Life stress and symptom pattern in out-patient depression

Epidémiologie de la dépression: données récentes. II — Epidémiologie analytique et épidémiologie d’évaluation

Les données provenant d’études récentes concernant l’épidémiologie analytique et l’épidémiologie d’évauation de la dépression sont examinées.

Au sujet de l’épidémiologie analytique, il peut être conclu de la littérature actuellement disponible que les principaux facteurs de risques pour la dépression majeure sont : a) Sociodémographiques, à savoir : être une femme, être jeune, séparé, divorcé, ou avoir des problèmes conjugaux (les auteurs soulignent en particulier que les différences entre les sexes, concernant la dépression, sont réelles et ne sont pas un artefact lié à la façon de relater les troubles ou au comportement vis-à-vis des soins), b) Environnementaux: facteurs prédisposants (les événements de vie qui surviennent durant l'enfance peuvent prédisposer un individu à la dépression à l’âge adulte) et facteurs précipitants (il y a relation entre la survenue d’événements de vie pendant la vie adulte et le début de la dépression), c) Familiaux: il y a une multiblication par 2 à 5 du taux de dépression majeure chez les parents de premier degré de sujets témoins par rapport à es témoins non malades. L’influence de l’hérédité génétique est supportée par des études de jumeaux et des études adoption, mais une large part de la variance ne peut être expliquée, d) Divers: qui concernent le cycle de reproduc- 10n de la femme (la période du post partum entraîne une augmentation du taux de dépression) et les variations saisonneres (pics au printemps et en automne).

En ce qui concerne l'épidémiologie d’évaluation, il est habituel de distinguer prévention primaire, secondaire et teriaie. Leur dessein est respectivement de diminuer l’incidence de la dépression, la prévalence de la dépression et les isques de chronicité et de la récurrence. Les efforts concernant la prévention primaire ne peuvent avoir qu’un effet imité. Dans le domaine de la prévention secondaire, il peut être conclu que beaucoup de dépressifs ne sont pas diaglostiqués ou ne sont pas traités. La prévention tertiaire a pour dessein de limiter deux risques: la chronicité et la récurence. Le premier pourrait être évalué à 15-20% à chaque épisode. Les facteurs de risques principaux seraient la présence un trouble psychiatrique non affectif ou d’une affection physique, une personnalité névrotique et un faible niveau le traitement. Pour le risque de récurrence, on peut considérer qu’entre 50 et 85% des patients ayant fait un épisode lépressif majeur feront au moins un autre épisode de dépression ultérieurement. Les facteurs de risques pourraient re un Stand nombre d’épisodes antérieurs, un antécédent d’épisode maniaque ou d’hypomanie, l’association à un ésordre psychiatrique non affectif ou à une affection physique, une histoire familiale de maladie affective ou un âge e début tardif.

Animal models of anxiety disorders and stress

Anxiety and stress-related disorders are severe psychiatric conditions that affect performance in daily tasks and represent a high cost to public health. The initial observation of Charles Darwin that animals and human beings share similar characteristics in the expression of emotion raise the possibility of studying the mechanisms of psychiatric disorders in other mammals (mainly rodents). The development of animal models of anxiety and stress has helped to identify the pharmacological mechanisms and potential clinical effects of several drugs. Animal models of anxiety are based on conflict situations that can generate opposite motivational states induced by approach-avoidance situations. The present review revisited the main rodent models of anxiety and stress responses used worldwide. Here we defined as "ethological" the tests that assess unlearned/unpunished responses (such as the elevated plus maze, light-dark box, and open field), whereas models that involve learned/punished responses are referred to as "conditioned operant conflict tests" (such as the Vogel conflict test). We also discussed models that involve mainly classical conditioning tests (fear conditioning). Finally, we addressed the main protocols used to induce stress responses in rodents, including psychosocial (social defeat and neonatal isolation stress), physical (restraint stress), and chronic unpredictable stress.

Life events and dementia: what is the nature of their relationship?

The current study analyzed the life events reported by 1271 demented patients vs. 140 cognitively healthy elderly subjects. The Life Change Unit (LCU) method was used to quantify the results. When all the events were included in the analysis, the two groups had similar LCU scores (61.26 vs. 63.42). However, when events causally related to dementia (e.g. stroke) are excluded, demented patients were found to experience half of the LCU load in comparison to controls (30.70 vs. 63.42). In both groups the level of LCU load is far below 100 which is the threshold suggested for the induction of psychosomatic disorders. Conclusively, the current study suggests that there is no causal role for life events in the etiopathogenesis of dementia. On the contrary, demented patients even the last few months before the clinical onset of dementia experience low life-events-related stress, possibly because of subclinical impairment which is already present.

