Adjustment disorders - detailed technical article.
- Definition and history
- Changes in the criteria for adjustment disorder in DSM-IV
- Problems with the adjustment disorder diagnosis
- Associated features of adjustment disorder
- Aetiology—the role of stress Nature of the stressor
- Modifiers of stress
- Confounds: assessing stress
- Clinical features
- Psychotherapy and counselling
- Course and prognosis
- Adults and adolescents
The psychiatric diagnoses that arise between normal behaviour and the major psychiatric morbidities constitute the problematic subthreshold disorders. These subthreshold entities are also juxtaposed between problem-level diagnoses and more clearly defined disorders. They present major taxonomical and diagnostic dilemmas in that they are often poorly defined, overlap with other diagnostic groupings, and have indefinite symptomatology. It is therefore not surprising that issues of reliability and validity prevail. One of the most commonly employed subthreshold diagnosis that has undergone a major evolution since 1952 is adjustment disorder (AD). The advantage of the indefiniteness of these subthreshold disorders is that they permit the classification of early or prodromal states when the clinical picture is vague and indistinct, and yet the morbid state is in excess of that expected in a normal reaction and this morbidity needs to be identified. Therefore AD has an essential place in the psychiatric taxonomy:
- normal state
- problem-level diagnoses (V Codes)
- adjustment disorders
- disorders not otherwise specified (NOS)
- major psychiatric disorders.
Adjustment disorder would ‘trump' problem-level disorders, but would be ‘trumped' by a specific diagnosis even if it were in the NOS category. The aetiological and dynamic attributes of the diagnosis of AD make it an important diagnostic category that bridges normality and pathology.
Many questions prevail with regard to the concept of the adjustment disorder diagnosis: the role of stressors and the place of specific stressors; the importance of age; the role of concurrent medical morbidity, for example comorbidity of Axis I and Axis III morbidity; the specificity of the diagnostic criteria; the unavailability of a list of symptoms; uncertainty as to optimal treatment protocols; undocumented outcomes or prognosis. Research data regarding these questions will be examined.
AD is a stress-related phenomenon in which the stressor precipitates maladaptation and symptoms that are time limited until the stressor is attenuated or a new state of adaptation occurs. At the same time that AD was evolving, other stress-related disorders, for example post-traumatic stress disorder and acute stress disorder, have been described. (Acute stress disorders were formulated by Spiegel during the development of the DSM-IV. (1,2)) Acute stress reactions could result from involvement in a natural disaster such as a flood, or an avalanche, or a cataclysmic personal event, for example loss of a body part. The stress-related disorders are unique in that they are psychiatric diagnoses with a known aetiology—the stressor—and this aetiology is central to the diagnosis. (Three other diagnostic categories also invoke aetiology in their diagnostic criteria: organic mental disorders (aetiology—physical abnormality); substance abuse disorders (aetiology—ingestion of substances); post-traumatic and acute stress disorders (aetiology—an identifiable stressor).) The DSM was conceptually designed with an atheoretical framework to encourage psychiatric diagnoses to be derived on phenomenological grounds with an avowed dismissal of pathogenesis or aetiology as diagnostic imperatives. In frank contradiction of this atheoretical algorithm, the stress-induced disorders require the inclusion of an aetiological significance to a life event—a stressor—and the need to relate the stressor's effect on the patient in clinical terms.
The diagnosis of AD also requires a careful titration of the timing of the stressor in relation to the adverse psychological sequelae that ensued. Maladaptation and disturbance of mood should obtain within 3 months of the patient experiencing the stressor. Until the DSM-IV criteria, the ADs were regarded as transitory diagnoses that should not exceed 6 months in duration. Thereafter, that appellation could not be employed and the diagnosis had to be changed to a major psychiatric disorder or dropped.
In summary, AD is a diagnosis that has been insufficiently researched but is, however, commonly employed in clinical practice. The utilization of the AD diagnosis is attributable to several issues:
- in a diagnostic system that is principally atheoretical, AD remains one of the few conditions which is linked to an aetiologic event;
- the concept that adjustment problems stem from stressful events has been a precept of psychodynamic thinking, and often underlies the approach of psychotherapeutic treatment;
- because the development of transient psychiatric symptoms in the context of stress is virtually a universal experience, AD is considered by many to be a non-stigmatizing diagnosis to assign when making a patient's psychiatric status public.
Definition and history
Wise,(3) who described the history of AD since 1945, states that the AD diagnostic concept initially included the idea of a transient situational disturbance, classified by developmental epochs. It then evolved to embody a disorder of adjustment characterized by maladaptation, for example work (or academic) inhibition, accompanied by a mood state, for example depressed, anxious, mixed (DSM-III).(4) In 1987 with the advent of DSM-IIIR, another category of adjustment disorders was included, i.e. physical complaints. (5)
With the opportunity in 1994 to develop another evolutionary step of the DSM, i.e. DSM-IV, (4) the authors were asked to re-examine the subthreshold diagnostic category of AD with the goal of improving its acknowledged ‘shortcomings.' The research included review of the literature, reanalysis of existing studies of AD and their data sets, and examination of field studies (e.g. minor depression, minor anxiety) to observe if there was sufficient differentiation among these minor disorders from the ADs (e.g. how often was a stressor identified in those patients assigned the diagnosis minor depression or minor anxiety?). From these three sources and consultations, modifications for DSM-IV and their rationale were formulated based on the best evidence extant.
Changes in the criteria for adjustment disorder in DSM-IV
The review of the literature and the reanalysis of the Western Psychiatric Institute and Clinic data supported the following changes in the DSM-IV.
- Enhance the understanding of the language.
- Describe the time of the reaction to reflect duration: acute (less than 6 months) or chronic (6 months or greater).
- Allow for the continuation of the stressor for an indefinite period; psychological reactions to chronic stress states (e.g. chronic arthritis, HIV, abuse by an alcoholic spouse) do not necessarily terminate at 6 months, nor do they necessarily lead to a major psychiatric disorder.
- Eliminate the subtypes of mixed emotional features, work (or academic) inhibition, withdrawal, and physical complaints (as they were rarely employed by diagnosticians).
