Psychiatric disorder in London and North Uist

Mental health and female sterilization

In order to evaluate the mental health aspects of voluntary sterilization for family planning purposes, WHO initiated in 1978 a multi-centre prospective controlled study. A total of 926 healthy women with at least two living children, who requested sterilization (interval, post-partum or post-abortion), and 924 controls using other methods of contraception, were selected for a prospective investigation which included pre-operative and post-operative (6 weeks, 6 months, and 1 year) standardized assessment of mental state and a range of variables related to physical health and psychosexual functioning. Although pre-operatively a high proportion of subjects in both index and control groups reported minor non-specific psychiatric symptoms the percentage of women meeting defined criteria of psychiatric disorder was low to moderate (less than 15%) in all centres except Cali (26%). In all centres the pre-operative rates of psychiatric disorder were somewhat higher in the sterilization groups than in the control groups, and in two centres (Cali and Chandigarh) this difference was statistically significant. Thus, the data indicate that women choosing sterilization for birth control tend to show pre-operatively more psychiatric disturbance than their control counterparts who opt for other methods of contraception.

Depression and modernization: a cross-cultural study of women

BACKGROUND

This study concerns depression among women living in developing and developed parts of the world. Using a continuum of environments conceptualized as ranging from traditional to modern, the goal is to explore the hypothesis that the prevalence of depression will be higher among those living in modern compared to traditional societies. This issue is examined among samples from West Africa and North America.

METHODS

The subjects are 657 women who reside in four locations. An operational definition of modernization is used to place the locales as a continuum in the following order: rural Yorubaland in Nigeria, Yorubas living in urban Nigeria, rural Canada, and urban United States. Variables employed include education, religious orientation, and the role of women as mothers and workers. Depression is assessed using an algorithm based on generally acknowledged criteria. Multivariate logistic regression is used to generate point estimates and confidence intervals for prevalence odds ratios, to adjust for potential confounders, and to assess effect modification.

RESULTS

The prevalence of depression was lowest among rural Nigerians and highest among urban residents in the United States. The association of depression with the proposed continuum was strongest among women under the age of 45 who had living children (Odds Ratio: 2.1; 95% Confidence Interval: 1.6-2.9).

CONCLUSIONS

In the areas studied, a traditional way of life seems to offer protection against some of the stresses associated with modernization although it does not appear to compensate for the adversity of childlessness. Level of modernization may be a useful concept for understanding differences in rates of depression in different parts of the world.

Higher prevalence of mental disorders in socioeconomically deprived urban areas in The Netherlands: community or personal disadvantage?

OBJECTIVE

Major mental disorders occur more frequently in deprived urban areas. This study examines whether this occurs for all mental disorders, including less serious ones. It further assesses whether such a concentration can be explained by the socioeconomic status (SES) of the residents concerned or that a cumulation of problems in deprived areas reinforces their occurrence.

DESIGN

Mental disorders were assessed by means of the General Health Questionnaire (GHQ) among 4892 residents. Additional data were obtained on area deprivation, and on individual SES. Multilevel logistic regression models were used to take the hierarchical structure of the data into account, residents being nested in boroughs.

SETTING

General population of the city of Amsterdam, the Netherlands.

MAIN OUTCOME MEASURE

Prevalence of an increased (> or = 2) score on the GHQ, 12 item version.

RESULTS

Mental disorders occur more frequently in deprived areas but this can be explained by the lower SES of the residents concerned.

CONCLUSIONS

The cumulation of mental disorders in deprived urban areas is mainly a result of a concentration of low SES people in these areas. Contextual factors of deprived urban areas give hardly any additional risk above that resulting from a low individual SES.