The emerging modern face of mood disorders: a didactic editorial with a detailed presentation of data and definitions

The present work represents a detailed description of our current understanding and knowledge of the epidemiology, etiopathogenesis and clinical manifestations of mood disorders, their comorbidity and overlap, and the effect of variables such as gender and age. This review article is largely based on the 'Mood disorders' chapter of the Wikibooks Textbook of Psychiatry http://en.wikibooks.org/wiki/Textbook_of_Psychiatry/Mood_Disorders.

Stressful life events, chronic difficulties, and the symptoms of clinical depression

Major life events and chronic difficulties have been found to be associated with the onset of depression. Little is known, however, about how exposure to such stressors is related to the clinical presentation of this disorder. We addressed this issue by administering an interview-based measure of life stress, the Beck Depression Inventory, and the Global Assessment of Functioning scale to 100 adults diagnosed with major depressive disorder. Participants who experienced a preonset severe life event exhibited greater overall levels of depression severity, endorsed more cognitive and somatic symptoms of depression, and functioned at lower levels than did their counterparts without preonset severe life events. In contrast, exposure to a preonset severe difficulty was unrelated to participants' severity of depression, cognitive and somatic symptoms, or level of global functioning. These findings highlight the potentially greater importance of acute stress compared with chronic stress for influencing these key clinical features of depression.

Life events and clinical subtypes of major depression: a cross-sectional study

The present study aimed to investigate the relationship between stressful life events and depression. Ninety major depressive patients, aged 36.57 +/- 12.02 years, and 121 matched controls, aged 27.11 +/- 10.59 years, took part in the study. The life change units (LCU) according to Rahe et al. were used for the quantification of life events. The results suggested that males reported significantly more life events than females. Atypical males differed from all other gender-by-type groups, while no other differences emerged. LCU correlated with melancholic indices but did not correlate with anxiety or functioning. The results suggest that subgroups of depressive patients are characterized by the presence of stressful life events. Male atypical patients reported the highest load of life events. The question that arises is whether this finding is valid or whether these patients (who have higher personality psychopathology and interpersonal rejection sensitivity) tend to over-report life events.

Severe melancholic depression is more vulnerable than non-melancholic depression to minor precipitating life events

BACKGROUND

The present study examines the moderating role of global depression severity on the relation of melancholic versus non-melancholic depression to severe and non-severe levels of stress.

METHOD

A community sample of 50 women with unipolar major depressive disorder, of which 54% met Research Diagnostic Criteria for melancholic depression, were interviewed regarding stressful life events experienced prior to onset. Events were coded as severe or non-severe based on the rigorous Bedford College contextual rating system.

RESULTS

Greater severity of depression was related to a higher likelihood of a severely stressful event prior to onset only for women with non-melancholic major depression. By contrast, greater severity of depression was related to a higher likelihood of a non-severe, more minor, stressful event prior to onset only for women with melancholic major depression.

LIMITATIONS

The present study was limited by its use of a female volunteer sample, which might not be entirely representative of the population of individuals with major depression. In addition, the study employed a cross-sectional design, which limits conclusions relating to the causal relation of stress to melancholic versus non-melancholic depression.

CONCLUSIONS

Far from being autonomous of stress, individuals with severe melancholic depression may be especially sensitive to stress, such that their episodes are influenced by more minor stressors than those of individuals with non-melancholic depression.

Life events, social support, and onset of major depressive episode in Finnish patients

We investigated differences in life events and social support between subgroups of depressed patients and the distribution of life events in phases preceding or during depression. In the Vantaa Depression Study, 269 psychiatric patients with DSM-IV major depressive disorder were diagnosed with Schedule for Clinical Assessment in Neuropsychiatry, Version 2.0, and Structured Clinical Interview for DSM-III-R personality disorders (SCID-II). Life events during the 12 months preceding the interview were investigated with the Interview for Recent Life Events, and social support with the Interview Measure of Social Relationships and the Perceived Social Support Scale-Revised. Nearly all patients (91%) reported life events, on average 4.1 per preceding year. No major differences between sociodemographic or clinical subgroups were found; the frequency of events was somewhat greater among the younger subjects, whereas those with comorbid alcoholism or personality disorders perceived less social support. Although events were distributed evenly between the time preceding depression, the prodromal phase, and the index major depressive episode, two thirds of the patients attributed their depression to some event. Despite clinical and sociodemographic heterogeneity, patients with major depressive disorder are fairly homogeneous in terms of life events during the preceding year. Events do not cluster in any particular phase of the progression to an episode.