Although it might be argued that ADs could be placed in a new category of ‘stress response syndromes', the literature and research reports did not support this taxonomical organization. AD could be eliminated altogether, with the advantage of maintaining the atheoretical approach of DSM, and substitute in its place the appropriate minor or NOS categories. However, this solution does not seem beneficial in view of the recent findings that demonstrate AD to be a valid frequently employed diagnosis.(6) AD was diagnosed in over 60 per cent of burned inpatients, (7) over 20 per cent of patients in early stages of multiple sclerosis, (8) and over 40 per cent of poststroke patients.(9) Furthermore, evaluations of patients in a psychiatric walk-in clinic showed a significant difference in the symptom profile of those assigned AD and the others, including minor diagnoses. (10) (The McArthur field trials on the prospective assessment of minor depressive and anxiety disorders and which collected data on the occurrence of stressors immediately preceding the outbreak of symptoms are important databases that need further study to establish whether stress per se is a distinguishing characteristic between AD and the other minor mood disorders.)
Problems with the adjustment disorder diagnosis
The symptom profile
Two issues remain as fundamental confounds in the diagnosis of AD. First, because of the lack of any quantitative behavioural or operational criteria, the problem of reliability and validity are paramount. Criterion reference was evaluated by Aoki et al. (11) who reported that three psychological tests, Zung's Self-Rating Anxiety Scale, (12) Zung's Self-Rating Depression Scale, (13) and the Profile of Mood States, (14) were useful tools for the diagnosis of AD in physical rehabilitation patients. While these measures succeeded in reliably differentiating AD patients from normal patients, they were not able to distinguish them from patients with major depression or post-traumatic stress disorder. The second confounding factor is that the classification of syndromes that do not fulfill the criteria for a major mental illness but present with serious (or incipient) symptomatology that requires intervention and/or treatment may be viewed, by default, as ‘subthreshold' and hence attract a subthreshold interest by health-care workers and third-party payers. Thus the construct of AD is designed as a means for classifying psychiatric conditions having a symptom profile that is (as yet) insufficient to meet the more specifically operationalized criteria for the major syndromes but that is:
- clinically significant and deemed to be in excess of a normal reaction to the stressor in question;
- associated with impaired vocational or interpersonal functioning;
- not solely the result of a psychosocial problem (V Code) requiring medical attention (e.g. non-compliance, phase of life problem, etc.).
Attention to minor mental symptomatology (and psychiatric morbidity) may forestall the evolution to more serious disorders and allow remediation before relationships, work, and functioning have been so impaired that they are disrupted or permanently impaired. Yet in the ‘grey area' where early diagnosis may take place and has enormous value with modest therapeutic investment, guidelines are the most tenuous. It is the professionals at the ‘front door'—those involved in primary care, triage, and emergency room treatment—who must be assisted to make this most difficult call: is there sufficient psychiatric morbidity to warrant mental health assessment and/or intervention? A number of studies state that because of the subthreshold nature of the AD diagnosis and the absence of a symptom checklist, non-psychiatric physicians and nurses have more difficulty making the AD diagnosis than that for a major psychiatric disorder. (7,15,16 and 17) However, the concurrence of medical illness, which is a frequent comorbidity of AD, does not have the same impact on evaluating a symptom profile as occurs in the major disorders with checklists. For example, in major affective disorders, if seminal symptoms which are key to the diagnosis (e.g. those of appetite, sleep, energy, libido) are attributable to a medical diagnosis, they cannot be used to support a psychiatric diagnosis. At times the attribution of the symptoms (physical or psychological) cannot be determined and therefore the quantitative assessment of prescribed symptoms supporting one diagnosis or another is problematic. However, field studies are being performed (18) to assess whether a reliable checklist from an elaborate list of symptoms associated with AD can be constructed. (The V Codes—a problem level of diagnoses—are understandably devoid of a symptom-based diagnostic schema.)
The meaning of ‘maladaptive'
The imprecision of the diagnostic criteria for AD is immediately apparent in the DSM-IV description of this disorder as a maladaptive reaction to an identifiable psychosocial stressor, or stressors, that occurs within 3 months after onset of the stressor. It is assumed that the disturbance will remit soon after the stressor ceases or, if the stressor persists, when a new level of adaptation is achieved. (1) In addition to the problem of no symptom checklist, difficulties are inherent within each of these diagnostic elements.
First, with regard to the ‘maladaptive reaction', it is unclear how this concept can or should be operationalized. The social, vocational, and relationship dysfunctions that are qualitatively or quantitatively unspecified lend themselves neither to reliability, nor to validity, nor even to clinical agreement when maladaptation is present. The concept of a maladaptive reaction is confounded by physical status, perception, and especially by culture, i.e. the reactions expected to occur within a specific cultural environment, gender responses, developmental level differences, and the ‘meaning' of events to an individual. The concepts of ‘average expectable environment' and ‘the patient's explanatory belief model' are examples of assessing cultural and subjective differences that can effect an individual's mental state and reaction. (19) The variable of culture was not included in the decision-making algorithm of DSM-IV. Is the assessment of maladaptation subjective or objective? Who makes the decision—a third party, a mental health professional, the patients themselves, or an admixture of these? Is this decision ‘culture bound?' Succinctly when does an individual cross the threshold into ‘patienthood,' and who will make the decision?
The patient's functional status evaluation (Axis V) is not linked via an algorithm to the AD construct in DSM-IV. Fabrega et al. (20) contend that both subjective symptoms and decrement in social function can be considered ‘maladaptive,' and that the severity of either of these is subject to great individual variation. Utilizing data from Axis V and their newly constructed ‘Axis VI'—an additional and more specific functional status axis on their Initial Evaluation Form (10)—these authors could not conclude that the level of psychopathology correlates with impaired functioning.(The functional status measure, involving seven levels of impairment, is used to assess ‘current functioning' of patients in three dimensions: ‘at work or at school, with family, and with other person or groups'. (20)) However, Bodlund et al. (21) found that the use, according to Axis V, of the Global Assessment of Functioning Scale self-report was a poor predictor for an AD with depressive symptoms as these patients, more than others, tended to score themselves lower.