Depression in female health visitor consulters: social and demographic facets

The primary health care setting has been established as a key venue for identifying and working with depression. Despite this, and the high risk of depression experienced by women in the postnatal period, maternal depression has been little examined in the work of health visitors. Furthermore, although research has been undertaken on social factors related to depression in the general population, there is nothing specifically on the population of health visitor clients. This study focuses on social and demographic factors related to maternal depression amongst users of health visitor services. In a study of 701 women in rural and urban areas, maternal depression was found to be significantly associated with the absence of receipt of further or higher education, housing status, employment status, reliance on state benefit, family size, family breakdown (reconstituted families and single divorced or separated women), perceptions of support and difficulty in getting children off to sleep, particularly after the first year of the child's life. Loglinear analysis was used to identify the most parsimonious model of these relationships, and three key areas emerged: social and economic disadvantage, family size and history of family disruption. The model suggested that 'behind' these data may have been poor life chances, as evidenced by data on further and higher education. These data identify elements of objective life circumstances important for depression in this group. They have considerable significance for the organization of health service resources and these are discussed.

Depression in the work of British health visitors: clinical facets

The primary health care setting has been established as a key venue for identifying and working with depression. Despite this, and the high risk of depression experienced by women in the post-natal period, maternal depression has been little examined in the work of health visitors. This study focuses on clinical facets of this work, including the rate and content of depression amongst health visitor attenders, the capacity of health visitors to identify accurately the presence of depression, the relationship between depression and child abuse and child behavioural issues, variations in the practice of health visitors and work with other professionals. In a cohort of 701, 11% of women were depressed, with distinguishing symptoms including fatigability, disgust/hate of herself and a sense of failure. Health visitors were not generally accurate in their identification of depression, were significantly more likely to see depressed women at home (than at clinic), but there was little difference in mean frequency of consultations according to whether or not the women were depressed. Urban health visitors had a higher mean frequency of consultations, but rural health visitors showed a rather greater tendency to increase frequency of consultation with the presence of depression. Child abuse concerns and behavioural problems were significantly associated with depression and these were three times as frequent amongst depressed women with no children aged under one compared with women in their post-natal year. Depressed women were far more likely to be involved with other agencies, but the GP was by far the most likely other professional to be involved. Health visitors are in a strategic position to help women with depression, and it is important that they are able to identify its presence. The association with child abuse is very important and indicates the need for multi-professional involvement, particularly with social workers, to a greater degree than was evident. The higher rate of depression in women beyond the post-natal year, was a surprising finding, and particularly in view of its relationship with child abuse concerns, suggests that this is a group of attenders who should be carefully monitored.

The homeless

The medical literature on homeless people tends to concentrate on their biographic characteristics or clinical problems without enough attention being given to the social and environmental contexts which they have to survive. This article summarises the literature on the role of social factors in the causation and outcome of health problems of the homeless and emphasises the importance of addressing the social context in effecting intervention. Services that deal with the social needs of the homeless will be more successful in meeting their needs and reducing their distress.

Psychiatric morbidity among repatriated Greek migrants in a rural area

This paper reports on psychiatric case identification by the application of the Structured Clinical Interview for DSM-III-R (SCID) in a sample of 198 Greek migrants repatriated from western europe in a northwestern province of Greece. The current (1 month) prevalence of psychiatric morbidity, based on the total number of diagnosed cases, was found to be 43.4%. Lifetime prevalence of psychiatric disorders was found to be higher (49.4%). The majority of the sample were diagnosed as suffering from anxiety disorders and dysthymia. Psychiatric disorders were found to be more prevalent among middle-aged respondents. Duration of stay in the foreign country was a factor correlated with psychiatric morbidity. Of short-term migrants 54% were found to suffer from specific nosological entities, whereas 32% of long-term migrants were diagnosed as cases. The results are discussed within the framework of the existing sociocultural context of emigration and repatriation.

The Dubai Community Psychiatric Survey: acculturation and the prevalence of psychiatric disorder

Dubai, an Emirate in the Gulf region, has experienced spectacular social change as a result of the exploitation of its oil reserves. The Dubai Community Psychiatric Survey was designed to study the effects of this social change on the mental health of female nationals. In this paper, we approach the problem by quantifying social change in two main ways: the first focused on social change at the individual level as measured by the Socio-cultural Change Questionnaire (Bebbington et al. 1993). The second examined the effect of social change at the community level by identifying areas of residence at different levels of development. We hypothesized that attitudes and behaviours markedly at odds with traditional prescriptions would be associated with high rates of psychiatric morbidity. On the individual level, the association between psychiatric morbidity and the amount of social change reflected in the behaviours and views of the subjects was not significant. However, there was a significant association between morbidity and between social attitudes and behaviours. At the community level, in contrast, the relationship between psychiatric morbidity and social change was significant: there was more psychiatric morbidity in areas at the extremes of the social change continuum. The hypothesis put forward in this study must be modified accordingly.