Life events and affective disorders

OBJECTIVE

To summarize research in life events and affective disorders

METHOD

Review of the literature.

RESULTS AND CONCLUSION

Many studies have shown that episodes of unipolar depression are preceded by life events at higher rates than in control samples. A variety of stressful events are involved with only limited specificity but some relationship to social exit events. Life events also affect remission and relapse of depression. Effects of life events are less when the disorder has already been recurrent, particularly where illness is severe. Effects of life events are also weaker in bipolar disorder than unipolar, but major life events may be important in first onset. Recent evidence suggests a specific role of social rhythm disruption events in manic relapses. Causative chains are multifactorial and complex, and genetic factors can influence life event exposure.

Which depressions are related to life stress?

This paper examines the relationship of recent life events to specific kinds of depression using published studies and the author's own work. An overall effect of life events on depression has been found consistently and is moderate in degree. In suicide attempts there are stronger and more immediate effects than in depression. Life events precede both non-melancholic and melancholic depressions. It is only in recurrent depressions that life events are less common with melancholic pictures. Life events influence bipolar disorder as well as unipolar. Mania may be preceded by life events, particularly those involving social rhythm disruption, but it is harder to rule out events which are consequences of insidious development of illness. There are strong effects of life events and social support in postpartum depressions but in postpartum psychoses these effects are absent. Events precede depression comorbid with other disorders as well as pure depression. The course of depression is also influenced by life stress with less remission where negative events occur after onset and better outcome where earlier adverse events are neutralized. Relapse is related to immediately preceding life events. However, where depressions are both severe and recurrent life stress effects weaken and as the number of episodes increases preceding life events lessen. These findings suggest that some kinds of depression are more related to psychosocial causation and some are more biological in origin.

Life stress and the symptoms of major depression

Life stress has been found to be associated with onset of depression and with greater severity of depressive symptoms. It is unclear, though, if life stress is related to particular classes or specific symptoms in depression. The association between severe life events and depressive symptoms was tested in 59 individuals diagnosed by Research Diagnostic Criteria with endogenous primary nonpsychotic major depression. As predicted, life stress was associated principally with cognitive-affective symptoms, not somatic symptoms. There also was a consistent association across different assessment methods between severe events and suicidal ideation. Finally, associations held specifically for severe events occurring before onset, not for severe events occurring after onset. Symptom variation in major depression is related specifically to severe stressors before onset and includes primarily cognitive-affective types of symptoms. There is an especially pronounced association of prior severe stress with suicidal ideation. The implications of stress-symptom associations are addressed for enlarging understanding of symptom heterogeneity and subtype distinctions in major depression.

Program and trainee lifestyle stress: a survey of AAMFT student members. American Association for Marriage and Family Therapy

Student members of the American Association for Marriage and Family Therapy (n = 900) were randomly selected and surveyed regarding perceptions of certain MFT training program and graduate student lifestyle stressors. The return rate was 329 (37%) of the 900 students surveyed. The sample was 70% female and 30% male. Sixty-five percent of the sample were married. Sixty-one percent were in master's, 25% were in doctoral, and 12% were in other postgraduate programs. Potential high stressors were the number of hours students worked to support themselves and the number of hours spouses or significant others worked. Twenty-seven percent of students had considered dropping out due to program demands. Only 11.2% had actually done so. Nearly 11% reported current antidepressant use. Recommendations for further research directions on the topic of MFT trainee stress are provided.

Psychological sequelae of torture and organised violence suffered by refugees from Iraq. Trauma-related factors compared with social factors in exile

BACKGROUND

Refugees who have suffered traumatic events present complex therapeutic challenges to health professionals. There is little research into post-exile factors that may be amenable to change, and therefore reduce morbidity. We examined the importance of social factors in exile and of trauma factors in producing the different elements of psychological sequelae of severe trauma.