Obviously, the stressor and its effect are central to the AD diagnosis. The second major confound emanates from the fact that the DSM-IV presents no criteria or ‘guidelines' to quantify stressors or to assess their effect or meaning for a particular individual at a given time. Furthermore, the assessment of stress is not linked by an algorithm to Axis IV—a statement of stress during the previous year—and so internal consistency or reinforcement within the diagnostic lexicon is not mandated (D. Schafer, personal communication, 1990). Many of the confounds expressed regarding the assessment of maladaptation above apply to the evaluation of stressors as well. (22,23 and 24)
Mezzich et al. (10) attempted to classify and quantify the psychosocial stressors in 13 specific domains: health, bereavement, love and marriage, parental, family stressors for children and adolescents, other familial relationships, other relationships outside the family, work, school, financial, legal, housing, and miscellaneous. Such specificity has not been defined in DSM-IV and the construct is vague and generic with minimal opportunity to achieve quantification. Despland et al. (25) observed that the type of stressor may indeed be of help in diagnosing AD. His study demonstrated that AD with depressed mood and mixed mood was associated with more marital problems than major depressive disorders. AD with anxiety could be distinguished from the major anxiety disorders by the quantity of family and marital problems.
The time course
The time course and chronicity of both stressors and their consequent symptoms were left vague in DSM-IIIR and were not consistent with the clinical situation. The modifications introduced in DSM-IV, which differentiate between acute and chronic forms of AD, solved the problem of the 6-month limitation of the AD diagnosis in DSM-IIIR and was more in keeping with what is observed in the clinical situation. This change was validated by Despland et al. (25) who observed that 16 per cent of patients with AD required treatment longer than 1 year—the mean exceeded the prior limitation of 6 months.
Other problems of definition
With all the limitations presented—the diagnosis of AD is not scientifically rigorous—it is just this imprecision that paradoxically makes the diagnosis so empirically useful to psychiatry. The identification of an emerging illness in its early stages, where the diagnosis of AD can serve as a ‘temporary' diagnosis modifiable with evolving information from longitudinal evaluation and treatment, allows the ‘marking' of an individual for possible difficulty, before the psychiatric morbidity becomes more apparent and at times destructive.
Even serious symptomatology (e.g. suicidal behaviour) that is not regarded as part of a major mental disorder needs treatment and a ‘diagnosis' under which it can be placed, for example a V Code, ‘Phase of Life Problem,' AD, acute stress response, etc. De Leo et al. (26,27) reported on AD and suicidality. Recent life events, which would constitute an acute stress, were commonly found to correlate with suicidal behaviour in a patient cohort which included those with AD. (28) Spalletta et al. (29) observed the assessment of suicidal behaviour to be an important tool in the differentiation among major depression, dysthymia, and AD. AD patients were found to be among the most common recipients of a deliberate self-harm diagnosis, with the majority involving self-poisoning. (30) Thus deliberate self-harm is more common in AD patients,(30) while the percentage of suicidal behaviour was found to be higher in AD patients with depressed mood. (29)
The AD DSM-IV Work Group suggested that suicide and deliberate self-harm could be subtypes of AD. However, there were concerns that patients with other diagnoses, for example major affective disorder, borderline personality disorder, etc. and suicide behaviour, would be assigned the AD diagnosis since there was a specific placement for suicidal ideation and behaviour and that would be a predominant reason to use AD. The final decision was to place the problem of suicidal symptomatology without a psychiatric diagnosis in the DSM-IV F Code section for other problems ‘that may be a focus of clinical attention.' A subthreshold diagnosis, AD, does not necessarily imply the presence of subthreshold symptomatology!
Boundaries with other disorders
Finally, the issue of boundaries among depression NOS, anxiety NOS, and AD remains problematic. How often are the major syndromes associated with a stressor and maladaptation? How different are the symptom profiles of depression and anxiety NOS from those of AD? How often does AD have a vegetative symptom (e.g. decreased appetite, libido, fatigue) or an ideational symptom (e.g. guilt, suicidality, hopelessness)? Much uncertainty about diagnostic boundary definition remains. The diagnosis of AD was consistently associated with shorter length of stay compared with major psychiatric diagnoses. (25,31) But, while Despland et al. (25) found a significantly greater number of Axis II comorbidities in AD patients compared with patients with other psychiatric diagnosis, Spalletta et al. (29) observed the prevalence of Axis II personality disorder to be lower in patients with AD with depressed mood than in those diagnosed as having major depression or dysthymia.
‘Splitting' and ‘lumping' continues; the subthreshold diagnosis of mixed anxiety–depressive disorder is a new category included in the DSM-IV. This disorder is very similar to AD with mixed mood; a boundary between the two is difficult to demarcate. The main difference between the two diagnoses was the chronicity of the mixed anxiety–depressive disorder (as was noted in the mixed anxiety–depression field trial). (32) The change in criterion C for AD—allowing a chronic or recurrent disturbance—confounds the differentiating of these two subthreshold diagnoses. This uncertainty is further complicated by the question of treatment. Is this an anxiety accompanied by depression which should be treated with anxiolitics, such as benzodiazapines, or is this a depression accompanied by anxiety, which should be treated with an antidepressant, such as a selective serotonin reuptake inhibitor? Furthermore, it is commonly viewed that the majority of patients with AD should be treated with psychotherapy or counselling as the initial approach.
Another potential mood disorder, subsyndromal symptomatic depression (SSD), has been suggested.(33) It joins AD in the grey area of subthreshold diagnoses. However, there are two critical differences between SSD and AD: SSD employs a symptom checklist, and it is not associated with a stressor. As was stated, AD trumps a problem-level diagnosis, but will be trumped by a specific psychiatric diagnosis, even if it is in the NOS category. By definition, ‘SSD is the simultaneous presence of any two or more symptoms of depression, persistent for most or all of the time for a duration of at least two weeks, associated with social dysfunction, and occurring in patients who do not meet the criteria for minor depression, major depression, and/or dysthymia'. (33) In some cases, the SSD diagnosis is the same as the DSM-IV diagnosis for minor depression, termed by the authors ‘SSD with mood disturbance', and has to be documented as such. In other cases, the disorder is SSD ‘without mood disturbance'. By definition SSD with mood disturbance, i.e. minor depression, would trump AD. But should SSD without mood disturbance trump AD? Research is necessary to demarcate the boundaries between the problem level, subthreshold, minor, and major disorders, and, in particular, with regard to the role of stressors as aetiological precipitants, concomitants, or essentially unrelated variables.