Rural-urban child psychopathology in a Northeastern U.S. state: 1986-1989

Parent and teacher symptom reports from two epidemiological surveys of 2,519 Connecticut children were used to study rural-urban differences in childhood psychopathology. Parents and teachers of girls in cities reported elevated total disturbance and social withdrawal. Parents of urban girls also reported higher rates of behavioral disturbance. For boys, urban excesses were primarily observed in emotional disturbance. Rural-urban variation was largely associated with economic and cultural differences between sites and not with urbanization per se. Findings suggest that certain assumptions about rural-urban differences in specific forms of psychopathology, such as delinquency, should be reevaluated.

Social support and its interactions with personality and childhood background as predictors of psychiatric symptoms in Scottish and American medical students

Two samples of first year medical students were studied, one in Houston, Texas, the other in Edinburgh, Scotland. Personality, measured on entry, and social support, measured prior to first year examinations, were used to predict scores at the latter time on anxiety self-ratings, depression self-ratings and the General Health Questionnaire. At both centres social support was implicated in symptomatology, but many significant effects were centre specific. Presence of a local friend in whom to confide, was important in Houston, while support from a relative and superficial support from classmates were salient in Edinburgh. In general, presence of support was associated with lower symptoms, but there were important exceptions similar in both centres. Students who did not have a boy/girl friend during the study period were less anxious than the rest. Regarding personality, students high on suspiciousness were more anxious and depressed when they had high levels of support. Among reserved (aloof and introspective) students both gain and loss of a boy/girl friend during the year were strongly associated with depressed mood. It is concluded that, particularly for students high in suspiciousness, reserve and shyness, the costs of social relationships sometimes outweigh the benefits.

Affective disorders among women in the general population and among those referred to psychiatrists. Clinical features and demographic correlates

In a study comparing depressive disorders detected in a field survey (n = 90) with patients referred to a specialist treatment setting (n = 63), the clinical features and demographic correlates of 'cases' of affective disorders proved to be similar. However, those in treatment settings appeared to have more people achieving definite case status. Hospital-referred cases were also more likely than community cases to be older and single, and this difference persisted even after controlling for chronicity of symptoms.

The influence of social factors on common mental disorders. Destabilisation and restitution

This study distinguishes between processes that cause individuals to experience symptoms--destabilisation--and those that are associated with loss of symptoms over time--restitution. It is shown that different clinical, social, and personality variables are associated with each of these processes. Where destabilisation is concerned, it is shown that different variables were associated with the development of symptoms of anxiety and those of depression. Different variables were associated with restitution, and they did not show the same relationship with the symptom dimensions of anxiety and depression as those which were associated with destabilisation.

Marriage, motherhood and psychiatric morbidity in New Zealand

The prevalence of psychiatric disorder, as assessed by both the GHQ-28 and the short PSE, is described for a random community sample of New Zealand women. In contrast to previous studies, married and widowed women and mothers showed lower rates than the never married and childless women. A plausible explanation is provided by available analysis of New Zealand gender roles. Such an explanation would reconfirm the importance of socio-cultural factors in community psychiatric disorder.

Psychiatric epidemiology

In this paper, some of the basic concepts of epidemiology are presented. Their applicability to psychiatry is outlined, and the definitions of terms are given. Various methods of epidemiological investigation are discussed. The background for the development of recent prevalence studies, including advances in diagnostic methods and case ascertainment, are given in some detail, and some recent results are presented. Additional areas of investigation are discussed more briefly.

