of leaving home or going out unaccompanied and nightmares around the ‘theme’ of separation often characterise other anxiety disorders (e.g., generalized anxiety disorder, panic disorder, agoraphobia and posttraumatic stress disorder). If the symptoms of prominent separation anxiety are better understood in the context of another disorder or interpreted as a manifestation of such a condition, there is no reason to use a diagnosis of adult SAD.

In the vast majority of cases, adult SAD co-occurs with various other anxiety and mood disorders (Pini et al., 2010; Shear et al., 2006), but its diagnostic primacy would be justified only if its symptoms were more disabling or if they were the main reason for seeking professional help. In diagnostically ambiguous situations, why would clinicians diagnose adult SAD in addition to another disorder, also given the lack of well-established treatments for adult SAD? Would it make any difference to the outcome or prognosis, for example, in terms of suggesting a poorer response to treatment? Some evidence suggests that this may be the case (e.g., Aaronson et al., 2008), but it requires replication and development and testing of the specific treatment procedures addressing pathological separation anxiety.

More fundamentally, a question to be posed is whether a phenomenon that is as basic and widespread as separation anxiety, with fuzzy boundaries with its pathological forms, should be conceptualised as the key feature of any disorder. An analogy could be made with death anxiety, another basic and widespread fear. Pathological forms of death anxiety have not led to a proposal to introduce a diagnosis of ‘death anxiety disorder’, as manifestations of pathological death anxiety can be found across a range of adult psychopathology.

What is there to gain from the ‘promotion’ of an adult form of separation anxiety disorder?

With the proposed DSM-5 diagnostic criteria for SAD, the diagnostic threshold will be lowered and it will be easier to diagnose SAD in adults. However, this does not necessarily mean that adult SAD will be more readily recognised. If adult SAD is not conceptually tightened, its clinical utility will not improve, clinicians will not perceive it as a particularly useful diagnosis, and it will not be taken seriously and will probably continue to be ignored.

Finally, a point should be made about a relentless process of splitting diagnostic categories in psychiatry and within anxiety disorders in particular. The DSM-5-sponsored ‘promotion’ of adult SAD as a separate condition seems to ignore the common vulnerability of many anxiety and related disorders and other similarities between them. This is only likely to increase the overlap between the putatively distinct anxiety disorder categories, make boundaries between them less clear and further raise co-occurrence rates. It is hard to see how this will constitute scientific progress and advance our understanding of the psychopathology.

See Viewpoint by Marnane and Silove, 2013, 47(1): 12–15.