Age and medical comorbidity
In contrast with DSM-IIIR, DSM-IV has tried to accommodate the presence of comorbid medical illness. DSM-IIIR was regarded as ‘medical illness and age unfair' (i.e. inadequate consideration of age and/or medical illness) (L. George, personal communication, 1981). (34) To enhance reliability and validity, there will need to be a psychiatric taxonomy that takes into account medical illness and symptomatology and developmental epochs (e.g. children and adolescents, adults, ‘young' elderly, and ‘old' elderly). It is clear from the Western Psychiatric Institute studies that the symptom profile for children and adolescents is very different from that for adults with regard to the entire spectrum of diagnoses. It is also apparent that symptoms secondary to a medical illness are not to be ‘counted' in the algorithm for psychiatric diagnoses (e.g. anorexia, decreased energy) but they are extremely difficult to differentiate from the same symptoms which may be psychiatrically induced. As a possible solution Endicott (35) recommends replacing vegetative with ideational symptoms when evaluating depressed patients with medical illness. Rapp and Vrana (36) confirmed Endicott's proposed changes in the diagnostic criteria for depression in medically ill elderly persons and observed a maintenance of specificity and sensitivity respectively when substituting ideational for vegetative symptoms. With regard to age, recent studies report AD patients to be significantly younger compared with those with major psychiatric diagnosis.(25,37) Zarb's study(38) suggests that in cognitively impaired elderly, using individual items of the Geriatric Depression Scale, AD could be differentiated from major depression. In addition, Despland et al. (25) showed that patients labelled AD with depressive or mixed symptoms included more women: a sex ratio resembling that seen in major depression or dysthymia. The future evolution of the DSM needs to consider the effect of developmental epochs, gender, and medical comorbidity on symptom profiles in the various diagnostic categories.
The Epidemiologic Catchment Area Study did not include AD in its historic survey of patients in the population of five major settings in the United States. Most studies are of smaller or more discrete samples and have the problem of generalization. Andreasen and Wasek (39) reported that 5 per cent of an inpatient and outpatient sample at the university hospital and clinics in Iowa were labelled as having AD. Fabrega et al. (20) reported that 2.3 per cent of a sample of patients presenting to a walk-in clinic (diagnostic and evaluation centre) met criteria for AD, with no other diagnoses on Axis I or Axis II; when patients with other Axis I diagnoses (i.e. Axis I and II comorbidities) were included, 20 per cent had the diagnosis of AD. In general hospital inpatient psychiatric consultation populations, AD was diagnosed as 21.5 per cent, 18.5 per cent and 11.5 per cent respectively. (40,41 and 42) D. Schafer (personal communication, 1990) noted that up to 70 per cent of children in the psychiatric setting may be given the diagnosis of AD in a variety of mental health care environments. Faulstich et al. (43) reported the prevalence of DSM-III conduct and AD for adolescent psychiatric inpatients. Andreasen and Wasek, (39) utilizing a chart review, reported that more adolescents than adults experienced acting out and behavioural symptoms, but adults had significantly more depressive symptomatology (87.2 per cent versus 63.8 per cent). Anxiety symptoms were frequent at all ages.
Mezzich and coworkers(10,20) evaluated 64 symptoms currently present in three cohorts: subjects with specific diagnoses, those with AD, and those who were not ill. Vegetative, substance use, and characterological symptoms were greatest in the specific-diagnosis group, intermediate in the AD group, and least in the group with no illness. The symptoms of mood and affect, general appearance, behaviour, disturbance in speech and thought pattern, and cognitive functioning had a similar distribution. The AD group was significantly different from the no-illness group with regard to more ‘depressed mood' and ‘low self-esteem' ( p £0.0001). The AD and no-illness groups had minimal pathology of thought content and perception. A positive response on the suicide indicators was obtained in 29 per cent of AD compared with 9 per cent of the no-illness group. The three cohorts did not differ on the frequency of Axis III disorders.
Associated features of adjustment disorder
Andreasen and Wasek (39) observed that in their AD cohorts 21.6 per cent of the adolescents' fathers and 11.8 per cent of the adults' fathers had problems with alcohol. Greenberg et al. (31) report more substance abuse in adults with diagnosed AD compared with all those with other diagnoses. Breslow et al., (44) comparing patients with AD and other psychiatric diagnoses, observed that alcohol or substance use/abuse did not help to differentiate between diagnostic groups. Thus, higher rate of substance use at this time does not serve as an incontrovertible prediction factor for the diagnosis of an AD diagnosis.
Aetiology—the role of stress
Nature of the stressor
Hirschfeld (45) and Winokur (46) discussed both sides of the controversy regarding ‘neurotic' depression (i.e. related to a stressor) and ‘endogenous' depression (i.e. not related to a stressor). From the examination of several studies, it has been difficult to demonstrate a significant temporal link between the onset of an identified stressor and the occurrence of depressive illness. (45,46,47,48,49,50 and 51)
Andreasen and Wasek (39) described the differences between the chronicity of stressors found in adolescents compared with those in adults: present for a year or more, 59 per cent and 35 per cent; present for 3 months or less, 9 per cent and 39 per cent. Fabrega et al. (20) reported that their AD group had greater registration of stressors compared with other diagnoses and the non-illness cohorts. Compared with other diagnoses and the non-illness patients, AD was over-represented in the ‘higher stress category'. In their consultation cohort, Popkin et al. (40) found that in 68.6 per cent of the cases the medical illness itself was judged to be the primary psychosocial stressor. Snyder and Strain (42) observed that stressors as assessed on Axis IV were significantly higher ( p = 0.0001) for consultation patients with AD than for patients with other diagnostic disorders.
