Adjustment disorder entered the DSM-II nomenclature in 1968 and was recognized in ICD-9 in 1978. Before then the term ‘transient situational disturbance’ was applied to such conditions. The addition of adjustment disorder to the ICD classification was in response to the confusion generated by the older concepts of reactive and endogenous depression. Both DSM-IV (American Psychiatric Association, 1994) and ICD-10 (World Health Organization, 1992) retain the category of adjustment disorder, which has utility as a clinical concept. However, it has been eclipsed by the focus on mood disorder among research and policy-makers. A consequence of this is the danger of exaggerating the need for expensive and sometimes unpredictable mental health interventions in those whose problems are likely to resolve spontaneously.

LOOSE DEFINITION

The psychiatric disorders classified in ICD-10 and DSM-IV as being caused by stress are acute stress reactions, post-traumatic stress disorder (PTSD), adjustment disorder and bereavement (in DSM-IV only). Acute stress reactions and PTSD develop in response to exceptionally threatening experiences but the former subsides within days and the latter is more protracted. Adjustment disorders are conceived of as developing in response to a variety of causal stressful events, the symptoms representing an adaptation to these stressors or to their continuing effects. The ICD-10 states that the diagnosis depends on a careful evaluation of the relationship between: form, content and severity of symptoms; previous history and personality; and stressful event, situation or life crisis. The latter should be established clearly before the diagnosis can be made and there should be strong presumptive evidence that the disorder would not have arisen without it. The definition in DSM-IV is similar.

Two border disputes attach to the diagnosis of adjustment disorder. One is the indistinct separation between the varied manifestations of adjustment disorder and normal adaptive reactions. The ICD-10 points to “usually interfering with social functioning and performance” and “some degree of disability in the performance of daily routines”, whereas DSM-IV suggests “marked distress that is in excess of what would be expected given the nature of the stressor or by significant impairment in social or occupational (academic) functioning”.

The second border dispute is the problem of overlap with other disorders. Both ICD-10 and DSM-IV attempt to overcome this problem by specifying that if criteria for another disorder are met, then the diagnosis of adjustment disorder should not be made, relegating it to subordinate status. In practice, the diagnosis of adjustment disorder will be undermined by the mechanistic and rigid application of diagnostic criteria (e.g. in DSM-IV, major depressive episode is diagnosed when five or more depressive symptoms have been present for longer than 2 weeks, irrespective of the close temporal relationship between an identifiable stressor and symptoms).

THE VALIDITY DEBATE

Adjustment disorder has always been controversial and has been described as a “waste-basket diagnosis, used in such a vague and all encompassing manner as to be useless” (Reference Fard, Hudgens and WelnerFardet al, 1979), as a “cryptic form of disease entity” and as unstable (Reference Greenberg, Rosenfeld and OrtegaGreenberget al, 1995).

However, content validity studies show that those with adjustment disorder differ from those with no diagnosis and those with mood disorders on a number of parameters, including differences in the nature of the stressors in comparison with other mood disorders (Reference Snyder, Strain and WolfSnyder et al, 1990; Reference Despland, Monod and FerreroDespland et al, 1995). Differences in vulnerability also emerged, both in relation to social isolation (Reference Snyder, Strain and WolfSnyder et al, 1990) and to personality disorder (Reference Spalletta, Troisi and SaraccoSpalletta et al, 1996), with personality disorder being much lower in those with adjustment disorder than in those with major depression or dysthymia. Evidence for its construct validity has come from out-patient studies comparing adjustment disorder with major depression and dysthymia on quality of life measures.

The predictive validity of the diagnosis of adjustment disorder has been confirmed also among adult in-patients (Reference Andreasen and HoenkAndreasen & Hoenk, 1982), with 79% of adults being well 5 years after the index admission. In those with adjustment disorder most symptoms resolve rapidly (Reference Snyder, Strain and WolfSnyder et al, 1990; Reference Despland, Monod and FerreroDespland et al, 1995), with fewer than 17% developing a chronic course (Reference BronischBronisch, 1991; Reference Greenberg, Rosenfeld and OrtegaGreenberg et al, 1995).