METHOD

Eighty-four male Iraqi refugees were interviewed. Adverse events and level of social support were measured. Various measures of psychological morbidity were applied, all of which have been used in previous trauma research.

RESULTS

Social factors in exile, particularly the level of "affective" social support, proved important in determining the severity of both post-traumatic stress disorder and depressive reactions, particularly when combined with a severe level of trauma/torture. Poor social support is a stronger predictor of depressive morbidity than trauma factors.

CONCLUSIONS

Some of the most important factors in producing psychological morbidity in refugees may be alleviated by planned, integrated rehabilitation programmes and attention to social support and family reunion.

Life events, social support and marital relationships in the outcome of severe depression

The effects of life events, social support and marital relationships on outcome were examined in a predominantly recurrent in-patient sample of depressives studied longitudinally every 3 months to remission and up to 15 months thereafter. Outcomes examined were length of time to remission, presence of residual symptoms at remission, and subsequent relapse. There were few associations between these outcomes and the social variables. These findings add to other recent evidence that psychosocial factors are relatively unimportant in the subsequent course of severe and recurrent depressions, in contrast to their contribution to onset of such depressions and subsequent outcome of milder depressions.

Organised violence and the stress of exile. Predictors of mental health in a community cohort of Vietnamese refugees three years after resettlement

BACKGROUND

The prevalence and course of mental disorders among Vietnamese refugees were studied, using a model including variables from different research traditions.

METHOD

A consecutive community cohort of 145 Vietnamese boat refugees aged 15 and above were personally interviewed on their arrival in Norway and three years later.

RESULTS

Three years later, there was, unexpectedly, no decline in self-rated psychological distress (SCL-90-R), almost one in four suffered from psychiatric disorder and the prevalence of depression was 17.7% (Present State Examination). Female gender, extreme traumatic stress in Vietnam, negative life events in Norway, lack of a close confidant and chronic family separation were identified as predictors of psychopathology.

CONCLUSIONS

The effects of war and persecution were long-lasting, and compounded by adversity factors in exile. A uniform course of improvement in mental health after resettlement cannot be expected in all contexts. The affected refugees need systematic rehabilitation.

Life events, social support and depression

This paper reviews current findings regarding social stress and support in clinical depression. Comparisons of recent life events at depressive onset and in general population controls show consistently raised event rates. The events span a range of threatening and undesirable experiences, with limited selectivity to exit events and interpersonal losses. Effects are similar in endogenous and non-endogenous symptom pictures, and there are suggestive findings in bipolar disorder, but these require further study. Events are also related to outcome and to relapse. Effects are moderate in degree, but relatively short-term of over six months to a year. For social support there are greater problems in the extent to which social support may be determined by the individual's own behaviour. Absence of social support appears to be associated with onset and relapse of depression, both acting independently and modifying effects of life events. Social stress findings have implications for prevention. The occurrence of major life events signals a period of increased risk when supportive interventions may prevent evolution of distress to disorder.

A cross-cultural comparative study of depressive symptoms in British and Turkish clinical samples

A cross-cultural comparison of the symptomatology of depression in Britain and Turkey was conducted in outpatient samples selected according to the same criteria, and matched for age, sex and severity; the same standardized rating instruments were also used with each sample. Comparison of the frequency and severity of individual symptoms showed higher mean ratings in the British sample for core depressive symptoms and higher ratings in the Turkish sample for symptoms reflecting somatization and a tendency to emphasize symptoms. Principal component analysis produced similar dimensions, confirmed after factor rotation, but the principal components emerged in a different order, reflecting differing contributions to the variance. In the Turkish sample, the first, general component, accounting for the greatest variance, was the somatization factor, while in the British sample it was the component that reflects core depressive symptoms. These findings indicate that there are some similarities in the symptoms of depression in the two cultures, but also differences in their predominant mode of expression.

Quality of life and antihypertensive therapy in men. A comparison of captopril with enalapril. The Quality-of-Life Hypertension Study Group

BACKGROUND

We conducted a multicenter trial comparing two angiotensin-converting-enzyme inhibitors to determine whether effects on quality of life during antihypertensive therapy are uniform within this pharmacologic class of agents, and to relate the effects of the drugs on quality of life to objective adverse events, such as the loss of a job or the death of a spouse.