Modifiers of stress
Stress has been described as the aetiological agent for AD. Vulnerability to stress is another risk factor. Diverse variables and modifiers are involved regarding who will experience AD following a stress. Cohen (22) argues as follows:
- acute stresses are different from chronic stresses in both psychological and physiological terms;
- the meaning of the stress is affected by ‘modifiers' (e.g. ego strengths, support systems, prior mastery);
- manifest and latent meanings of the stressor(s) may be associated with differential impact (e.g. loss of job may be a relief or a catastrophe).
AD with maladaptive denial of pregnancy, for example, can be a consequence of a stressor such as separation from a partner. (52) An objectively overwhelming stress may have little impact on one individual, whereas a minor stress could be regarded as cataclysmic by another. A recent minor stress superimposed on a previous underlying (major) stress that has no observable effect on its own may have a significant additive impact (i.e. concatenation of events) (B. Hamburg, personal communication, 1990).
The chronological relationship of the stressor and symptoms has been examined less extensively. Depue and Monroe (53) and Skodol et al. (54) identified significant methodological problems in evaluating the quality, quantity, and timing of both stressors and symptoms. Depue and Monroe (53) and Rahe (55) state that the model of a single stressor impinging on an undisturbed individual to cause symptoms at a single point in time is insufficient to account for the many presentations of stress and illness in the clinical situation.
Confounds: assessing stress
Limitations of the current construct of stress for research have been described. (22) Holmes and Rahe(56) assigned relative values to specific stressors, but there has been much concern about their methodology and the results obtained. (22) Other life-event scales(51,57,58) have also been shown to be inconsistent in their ability to link stress and illness. As discussed above, many authors have cautioned that the vulnerability of the individual (e.g. ego strengths, support system, underlying personality disorders, the timing and concatenation of the stress(es), the issue of control over the stressor, and the desirability of the event, etc.) need to be evaluated to ascertain the impact of the stressor on the individual. Axis IV of DSM-III was included to allow the clinician to assess the presence of stress in the multiaxial diagnoses of psychiatric disorders, but it has been confounded by low reliability. (59,60 and 61) Despland et al. (25) reported that stressors were present on Axis IV in 100 per cent of those assigned AD with depressed mood, while it was present in 83 per cent of those with major depression, 80 per cent of those with dysthymia, and 67 per cent of those with non-specific depression, which emphasizes the importance of stressors in the AD diagnosis.
Nine different types of AD are listed in DSM-IIIR. (1) As in DSM-III, AD is classified in DSM-IIIR according to the predominant symptom picture. In DSM-IV, AD has been reduced to six types that, again, are classified according to their clinical features:
- AD with depressed mood;
- AD with anxious mood;
- AD with mixed anxiety and depressed mood;
- AD with disturbance of conduct;
- AD with mixed disturbance of emotions and conduct;
- AD not otherwise specified.
In their study, Despland et al. (25) suggested reducing the subtypes even further, demonstrating identical profiles for AD with depressed mood and AD with mixed mood, and proposing assimilation of mixed mood into the depressed mood category. Fifty-seven per cent of their sample were represented by these two groups; the remainder were accounted for by AD with ‘anxiety' and ‘other' categories.
Psychotherapy and counselling
Treatment of AD initially focuses on psychotherapeutic and counselling interventions to reduce the stressor, enhance the capacity to cope with a stressor that cannot be reduced or removed, and establish a system of support to maximize adaptation. The patient needs to be made aware of the significant dysfunction that the stressor has caused and consider strategies to manage the disability. Some stressors, for example taking on more responsibility than can be managed by the individual, or putting oneself at risk (e.g. unprotected sex with an unknown partner), can be avoided or minimized. Other stressors may elicit an over-reaction on the part of the patient (e.g. abandonment by a lover). The patient may attempt suicide or become reclusive, damaging his or her source of income. In this situation, the therapist would assist the patient to verbalize his or her disappointed feelings and rage rather than behaving destructively. The role of verbalization in minimizing the discomfort of the stressor and enhancing coping cannot be overestimated. It is necessary to clarify and interpret the meaning/reality of the stressor for the patient. For example, if a mastectomy has devastated a patient's feelings about her body and herself, it is mandatory to articulate that the patient is still a woman, capable of having a fulfilling relationship, including a sexual one, and that recurrence of the cancer may not occur. Without the correction of distortions, the patient's pernicious fantasies'—all is lost'—may occur as sequelae to the stressor (i.e. the mastectomy) and intensify incapacitation at work and/or sex, as well as a profound disturbance of mood.
Counselling, psychotherapy, crisis intervention, family therapy, and group treatment are utilized to encourage the verbalization of fears, anxiety, rage, helplessness, and hopelessness to the stressors imposed upon a patient. As mentioned above, the goal of treatment is to expose the concerns and conflicts that the patient is experiencing, identify means to reduce the stressor(s), enhance the patient's coping skills, and clarify the patient's perspective on the adversity, and enable the establishment of supporting relationships. The primary treatment for AD is talking.
Small doses of antidepressants and anxiolytics may sometimes be appropriate for AD patients when dysphoria remains profound despite several sessions of psychological treatment.
No randomized controlled trials of pharmacological treatment of AD are available. Although formal psychotherapy is presently the treatment of choice, (62) psychotherapy combined with benzodiazapines is utilized, especially for patients with severe life stress(es) and an unrelenting anxious component. (62) Tricyclic antidepressants or buspirone were recommended in place of benzodiazapines for AD patients with current or past excessive alcohol use because of their greater risk of dependence. (62) The use of antidepressants may assist some patients if their maladaptation is debilitating and the accompanying mood is pervasive.
Course and prognosis
Adults and adolescents
Andreasen and Hoenk (63) report that the long-term outcome of AD has a good prognosis for adults, but that a majority of adolescents eventually have major psychiatric disorders. Follow-up at 5 years after original diagnosis of AD revealed that 71 per cent of adults were completely well, 8 per cent had an intervening problem, and 21 per cent had developed a major depressive disorder or alcoholism. However, in adolescents at 5-year follow-up, only 44 per cent were without a psychiatric diagnosis, 13 per cent had an intervening psychiatric illness, and 43 per cent went on to develop major psychiatric morbidity (e.g. schizophrenia, schizoaffective disorders, major depression, bipolar disorder, substance abuse, and personality disorders). In contrast with the adults, the chronicity of the illness and the presence of behavioural symptoms in the adolescents were the strongest predictors for major psychopathology 5 years after the initial AD diagnosis. The number and type of symptoms were less useful than the length of treatment and chronicity of symptoms as predictors of future outcome.