Period prevalence of psychiatric disorders in Edmonton

3,258 randomly selected adult household residents of Edmonton were interviewed by trained lay interviewers using the Diagnostic Interview Schedule (DIS). Six-month prevalence figures for DIS/DSM III diagnoses are given, and selected figures for the one-month and one-year prevalence, and the one-year symptom-free rates. The six-month prevalence for any diagnosis is 17.1%, comparable to findings from other population studies using DSM III derived diagnoses, but lower than studies using the PSE. The prevalence rates for most disorders tended to be lowest in the elderly, but this was not as marked as the drop in lifetime prevalence. Men had higher prevalence for substance use disorders than women, but women had higher rates for affective disorders and anxiety/somatoform disorders. Prevalences for all disorders were either similar or lower in those who were married or living as though married, than in those who were not cohabiting. One-year symptom-free rates were highest for those with substance use disorders and lowest for those with anxiety/somatoform disorders - largely due to the persistence of phobias.

Investigations of the prevalence of psychiatric disorders

More than thirty studies of the prevalence of psychiatric disorders have been reported from Western countries in the last thirty years. Significant methodological advances have occurred, including sophistication of sampling techniques, improvements in caseness definitions and identification of cases, and clear definitions and specification of rules for diagnosis. The introduction of standardized diagnostic instruments for use by trained lay interviewers is expected to facilitate studies of psychiatric epidemiology. The methods used in each of the studies are briefly described.

The epidemiology of anxiety disorders: rates, risks and familial patterns

This paper reviews what we know about the epidemiology and familial patterns of anxiety disorders. Focus is on the current studies based on specified diagnostic criteria. Data are presented, when available, on the subclassifications of the anxiety disorders. Data from epidemiologic and family studies support the notion that anxiety disorders have a relatively high prevalence and are familial, that they are heterogeneous, and that some are related to depression. It suggests that there is an increased probability that a person with one anxiety disorder will have another or will have a major depression during his or her lifetime. Data also suggest that panic disorder has the most severe consequence in terms of morbid risk to first-degree relatives, particularly risk to children, and that there may be a relationship between adult and childhood anxiety disorders. Potential research areas are given.

The buffer theory of social support--a review of the literature

The buffer theory postulates that social support moderates the power of psychosocial adversity to precipitate episodes of illness. In this paper, we review the theory as applied to minor affective disturbances. Research in this area suffers because of the many disparate conceptualizations of social support and the resulting difficulty of deciding on the content of measures. Moreover, the meaning of the term buffering is itself unclear. These problems have not, however, inhibited research, and many cross-sectional and longitudinal studies have now been carried out. Our review leads to the conclusion that evidence for a buffering role of social support is inconsistent, reflecting methodological differences between studies but probably also indicating that buffering effects are not of dramatic proportions. Moreover, it is possible that the observed relationships are the result of spurious association or contamination of measures.

The Zurich Study. V. Anxiety and phobia in young adults

From data collected within the frame of a longitudinal epidemiologic study of a representative sample population of young adults (the Zurich Study), anxiety disorders--anxiety and phobic states--were analyzed. The current prevalence rate was found to be 2.9% for anxiety states and 4.3% for phobic states, totaling 7.2%. Because of their anxiety disorders, one-fifth of the cases had undergone treatment during the preceding 12 months. Substantial difficulties arose, from the point of view of classification. The currently used categories, such as anxiety states, panic attacks, agoraphobia, simple phobia, social phobia, have more in common than differing symptoms. On a syndromal level, numerous overlapping configurations resulted which can only be artificially forced into the aforementioned diagnostic classes of anxiety disorders. On the symptom level, merely a few operationalized items defined these categories. In this way, most of a broad identical 'nonspecific' symptomatology was not taken into account. This was exemplified by a comparison of anxiety states and agoraphobia. In fact, these two groups did not differ significantly in many symptoms or in SCL-90 profiles, but did show a highly significant difference from control samples. Both groups suffered to the same great degree from depressive, cardiovascular, and gastrointestinal symptoms. We doubt whether any forced categorical diagnostic differentiation of anxiety and phobic states is at all reasonable.

Women and health in Europe: the scope and limits of epidemiology

Epidemiology is one tool for assessing women's health, at least in terms of the main diseases which cause morbidity and mortality. When employed for international comparisons across Europe, its value, even at this level, is diminished by the shortage of reliable data. Examples are chosen from the fields of cancer and cardiovascular disease. The latter part of the article draws attention to the inadequacies and limits of clinical epidemiology when applied to women's broader experience in respect to health.

