Background. To examine relationships between recent DSM-III-R psychiatric disorders and work impairment in major occupational groups in the US labour force.

Method. Data are from the US National Comorbidity Survey (NCS), a survey of respondents ages 15–54 in the US. Employed people are the focus of the report.

Results. There is substantial variation across occupations in the 30-day prevalences of NCS/DSM-III-R psychiatric disorders, with an average prevalence of 18·2% (range: 11·0–29·6%) for any disorder. The average prevalences of psychiatric work loss days (6 days per month per 100 workers) and work cutback days (31 days per month per 100 workers), in comparison, do not differ significantly across occupations. Work impairment is more strongly concentrated among the 3·7% of the workforce with co-morbid psychiatric disorders (49 work loss days and 346 work cutback days per month per 100 workers) than the 14·5% with pure disorders (11 work loss days and 66 work cutback days per month per 100 workers) or the 81·8% with no disorder (2 work loss days and 11 work cutback days per month per 100 workers). The effects of psychiatric disorders on work loss are similar across all occupations, while effects on work cutback are greater among professional workers than those in other occupations.

Conclusion. The results reported here suggest that work impairment is one of the adverse consequences of psychiatric disorders. The current policy debate concerning insurance coverage for mental disorders needs to take these consequences into consideration.

Background. This study investigated factors contributing to suspected offending behaviour by adults with a history of learning disabilities taken into custody at a city police station.

Methods. Adults charged with offences, and/or leaving custody, during a defined period, were identified as having a possible learning disability using a four-item questionnaire (Lyall et al. 1995a). A comparison group, of similar age, sex and IQ, was identified from a database of young people with learning difficulties. Information was obtained on interview about each individual's medical, psychiatric, social and family histories and psychological assessments were undertaken.

Results. In contrast to the comparison group, those in the ‘offending’ group were more likely to have a history of the following: losing contact with their father, forensic contact in one or more family members, past homelessness, illicit drug use, experiencing an excess of recent life events, self-reported behavioural problems at school, truancy, childhood police contact and contact with probation services. All had histories of repeated offending. There was also an increased rate of drug/alcohol dependence. Only two subjects in the study group had a full-scale IQ below 70.

Conclusions. These differences would suggest that the presence of childhood behavioural problems, offending behaviour by other family members, family separation and other social disruption and the development of drug and alcohol related problems are potentially the most important factors in trying to understand why one group engaged in criminal behaviour. The offending group had many characteristics in common with general offending populations.

Background. A number of recent epidemiological studies have shown that the prevalence and incidence of dementia are increased in population strata with low compared to high levels of education. This has been explained as a consequence of a greater ‘brain reserve capacity’ in people with a high level of education. Theoretically, however, brain reserve capacity is better reflected by intelligence than by level of education. Thus, the emergence of dementia will be better predicted by low pre-morbid intelligence than by low education.

Methods. This prediction was tested in a population based sample of elderly subjects (N=2063; age range 65–84; Amsterdam Study of the Elderly) who were followed over 4 years. Dementia was diagnosed using the Geriatric Mental State examination (GMS). Pre-morbid intelligence was measured using the Dutch Adult Reading Test (DART), a short reading test which gives a good estimate of verbal intelligence, and is relatively insensitive to brain dysfunction. The effects of age, gender, occupational level, number of diseases affecting the central nervous system and family history of dementia or extreme forgetfulness were also examined.

Results. Logistic regression analysis showed that low DART-IQ predicted incident dementia better than low level of education. A high occupational level (having been in charge of subordinates) had a protective effect.

Conclusions. This result supports the brain reserve theory. It also indicates that low pre-morbid intelligence is an important risk factor for cognitive decline and dementia. Use of reading ability tests is to be preferred over years of education as estimator of pre-morbid cognitive level in (epidemiological) dementia research.

Background. Patients suffering from bipolar affective disorders are generally reported to be able to detect prodromes. Insight is also said to be desirable for a good outcome. However, very little is known about the effect of insight and patients' spontaneous strategies for coping with prodromes on their social functioning.

Method. In a cross-sectional study 40 bipolar patients, who were not in an acute episode, were interviewed about their prodromes of depression and mania, their coping strategies for these prodromes, their levels of insight and their levels of social functioning.

Results. A quarter of subjects reported that they could not detect any early warnings of depression compared with only 7·5% of subjects who reported that they could not detect prodromes of mania. Subjects reported both spontaneous cognitive and behavioural strategies for coping with prodromes of depression but only behavioural strategies for prodromes of mania. Subjects' current levels of depression, how they coped with prodromes of mania and their ability to recognize early warnings for depression contributed significantly to their level of social functioning. Insight also had a weaker but significant contribution.