As Chess and Thomas (64) have reported, it is important to note that AD with disturbance of conduct, regardless of age, has a more guarded outcome. In agreement with the findings of Andreasen and Wasek. (39) Chess and Thomas (64) emphasize that:
a significant number [of AD adolescents] did not improve or even grew worse in adolescence and early adult life. It was not possible to predict the developmental course of the disorder in the early period after its first identification. Hence, we would suggest active appropriate therapeutic intervention in all cases but especially adolescents [and adequate follow-up].
Spalletta et al. (29) report that suicidal behaviour and deliberate self-harm are important predictors in the diagnosis of AD. As mentioned before, these are obviously not subthreshold symptoms; they can lead to the most dire consequence—death. This outcome, when reached, can be neither corrected nor resolved. These behaviours mandate immediate and protective interventions. The diagnosis of AD may suggest that the patient has minor symptomatology. Such assessment may be life-threatening. There needs to be a definite split from viewing a diagnosis as subthreshold, and therefore the attendant symptoms as subthreshold. It is similar to labelling a patient with hypochondriasis, which in some settings can influence a more casual physical assessment, when such a patient could have serious physical morbidity concomitant with their hypochondriacal Axis I pathology.
As mentioned earlier, the diagnosis of an AD may be in the early phase of an evolving disorder that has not yet developed to the extent that full-blown symptoms are evident to reach threshold for a major psychiatric disorder. If a patient continues to worsen, becomes more symptomatic, and does not respond to treatment, it is critical to review the diagnosis for the presence of a major disorder.
The domains of diagnostic rigour and clinical utility seem at odds for AD. Studies that employ reliable and valid instruments (e.g. depression or anxiety rating scales, stress assessments, length of disability, treatment outcome, family patterns, etc.) would enhance more exact specification of the parameters of the AD diagnosis. Identification of the time course, remission or evolution to another diagnosis, and the evaluation of stressors (characteristics, duration, and the nature of adaptation to stress) would enhance the understanding of the aetiology, mechanisms, and mediators of a stress-response illness.
Studies with adequate symptom checklists rated independently from the establishment of the diagnosis would clarify the threshold between major and minor depression and anxiety, as well as guide an entry threshold to employ the AD diagnosis. Although the upper threshold is established by the specified criteria for the major syndromes, the entry threshold between an AD and problem-level diagnoses and normality is undesignated with operational criteria. The careful examination of associated demographic and treatment outcome variables would also enable clinicians to describe more specifically the boundaries among subthreshold diagnoses, problem-level diagnosis, and normal behaviour. Associated features such as family history, biological correlates, treatment response, long-term course, and so forth, are all critical to establishing the authenticity of a diagnosis. The theory and practice of medicine have demonstrated the need for a comprehensive multidimensional formulation of all these physiological and functional variables to describe an illness and develop the most appropriate working diagnosis.
Subthreshold syndromes can encompass significant psychopathology that must not only be identified but treated (e.g. suicidal ideation/behaviour). Cross-sectionally, AD may appear to be the incipient phase of an emerging major syndrome. Consequently, AD, despite its questionable reliability and validity, serves an important diagnostic function in the practice of psychiatry. Problem- and subthreshold-level diagnoses are critical to the function of any medical discipline. Because this may be the initial phase, or a mild form, of a dysfunction that is not yet fully developed, there is a need to describe the relationship of this incipient state to other potential diagnoses. This lack of specificity and questionable reliability and validity are the hallmark of interface disorders and subthreshold phenomena, whether they be in diabetes mellitus, hypertension, or depression.
Should drugs be used in the treatment of AD? The pharmacological studies are not conclusive. The diagnostic dilemmas of the AD present sufficient difficulty in and of themselves.(65,66 and 67) It would be preferred that cautious psychotropic drug administration be employed, to avoid subjecting the patient to the risk of unfavourable drug–psychotrophic drug interaction. Psychotrophic drug treatment will not be necessary if the condition resolves. If it evolves into a major psychiatric illness then drug treatment needs to be considered.
As mentioned earlier, the characteristics of a mental disorder vary over the lifecycle, and this is clearly illustrated by AD. Certain developmental epochs may be associated with a particular symptom profile, as seen with acute myocardial infarction or appendicitis. The effect of the stressor may vary, and the assessment of functioning must be ‘measured' according to the demands of the developmental stage (e.g. school (adolescents), work (adults), self-care and maintenance (elderly)). The symptom characteristics and functional assessment of other diagnoses may also vary along the developmental schema from birth to senescence; illnesses such as major depressive disorders, organic mental disorders, sexual dysfunctions, and eating disorders need to be ‘recast' in another hierarchy to incorporate the stage of the lifecycle extant at the time of the assessment, and symptom profiles adjusted accordingly. The normal variations across developmental epochs would make AD and the other psychiatric disorders more reliable and valid across the lifecycle. Similarly there needs to be a consideration of a possible concomitant state of medical illness. The result would be a taxonomy tempered by the vicissitudes of development and medical illness.
A taxonomy which considers the development epoch and the presence of medical illness would be more useful to child psychiatrists, paediatricians, geriatricians, geriatric psychiatrists, and primary care specialists, who often are convinced that a patient does not conform with today's psychiatry's lexicon. A significant number of their patients remain at the problem level of diagnoses with their somatic complaints as well. It is not uncommon for a fever of unknown origin to not be diagnosed, or for a chest pain to remain unspecified. It is the art of medicine that makes it a profession, and a most difficult one, at the interface of medicine and psychiatry, or at the interface of normality and pathology. Anna Freud (68) has emphasized the difficulty of understanding normality and pathology in her assessments of childhood. This important advice would obtain across the lifecycle and be an important challenge to the developers of the subthreshold diagnoses (e.g. AD) and the future evolution of the DSM.