Prevalence of psychological disorders in the Athens area. Prediction of causal factors

A two-phase cross-sectional study on the prevalence of psychological disorders was carried out in a sample of 1,574 adult male and female residents of two boroughs in the greater Athens area. The analysis focused on the detection of any possible effects (risk factors) that personal data (such as sociodemographic profile, use of psychiatric services, medication and experience of stressful life events) may have on mental health; information about the latter was assessed by Langner's scale. The findings are that all the independent variables investigated affect the respondents' psychological status, although interactive effects between variables predicted different patterns of psychopathological symptom formation.

Developmental differentiation of anxiety and affective neuroses

One hundred and fifty anxiety neurotic and neurotic depressive twin probands were differentiated into three groups by means of discriminant analysis, 50 in each group. The groups were named pure anxiety neurosis, mixed anxiety-depression and pure neurotic depression. Childhood environmental factors seemed important in the development of pure neurotic depression, while mixed anxiety-depression seemed to be determined by environmental factors in adult life. Further investigation showed that loss of a parent before the age of 16, and moving before the age of 14 seemed important in the development of pure neurotic depression. Loss, or threat of loss, of love objects appeared to be the most common precipitating event for the neurotic depression and mixed anxiety-depression groups, while pregnancy and childbirth seemed to be of great significance in the development of pure anxiety neuroses. Furthermore, unmarried status was most frequent in the mixed anxiety-depressive group of probands.

Psychiatric disorder in a rural and an urban population: 3. Social integration and the morphology of affective disorder

This paper focuses on the morphology of affective disorder. We set out to confirm on the island of Lewis a cross-over result originally found on the island of North Uist, whereby in those sections of the population where the rates of depression were lowest the rates of anxiety were highest. These sections of the population varied in the degree to which their members were integrated into the traditional way of life. 'Integration' was characterized in terms of churchgoing and crofting (the word for farming in the Hebrides). It is predicted here that the churchgoing rather than the crofting component of the integration measure would account for any cross-over result found concerning the morphology of symptoms. This hypothesis, derived from a theoretical concern with links between repression and anxiety, is not confirmed. Instead, crofting is found to be the component of the integration index which predicts anxiety/phobia without depression. In seeking to explain this unexpected finding further significant associations between crofting, pure anxiety/phobia and the style of personal relationships are revealed. It is also confirmed that types of severe events among women are also found to vary with the degree of integration into traditional society. A perspective is then developed which might explain how the interaction between style of personal attachments and types of provoking event could produce variations in the morphology of affective disorder.

The Zurich study--a prospective epidemiological study of depressive, neurotic and psychosomatic syndromes. I. Problem, methodology

The purpose and methodology of a 4-year longitudinal study based on a cohort aged 20 years are presented. A two-stage procedure was chosen; in 1978, 2201 males and 2346 females, aged 19-20, were examined. This sample was representative of the respective age group in the Canton of Zurich. From high and low-scorers (SCL-90), 292 males and 299 females were randomly selected for interview and for a prospective study. Subsequent investigations were carried out by questionnaires and by a personal interview. The instruments chiefly consisted of a semi-structured interview (SPIKE), a clinical syndrome list (SL), a 90-item symptom check list (SCL-90R), a life-event-inventory, scales measuring coping behavior and dissimulation, and an extensive sociological interview dealing with sociodemographic characteristics and social adjustment. This paper gives an account of the methodological aspects of the study.