EPIDEMIOLOGICAL ERROR

The DSM-IV states that adjustment disorder is the principal diagnosis in 5-20% of individuals in out-patient treatment. However, others have found the diagnosis much less common in this setting, estimated at 2.3% of such patients (Reference Fabrega, Mezzich and MezzichFabrega et al, 1986). A much higher prevalence has been shown in medical settings and in primary care. A study of over 1000 consultationliaison patients found that this diagnosis was made in 12% of patients, being comorbid with personality disorder or organic mental disorder in 4.2% and the sole diagnosis in 7.8% of patients (Reference Strain, Smith and HammerStrain et al, 1998). Snyder et al (Reference Snyder, Strain and Wolf1990) made the diagnosis of adjustment disorder with greater frequency than that of major depression among a group of patients seen in a general hospital, whereas in primary care adjustment disorder has been found to be the most common diagnosis.

ACADEMIC NEGLECT

Over the past 25 years there have been fewer than 30 publications in peer-reviewed journals exclusively devoted to the diagnosis, and adjustment disorder is dealt with briefly in all major British textbooks, even the most recent publication (Reference Stein and WilkinsonStein & Wilkinson, 1998). This lack of interest is surprising in view of the prevalence of the disorder and especially when contrasted with the vast concurrent interest in related topics such as PTSD and life-events research into the precipitation of a range of mental disorders over the past two decades (Reference Kendler, Karkowski and PrescottKendler et al, 1999).

To date, epidemiological studies have shed little light on adjustment disorder. The Epidemiologic Catchment Area (ECA) study (Reference Myers, Weissman and DischlerMyers et al, 1984) did not include any assessment of the prevalence of adjustment disorder; neither did the US National Comorbidity Survey (NCS; Reference Kessler, McGonagle and ZhaoKessler et al, 1994) nor the National Psychiatric Morbidity surveys of Great Britain (Reference Jenkins, Lewis and BebbingtonJenkins et al, 1997). Our recent multi-set European study (the Outcome of Depression International Network, ODIN; Reference Dowrick, Casey and DalgardDowrick et al, 1998) included adjustment disorder within its scope. However, we identified adjustment disorder in only 1% of individuals in the community diagnosed as having a depressive disorder (Reference Ayuso-Mateos, Vázquez-Barquero and DowrickAyuso-Mateoset al, 2001). On the face of it, this finding might encourage the critics of adjustment disorder.

WHAT IS DEPRESSION?

We believe that the marginalisation of adjustment disorder stems from an overelastic concept of ‘depression’ and from the over-rigid diagnostic guidelines and criteria in ICD-10 and DSM-IV. Depression is an over-inclusive term with a lack of conceptual clarity between symptom, syndrome, episode and illness. It encompasses a range of feelings, described by Snaith (Reference Snaith1987) in his critique of ‘mild depression’ — “Here are the states of grief at loss, frustration of failed aspiration, the gloom of despair, the accidie of disillusion, the demoralisation of the long suffering and the cynical outlook of the pessimist”. Here also is mood disorder that resolves spontaneously and mood disorder requiring specific treatment.

As psychiatry focuses more and more on ‘serious mental illness’ and increasingly allies itself with the biological sciences, in parallel with the discovery of effective antidepressants and other medications, psychiatrists understandably tend to view ‘depression’ as a biological entity. Even those who acknowledge that greater emphasis needs to be placed on psychological interventions do so in the belief that treatment is necessary in the first place. Thus, transient depressive responses to stressful events are increasingly regarded as illness requiring specific interventions.