1. American Psychiatric Association (1994). Diagnostic and statistical manual of mental disorders (4th edn). American Psychiatric Association, Washington, DC.
2. Spiegel, D. (1994). DSM-IV options book. American Psychiatric Association, Washington, DC.
3. Wise, M.G. (1988). Adjustment disorders and impulse disorders not otherwise classified. In American Psychiatric Press textbook of psychiatry (ed. J.A. Talbot, R.E. Hales, and S.C. Yudofsky), pp. 605–20. American Psychiatric Press, Washington, DC.
4. American Psychiatric Association (1980). Diagnostic and statistical manual of mental disorders (3rd edn). American Psychiatric Association, Washington, DC.
5. American Psychiatric Association (1987). Diagnostic and statistical manual of mental disorders (3rd edn, revised). American Psychiatric Association, Washington, DC.
6. Kovacs, M., Ho, V., and Pollock, M.H. (1995). Criterion and predictive validity of the diagnosis of adjustment disorder: a prospective study of youths with new-onset insulin-dependent diabetes mellitus. American Journal of Psychiatry, 152, 523–8.
7. Perez-Jimenez, J.P., Gomez-Bajo, G.J., Lopez-Catillo, J.J., et al. (1994). Psychiatric consultation and post-traumatic stress disorder in burned patients. Burns, 20, 532–6.
8. Sullivan, M.J., Winshenker, B., and Mikail, S. (1995). Screening for major depression in the early stages of multiple sclerosis. Canadian Journal of Neurological Science, 22, 228–31.
9. Shima, S., Kitagawa, Y., Kitamura, T., et al. (1994). Poststroke depression. General Hospital Psychiatry, 16, 286–9.
10. Mezzich, J.E., Dow, J.T., Rich, C.L., et al. (1981). Developing an efficient clinical information system for a comprehensive psychiatric institute, II: Initial Evaluation Form. Behavioral Research Methods and Instrumentation, 13, 464–78.
11. Aoki, T., Hosaka, T., and Ishida, A. (1995). Psychiatric evaluation of physical rehabilitation patients. General Hospital Psychiatry, 17, 440–3.
12. Zung, W. (1971). A rating intrument for anxiety disorders. Psychosomatics, 12, 371–9.
13. Zung, W. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63–70.
14. McNair, D.M., Lorr, M., and Doppelman, L.F. (ed.) (1971). Manual for the Profile of Mood States . Educational and Industrial Testing Service, San Diego, CA.
15. Margolis, R.L. (1994). Nonpsychiatrist house staff frequently misdiagnose psychiatric disorders in general hospital inpatients. Psychosomatics, 35, 485–91.
16. Fincannon, J.L. (1995). Analysis of psychiatric referrals and interventions in an oncology population. Oncology Nursing Forum, 22, 87–92.
17. Silverstone, P.H. (1996). Prevalence of psychiatric disorders in medical inpatients. Journal of Nervous and Mental Disease, 184, 43–51.
18. Strain, J.J., Newcorn, J., Mezzich, J., et al. (1998). Adjustment disorder: the McArthur reanalysis. In DSM-IV source book, Vol. 4, pp. 404–24. American Psychiatric Association, Washington, DC.
19. Kleinman, A. (1980). Patients and healers in the context of culture: an exploration of the borderland between anthropology, medicine, and psychiatry. University of California Press, Berkeley, CA.
20. Fabrega, H. Jr, Mezzich, J.E., and Mezzich, A.C. (1987). Adjustment disorder as a marginal or transitional illness category in DSM-III. Archives of General Psychiatry, 44, 567–72.
21. Bodlund, O., Kullgren, G., Ekselius, L., et al. (1994). Axis V—Global Assessment of Functional Scale. Evaluation of a self-report version. Acta Psychiatrica Scandinavica, 90, 342–7.
22. Cohen, F. (1981). Stress and bodily illness. Psychiatric Clinics of North America, 4, 269–86.
23. Perris, H., von Knorring, L., Oreland, L., et al. (1984). Life events and biological vulnerability: a study of life events and platelet MAO activity in depressed patients. Psychiatry Research, 12, 111–20.
24. Zilberg, N.J., Weiss, D.S., and Horowitz, M.J. (1982). Impact of Event Scale: a cross-validation study and some empirical evidence supporting a conceptual model of stress response syndromes. Journal of Consulting and Clinical Psychology, 50, 407–14.
25. Despland, J.N., Monod, L., and Ferrero, F. (1995). Clinical relevance of adjustment disorder in DSM-III-R and DSM-IV. Comprehensive Psychiatry, 36, 456–60.
26. De Leo, D., Pellegrini, C., and Serraiotto, L. (1986). Adjustment disorders and suicidality. Psychology Reports, 59, 355–8.
27. De Leo, D., Pellegrini, C., Serraiotto, L., et al. (1986). Assessment of severity of suicide attempts: a trial with the dexamethasone suppression test and two rating scales. Psychopathology, 19, 186–91.
28. Isometsa, E., Heikkinen, M., Henriksson, M., et al. (1996). Suicide in non-major depressions. Journal of Affective Disorders, 36, 117–27.
29. Spalletta, G., Troisi, A., Saracco, H., et al. (1996). Symptom profile: Axis II comorbidity and suicidal behaviour in young males with DSM-III-R depressive illnesses. Journal of Affective Disorders, 39, 141–8.
30. Vlachos, I.O., Bouras, N., Watson, J.P., et al. (1994). Deliberate self-harm referrals. European Journal of Psychiatry, 8, 25–8.
31. Greenberg, W.M., Rosenfeld, D.N., and Ortega, E.A. (1995). Adjustment disorder as an admission diagnosis. American Journal of Psychiatry, 152, 459–61.
32. Zinbarg, R.E., Barlow, D.H., Liebowitz, M., et al. (1994). The DSM-IV field trial for mixed anxiety–depression. American Journal of Psychiatry, 151, 1153–62.
33. Judd, L.L., Rapaport, M.H., Paulus, M.P., et al. (1994). Subsyndromal symptomatic depression: a new mood disorder? Journal of Clinical Psychiatry, 55 (Supplement), 18–28.
34. Strain, J.J. (1981). Diagnostic considerations in the medical setting. Psychiatric Clinics of North America, 4, 287–300.