Social support and mental health: direct, protective and compensatory effects

This study deals with the relationship between social support and mental health. This relationship was conceptualized as a multi-faceted one which may be understood in terms of direct (main), protective (buffering) and compensatory effects. Hypotheses relating to these effects were evaluated in terms of spousal/community support and job strains (pressure, autonomy, opportunities) among a sample of 455 married males who were employed on a full-time basis. Results indicated that spousal support generally had more important implications for psychological well-being than did community support. However, the relative impact of these different sources of support appeared to be related to the degree to which the mental health symptomology reported was affectively defined. With respect to protective effects, spousal support was found to moderate the relationship between certain job strains (pressure) and mental health. No protective effects were found for community support. Tests of the hypothesis that community support compensates for low levels of spousal support were made in terms of both direct and protective effects. No direct effect compensation was observed. However, protective effect compensation was observed for certain job strains (opportunities) and was relevant to a wide range of symptomology. The overall results suggest a primary/secondary hierarchy of supports and strains in which primary supports (e.g. spousal) assume a protective function with respect to secondary strains (e.g. job) while secondary supports (e.g. community) do not protect individuals from primary strains (e.g. spousal). Secondary supports have a protective function for secondary strains in the absence of primary supports. It is further suggested that sex roles may be important determinants of the availability and use of different support sources.

Effects of social support in the stressor-strain relationship: a Dutch sample

Social support may be defined as the social ties an individual has, and which lead the subject to believe that he is cared for. The present study describes research on the hypothesis that social support protects the person against the noxious effects of job stress and will keep the strain and health effects low. A short review of relevant literature is given and 6 hypotheses about the role of support are tested; 3 about the interactive effects and 3 about the main effects. The study resembles the work of LaRocco, House and French, although there are some differences. Like these authors, a moderated regression technique was used on data from a sample of 1246 employees of 13 different industrial organizations. The analysis shows that social support buffers the impact of work-related stressors on psychological and behavioral strains but that there is no such a buffering effect on health strains. A very striking result was the strong buffering effect against employee's high blood pressure (SBP and DBP), of the supervisor giving support in the case of psychological problems.

Epidemiology of mental disorders in Camberwell

A two-stage psychiatric survey of a random samples of adults aged 18-64 from Camberwell is described. Agency interviewers carried out the first stage (N = 800), using the shorter form of the Present State Examination (PSE). MRC interviewers, using the full PSE, saw a stratified sample of these (N = 310) in the second stage. A second interview was sought with all those of Index of Definition (ID) level 5 and above at the first interview ("cases") and with random sample of those below that level 20.9% refused or were never available for the first interview. Of the 800 subjects successfully interviewed, 10% refused a further interview and 124% of those finally selected for this interview were either unavailable or changed their minds. The MRC data, weighted to represent the whole sample, are used in our analyses. The prevalence of psychiatric disorder as defined in our study was calculated at 6.1% for men and 14.9% for women. Women showed a higher prevalence of disorder in the age-groups 25-34 and 45-54, but in men there was no significant association with age. In contrast to the findings of Brown & Harris (1978), social class did not have a strong association with disorder. Single men had much higher rates than married men, while the reverse was true in women. In both sexes employment was associated with lower rates of disorder. An attempt to explain the high prevalence in women in terms of their role in marriage and child-care was only partly successful.

Psychiatric disorder in a rural and an urban population: 1. Aetiology of depression

An earlier survey of a random sample of women in Camberwell in South London has shown that the majority of new occurrences of depressive disorders were brought about by certain kinds of life event and ongoing difficulty (provoking agent) and that the risk was increased under these circumstances by the presence of certain other social factors (vulnerability factors). Working- class women wer much more likely to develop depression because they experienced more of these factors. A new survey in a rural population in the Outer Hebrides has confirmed the importance of these factors in the genesis of depression, although provoking agents occurred much less frequently in this rural setting. However, integration into the traditional way of life, rather than a middle-class status, was related to a lower chance of developing depression, and this appears to be explained by the relationship of provoking agents and vulnerability factors to such integration in the Outer Hebrides, and to social class status in Camberwell.

Depression and anxiety in the community: replicating the diagnosis of a case

The diagnoses were reviewed of 866 women interviewed in 5 community surveys carried out by the Bedford College team. The diagnoses matched commonly held conceptions of who is a case in terms of both severity and type of syndrome. Two methods of replicating the diagnoses are prevented, one based on discriminant function analysis and the other based on a simple checklist.

What makes persons "patients" and patients "well?