The modern approach bases diagnosis on symptom thresholds. In DSM-IV, adjustment disorder represents a hybrid based on aetiology and rooted out when the severity and duration threshold for another disorder are reached. A person who is tearful, not sleeping, has poor concentration, has reduced appetite and feels tired for longer than 2 weeks following a diagnosis of cancer has, in DSM-IV terms, a major depressive episode. We believe that this is a low threshold for a specific diagnosis signalling a particular therapeutic approach and prognosis. All the more so because patients so diagnosed remain the foundation of antidepressant clinical trials. It is absurd to propose that a person adapting to a stressor may have only four rather than five of nine symptoms, or such symptoms for a duration of 2 weeks, after which the diagnosis becomes major depressive episode.

The absence of clinical nuance has been alluded to by Clarke & McKenzie (Reference Clarke and McKenzie1994) in their critique of cut-offs and thresholds when they state “The subtle balancing of consideration of quantity and quality of symptoms has occurred intuitively in clinical practice but is at risk of being lost as we ‘operationalise’ criteria”. We believe that psychiatrists need to embrace the understanding that the DSM-IV diagnostic criteria and ICD-10 guidelines are neither fully comprehensive nor intended to be applied in a cookbook fashion.

BLUNT INSTRUMENTS

The use of diagnostic instruments in research to identify the disorders classified in both DSM and ICD is an advance in epidemiological research. However, the technical advantages need to be set against the concerns about their clinical validity (Reference Brugha, Bebbington and JenkinsBrugha et al, 1999) because the diagnostic instruments inevitably incorporate the shortcomings of the classifications from which they derive, also vitiating cross-study comparisons. The particular relevance of this difficulty to adjustment disorder and mood disorders has been highlighted by Regier et al (Reference Regier, Kaelber and Rae1998), who pointed to the different prevalence rates for major depression found across the first-wave and second-wave ECA studies (Reference Myers, Weissman and DischlerMyers et al, 1984; Reference Regier, Narrow and RaeRegier et al, 1993) and the NCS (Reference Kessler, McGonagle and ZhaoKessler et al, 1994), ranging from 4.2% to 10.1%. He commented: “It is possible that many people with currently defined mental syndromes (in particular among the affective and anxiety disorders) not brought to clinical attention may be having appropriate homeostatic responses that are neither pathological nor in need of treatment”.

Wide variation exists in the approach of various diagnostic instruments to the measurement of adjustment disorder. Two widely used schedules have not encompassed the diagnosis: the Clinical Interview Schedule — Revised (CIS-R; Reference Lewis, Pelosi and ArayaLewis et al, 1992), used in the British National Psychiatric Morbidity surveys, and the Composite International Diagnostic Interview (CIDI; Reference Robins, Wing and WittchenRobins et al, 1988), used in the US National Comorbidity Survey. The Schedule for Clinical Assessment in Neuropsychiatry (SCAN; World Health Organization, 1994) includes adjustment disorders in section 13, which deals with inferences and attributions, but it provides no guidance on the application of this section. In addition, by placing this section at the end of the schedule there is an inference that adjustment disorder is an unlikely diagnostic consideration. Against this background it is little surprising to find a proliferation of mood disorders and a scarcity of adjustment disorders.

SYNTHESIS

The strength of epidemiological studies derives from the extent of their contribution to knowledge and clinical practice in respect of the identification, prevention and treatment of psychiatric disorders. Wide variations in the prevalence of common psychiatric disorders undermine the foundation of therapeutic interventions. The distinction between disorders requiring treatment and those that resolve spontaneously over time is more than a nosological nicety because it impinges upon clinical practice and policy via resource allocation. Psychiatric disorder in primary care has been divided into three broad categories — distress requiring no specific intervention, distress requiring intervention and major psychiatric disorder. Distress and disorder remain conflated in clinical practice and among the research community as we continue to medicalise the emotional vagaries of the human condition (Reference IllichIllich, 1977). Although some will argue that the labelling of adjustment disorder in current classifications reinforces the medicalisation of distress and will point to an inherent contradiction in our argument that adjustment disorder requires rehabilitation, we believe that its value lies in identifying those not requiring any treatment from those with similar symptoms and dysfunction who require and benefit from specific interventions. Whether or not DSM-V and ICD-11 will recognise these issues of clinical judgement when framing their criteria and guidelines remains to be seen.