METHODS

After a four-week washout period when they received placebo, 379 men with mild-to-moderately-severe hypertension were randomly assigned to receive captopril (25 to 50 mg twice daily, with or without hydrochlorothiazide) or enalapril (5 to 20 mg per day, with or without hydrochlorothiazide) for 24 weeks. Blood pressure, quality of life, and life events were monitored. Differences between treatments were evaluated by calibrating measures of quality of life with objective life events.

RESULTS

Throughout the treatment period, no differences were found in blood pressure, frequency of withdrawal of patients from the study, or major side effects. Patients treated with captopril had more favorable changes in overall quality of life, general perceived health, vitality, health status, sleep, and emotional control (P < 0.05 for each). The changes varied according to the quality of life at base line (P < 0.001); patients with a low quality of life at base line remained stable or improved with either drug, whereas those with a higher quality of life remained stable with captopril but worsened with enalapril. The quality-of-life scales correlated with life events and symptom distress (P < 0.001), and calibration analysis indicated that differences between treatments were clinically important.

CONCLUSIONS

Two angiotensin-converting-enzyme inhibitors, captopril and enalapril, indistinguishable according to clinical assessments of efficacy and safety, had different effects on quality of life. Calibration with life events showed that drug-induced changes are substantial and that the different effects of these two agents on quality of life can be clinically meaningful.

Life events and psychosis. Initial results from the Camberwell Collaborative Psychosis Study

Data from the Camberwell Collaborative Psychosis Study were used to examine the proposition that there is an excess of life events preceding the onset of psychoses of all types. Of 97 patients from the study who had episodes within the past year that were datable, 51 had developed psychotic symptoms from an essentially symptom-free state, 29 had been suffering only from neurotic symptoms, and 17 had experienced a marked exacerbation of psychotic symptoms. DSM-III diagnoses were collapsed into three major groups: 51 cases of schizophrenia; 31 cases of mania; and 14 cases of depressive psychosis. Life-event histories were taken for the six months before onset, and when these were compared with equivalent histories from a psychiatrically healthy sample from the local general population, there was a significant excess of life events, particularly in the three months before onset of psychosis. This was apparent in all groups, and remained even when events were restricted to the independent category. The excess of events began rather earlier than has been found in previous studies. In our view, this study provides some of the strongest evidence for a link between life events and the emergence of psychotic symptoms.

The similarities and dissimilarities between community and clinic cases of depression

It has been hypothesised that community cases of depression compared with clinic cases: (a) have fewer symptoms; (b) have milder symptoms; (c) have different symptoms; (d) are of shorter duration; (e) are less incapacitating; (f) are more environmentally and less biologically caused; and (g) respond better to psychological than pharmacological intervention. A review of the literature indicates that some of these hypotheses have not been tested, others have not been adequately tested, and none has been unequivocally confirmed. Further investigations of these important issues are required.

The prevalence of psychiatric morbidity among people aged 85 and over living at home. Associations with reported somatic symptoms and with consulting behaviour

A survey of all people aged 85 years and over, living in an inner London borough, was carried out. In 1987, 662 people who lived at home were traced from family practitioner committee records and interviewed. The General Health Questionnaire was administered to measure psychiatric morbidity (after excluding people with evidence of disorientation or confusional states). 27% of respondents were rated as having probable psychiatric morbidity (cases). These cases were more likely to report somatic health problems, particularly those associated with stress. There was no relationship with psychiatric morbidity and age, sex, social network type, or feelings of loneliness. However, fewer of those rated as cases, and who also independently reported problems with nerves, stress or depression, said they had reported these feelings to their general practitioners. Multiple regression analysis showed no significant relationship with General Health Questionnaire score and recency of contact with general practitioners.

The relationship between life stress and depression in an endogenous sample

The role of stress was examined in a sample of 68 endogenously depressed patients. Antecedent and concurrent life events and long-term difficulties, operationalized by different subcategories and scoring options, were related to depression severity, treatment response versus nonresponse, and maintenance versus relapse. Results showed an association between initial depression and various categories of subjective stress; a reduction in subjective (but not objective) event-related stress coincident with remission; positive relationships between life events and response and maintenance and between difficulties and relapse; and a higher event profile among unipolar depressive patients. Results are discussed in terms of the positive prognostic value of reactivity in endogenous depression, as well as the interactive relationship between continuing depression and long-term difficulties.