The transitions from person to patient and back represent social decision points rather than boundaries determined by shifting biological equilibria, as doctors so often suppose. Moreover, host resistance to pathogenic agents is weakened by social stress and strengthened by social support. Thus, the efficiency and effectivenss of medical care will be improved if the clinician, in assessing patient problems, systematically inquires into the social determinants of the decision to seek help. The probability of resolving the patient's difficulties will be enhanced by targeting treatment measures at the social components of the illness experience as well as at the pathophysiology of the disease process.

Prevalence rates for the neuroses: pitfalls in the evaluation of familiality

A survey was made of studies reporting population prevalence rates for the different types of neurotic disorders in order to determine the utility of these rates for familial and genetic research. Rates varied considerably among studies both for all neuroses and for particular types of neuroses, even among studies using similar methods examining similar populations. The method of case ascertainment (personal interview versus record searches) and the threshold for defining an individual as affected are important contributors to variation. The use of the published prevalence rates for quantifying familial and genetic effects on the neurotic disorders would be premature.

Notions of neurosis: a comparison of community cases with declared patients

Seven ways are discussed in which the concept of neurosis, as measured in community surveys, differs from the concept of neurosis among declared patients. These are different ratio of neurotic symptoms to dysthymic states (using Foulds' terms), the different ratio of mixed anxiety/depression to anxiety or depression alone, the duration, seriousness, and intensity of symptoms, and the two states of social dysfunction without symptoms, and symptoms without dysfunction. Attention is drawn to the need for epidemiologists and clinicians to describe neurosis on axes other than the presence or absence of symptoms alone.

The epidemiology of depressive disorder

The concept of depression is reviewed. Depression is defined as a syndrome and thus involves particularly subjective decisions by the diagnostician. The conceptual difficulties which arise from this are outlined. The rate of risk is reviewed, using both studies of treated cases and community interview surveys. The study of treated rates tells more of the variables affecting the process of declaration than of depression. It is proposed that case-finding has the prerequisite of precise case-definition with a standardized instrument. Only then can comparability be reached. The effect sociodemographic variables on rates of risk is analysed: it is concluded that they do not account for much of the variance because they are at best gross measures. The role of life events is assessed and it is argued that only by the study of their possible interactions with other factors will an adequate causal explanation of depression be achieved. The usefulness of this as a validation of the concept of depression is emphasized. In this paper we shall firstly look at how concepts of depression affect the process of arriving at a diagnosis and secondly at policies of case finding. We are then in a position to survey critically the findings of epidemiology in the study of depression. Finally, we will review work which throws light on the social causation of depression with the emphasis on recent developments.

The social causation of depression: a critique of the work of Brown and his colleagues

Brown and his colleagues have presented evidence for the social origins of depression from results of 2 community surveys in Camberwell. This paper examines critically their methodology, conceptualization of variables, mode of argument and statistical techniques. Log-linear analysis of their published data fails to substantiate their differentiation of risk variables into vulnerability factors and provoking agents.

The concept of a 'case' in psychiatric population surveys

A series of in-patients, a series of out-patients and a series of women from a general population sample, all aged under 65, and all living in the same area of south-east London, were examined using the ninth edition of the Present State Examination (PSE). An 'index of definition' (ID), based on number, type and severity of PSE symptoms, was constructed in order to define a threshold point above which sufficient information was available to allow classification into one of the functional psychoses or neuroses. The index specified 8 levels of definition of disorder; the threshold point for a 'case' lying between levels 4 and 5. The identification of 'cases' by this means in the general population series was highly concordant with global clinical judgement. When the 3 series were compared, the in-patient series contained the most definite cases, the outpatient series was intermediate, and the general population series was characterized mainly by below-threshold conditions. Disorders above the threshold point were classified using the CATEGO program. The clinical severity of the depressive disorders and anxiety states identified in this way in the 3 series was compared on the basis of symptom and syndrome profiles and PSE scores. As expected, the inpatients had the most severe, and the general population 'cases' the least severe disorders. It is suggested that the PSE-ID-CATEGO techniques can be used to help standardize certain aspects of case-finding and case-description in both referred and non-referred populations and thereby make the results of surveys more comparable.