35. Endicott, J. (1984). Measurement of depression in patients with cancer. Cancer, 53, 2243–9.
36. Rapp, S.R. and Vrana, S. (1989). Substituting nonsomatic for somatic symptoms in the diagnosis of depression in elderly male medical patients. American Journal of Psychiatry, 146, 1197–200.
37. Mok, H. and Walter, C. (1995). Brief psychiatric hospitalization: preliminary experience with an urban sort-stay unit. Canadian Journal of Psychiatry, 40, 415–17.
38. Zarb, J. (1996). Correlates of depression in cognitively impaired hospitalized elderly referred for neuropsychological assessment. Journal of Clinical and Experimental Neuropsychology, 18, 713–23.
39. Andreasen, N.C. and Wasek, P. (1980). Adjustment disorders in adolescents and adults. Archives of General Psychiatry, 37, 1166–70.
40. Popkin, M.K., Callies, A.L., Colón, E.A., et al. (1990). Adjustment disorders in medically ill patients referred for consultation in a university hospital. Psychosomatics, 31, 410–14.
41. Foster, P. and Oxman, T. (1994). A descriptive study of adjustment disorder diagnoses in general hospital patients. Irish Journal of Psychological Medicine, 11, 153–7.
42. Snyder, S. and Strain, J.J. (1989). Differentiation of major depression and adjustment disorder with depressed mood in the medical setting. General Hospital Psychiatry, 12, 159–65.
43. Faulstich, M.E., Moore, J.R., Carey, M.P., et al. (1986). Prevalence of DSM-III conduct and adjustment disorders for adolescent psychiatric inpatients. Adolescence, 21, 333–7.
44. Breslow, R.E., Klinger, B.I., and Erickson, B.J. (1996). Acute intoxication and substance abuse among patients presenting to a psychiatric emergency service. General Hospital Psychiatry, 18, 183–91.
45. Hirschfeld, R.M.A. (1981). Situational depression: validity of the concept. British Journal of Psychiatry, 139, 297–305.
46. Winokur, G. (1985). The validity of neurotic-reactive depression: new data and reappraisal. Archives of General Psychiatry, 42, 1116–22.
47. Akiskal, H.S., Bitar, A.H., Puzantian, V.R., et al. (1978). The nosological status of neurotic depression: a prospective three- to four-year follow-up examination in light of the primary-secondary and unipolar–bipolar dichotomies. Archives of General Psychiatry, 35, 756–66.
48. Andreasen, N.C. and Winokur, G. (1979). Secondary depression: familial, clinical, and research perspectives. American Journal of Psychiatry, 136, 62–6.
49. Benjaminsen, S. (1981). Stressful life events preceding the onset of neurotic depression. Psychological Medicine, 11, 369–78.
50. Garvey, M.J., Tollefson, G.D., Mungas, D., et al. (1984). Is the distinction between situational and nonsituational primary depression valid? Comprehensive Psychiatry, 25, 372–5.
51. Paykel, E.S., Prusoff, B.A., and Uhlenhuth, E.H. (1971). Scaling of life events. Archives of General Psychiatry, 25, 340–7.
52. Brezinka, C., Huter, O., Biebl, W., et al. (1994). Denial of pregnancy: obstetrical aspects. Journal of Psychosomatic Obstetrics and Gynecology, 15, 1–8.
53. Depue, R.A. and Monroe, S.M. (1986). Conceptualization and measurement of human disorder in life stress research: the problem of chronic disturbance. Psychological Bulletin, 99, 36–51.
54. Skodol, A.E., Dohrenwend, B.P., Line, B.G., et al. (1990). The nature of stress: problems of measurement. In Stressors and the adjustment disorders (ed. J.D. Noshpitz and R.D. Coddington), pp. 3–22. Wiley, New York.
55. Rahe, R.H. (1990). Psychosocial stressors and adjustment disorder: Van Gogh's life chart illustrates stress and disease. Journal of Clinical Psychiatry, 51 (Supplement), 13–19.
56. Holmes, T.H. and Rahe, R.H. (1967). The Social Readjustment Rating Scale. Journal of Psychosomatic Research, 11, 213–18.
57. Dohrenwend, B.S., Krasnoff, L., Askenasy, A.R., et al. (1978). Exemplification of a method for scaling life events: the PERI Life Event Scale. Journal of Health and Social Behaviour, 19, 205–29.
58. Tennant, C. (1983). Life events and psychological morbidity: the evidence from prospective studies (editorial). Psychological Medicine, 13, 483–6.
59. Rey, J.M., Stewart, G.W., Plapp, J.M., et al. (1988). DSM-III Axis IV revisited. American Journal of Psychiatry, 145, 286–92.
60. Spitzer, R.L. and Forman, J.B.W. (1979). DSM-III field trials, II: initial experience with the multiaxial system. American Journal of Psychiatry, 136, 818–20.
61. Zimmerman, M., Pfohl, B., Coryell, W. et al. (1987). The prognostic validity of DSM-III Axis IV in depressed inpatients. American Journal of Psychiatry, 144, 102–6.
62. Uhlenhuth, E.H., Balter, M.B., Ban, T.A., et al. (1995). International study of expert judgment on therapeutic use of benzodiazepines and other psychotherapeutic medications: III. Clinical features affecting experts' therapeutic recommendations in anxiety disorders. Psychopharmacology Bulletin, 31, 289–96.
63. Andreasen, N.C. and Hoenk, P.R. (1982). The predictive value of adjustment disorders: a follow-up study. American Journal of Psychiatry, 139, 584–90.
64. Chess, S. and Thomas, A. (1984). Origins and evolution of behavior disorders: from infancy to early adult life. Brunner–Mazel, New York.
65. Hosaka, T., Aoki, T., and Ichikawa, Y. (1994). Emotional states of patients with hematological malignancies: preliminary study. Japanese Journal of Clinical Oncology, 24, 186–90.
66. Oxman, T.E., Barrett, J.E., Freeman, D.H., et al. (1994). Frequency and correlates of adjustment disorder relates to cardiac surgery in older patients. Psychosomatics, 35, 557–68.