The Camberwell Collaborative Depression Study. I. Depressed probands: adversity and the form of depression

The Camberwell Collaborative Depression Study is an investigation of a series of 130 patients (76 female; 54 male) attending the Maudsley Hospital Services with unipolar depression of recent onset (the probands), and of their first-degree relatives. This paper describes the first element of the study, the investigation of the index cases or probands, which was carried out by members of the MRC Social Psychiatry Unit over the period 1982-1985. A description of the methods of the study is followed by an analysis of life events in relation to the symptomatic pattern of the depressive state. An 'endogenous' group was defined as cases of depression falling within Catego classes D and R, and compared with a 'neurotic' group conforming to classes N and A. The hypothesis that the 'endogenous' group of disorders would be relatively independent of prior life stress was not confirmed. Depressed women were more likely to have experienced life events or difficulties than their male counterparts, and there was some evidence that sex, but not age or social class, influenced the relationship between adversity and the type of depression. Examination of the timing of life events was strongly suggestive of a causal effect, with a pronounced rise in the month before onset. This was not limited to the most severe events. Differences between the 'endogenous' and 'neurotic' groups in the temporal patterning of events before onset are discussed. The findings are interpreted in terms of the literature on the topic.

Adversity and the symptoms of depression

Data from a community survey were used to test the proposition that pathological guilt and vegetative symptoms of depression were less likely to be associated with stressful life events and difficulties than non-specific symptoms like tension and worry. Two types of analysis were carried out: the first took only cases, defined as ID5+ on the Index of Definition program, the second was based on all subjects. Only the second analysis provided support for our predictions.

Depressive disorder in primary care

Since the pioneering study of psychiatric morbidity in primary care by Shepherd et al in 1966, it has become increasingly apparent that a substantial proportion (between 20% and 25%) of patients consulting their GP are suffering from some form of psychiatric disturbance (Goldberg & Blackwell, 1970; Hoeper et al, 1979). The composition of this psychiatric morbidity has been shown to be almost wholly affective in nature and largely mild in degree. In their important review Jenkins & Shepherd (1983) recently summarised the now extensive findings relating to overall minor psychiatric morbidity in primary care. However, recent collaborative studies between psychiatrists and GPs have identified that within this dilute pool of minor disorders, lurks a significant but poorly served population of patients suffering from depressive disorders which are by no means minor in degree. A number of crucial issues regarding this depression in primary care emerge which the present paper aims to review. In particular, how common is it, and how severe? How does it present and what, if any, are its special characteristics? What is the precise relationship between depressive symptoms and depressive illness presenting to the GP and what is the relationship between physical illness and depression? And finally, what is the course and outcome of depression in this setting and what are the indications for and effect of treatment?

Depression: distress or disease? Some epidemiological considerations

Surveys using clinical-type interviews have documented a high rate of depression among working-class women, and this is discussed in the light of a recent survey in an inner-city area. While women with caseness of depression contacting a psychiatrist did not differ in number of core depressive symptoms from those who did, they did in certain characteristics that would make them worrying for a general practitioner to deal with. It is concluded that there is a considerable overlap in the severity of depressive conditions between those seen by psychiatrists and those defined as cases in population surveys; any differences that do exist may relate more to the way symptoms are expressed than to the severity of the depressive disorder as such.

Life events in depression. Relationship to subtypes

Life events that had occurred in the 6 months before the onset of depression were recorded in 40 depressed patients and 41 normal controls. The depressed patients had experienced significantly more life events and significantly more undesirable life events than the controls. The 20 patients with a DSM-III diagnosed major depressive episode (MDE) without melancholia had experienced significantly more life events in the 6 months before the onset of depression than the 20 patients with a major depressive episode with melancholia. The patients with MDE without melancholia, but not the MDE with melancholia patients, had also experienced significantly more life events than a group of age- and sex-matched normal controls.

Depression in general practice: clinical features and comparison with out-patients

General practice (GP) depressives prescribed an antidepressant were compared with those given other treatment, and with antidepressant-treated psychiatric out-patient depressives. GP depressives were considerably less severely ill than out-patients, with fewer depressive symptoms and shorter illness, as well as less primary and less endogenous depression. The two groups of GP depressives differed less, but those receiving other treatment tended to have less severe depression than those receiving antidepressants and were less likely to satisfy diagnostic criteria for depression. Depressives in GP differ considerably in clinical characteristics from psychiatric out-patient depressives, and clinical features influence the GP's decision to treat with antidepressants.