Conclusion. No causal link was made in this study. However, it did show that patients' level of social functioning was related to their level of insight, and to how well they coped with the prodromes of mania and whether they could detect prodromes of depression. The results suggest that it is worth exploring ways of teaching patients to monitor their moods and to promote insight and good strategies for coping with their prodromes.

This paper attempts to construct a simplified system for the classification of personality disorders, and relates this system to normally distributed human personality characteristics. One hundred and forty-eight subjects with a variety of psychiatric diagnoses were evaluated using the SCID-II structured clinical interview for personality disorders. A four-factor solution of personality disorder symptoms was obtained and we labelled these factors ‘the four As’ : antisocial, asocial, asthenic and anankastic. The factors related to the four temperament dimensions of the Tridimensional Personality Questionnaire (TPQ), but less closely to Eysenck Personality Questionnaire (EPQ) dimensions. The four factors were similar to those identified in a number of studies using a variety of assessment methods and this lends some credibility to our findings. It suggests that a more parsimonious set of trait descriptors could be used to provide simpler, less overlapping categories that retain links with current clinical practice. In addition, these factors can be seen as extremes of normally distributed behaviours obtained using the TPQ questionnaire.

Background. Recent observations indicate that a certain pre-morbid personality profile (‘autonomic lability’, i.e. elevated neuroticism, frequent somatic complaints and increased inter-personal sensitivity) appears to be a valid antecedent of major depression. However, most of these prospective studies used samples drawn from the general population, which limits the power of any observed differences between subjects who developed a depressive disorder during the follow-up period and those who did not.

Methods. We investigated the psychometric profile of 54 high-risk probands (aged between 18 years and 45 years) without a current or lifetime diagnosis of any psychiatric disorder, but who had first-degree relatives with an affective disorder according to DSM-III-R criteria. Twenty-two control probands, matched for age and gender and without any personal or family history of psychiatric disorders, served as the reference group.

Results. As a group, the high-risk probands scored higher than the controls on scales that assessed neuroticism, rigidity, depressive cognitions, vegetative lability and stress. With an individual-orientated approach (cluster analysis), 30 high-risk probands were identified as conspicuous, characterized by elevated rigidity and increased ‘autonomic lability’. The remaining 24 high-risk probands showed a psychometric profile very similar to that of the controls.

Conclusions. The present findings in 54 probands at high risk for affective disorders not only strongly underline the assumption that the personality trait ‘autonomic lability’ is a valid antecedent of at least major depression, but also add the personality trait ‘rigidity’ as a further and potential candidate for a true vulnerability marker for affective disorders.

In order to examine factors that influence the time to recovery (TTR) from depressive episodes in women, we examined members of 1030 female–female twin pairs of known zygosity, ascertained from a population-based twin registry. We predicted, in a Cox model, TTR in 235 women with an onset of an episode of major depression (MD) in the last year meeting DSM-III-R criteria. The median and mean TTR for episodes of MD was 42 and 82 days, respectively ; only 2·2% of women had not recovered by 1 year. Four variables predicted TTR: financial difficulties, obsessive–compulsive symptoms, severe life events (SLEs), and genetic risk. Dividing all depressive episodes meeting symptomatic DSM-III-R criteria into early (5–28 days) and late (>28 days) phases, significant predictors of TTR early in the course of illness (income, parental protectiveness and separation, personality, lifetime traumas and SLEs) differed from those that predicted TTR later in the depressive episode (health, social support, obsessive–compulsive symptoms, SLEs and genetic risk). Including cases with chronic MD increased the strength of personality, financial problems and genetic risk as predictors of slow TTR. These exploratory analyses suggest that TTR from MD in women is influenced by multiple environmental, temperamental and genetic factors. Predictors of TTR early and later in the course of MD may differ qualitatively, suggesting different processes in recovery from brief versus prolonged depressions.

Background. Research on the dimensional structure and reliability of the Hospital Anxiety and Depression Scale (HADS) and its relationship with age is scarce. Moreover, its efficacy in determining the presence of depression in different patient groups has been questioned.

Methods. Psychometric properties of the HADS were assessed in six different groups of Dutch subjects (N = 6165): (1) a random sample of younger adults (age 18–65 years) (N = 199); (2) a random sample of elderly subjects of 57 to 65 years of age (N = 1901); (3) a random sample of elderly subjects of 66 years or older (N = 3293); (4) a sample of consecutive general practice patients (N = 112); (5) a sample of consecutive general medical out-patients with unexplained somatic symptoms (N = 169); and (6) a sample of consecutive psychiatric out-patients (N = 491).

Results. Evidence for a two-factor solution corresponding to the original two subscales of the HADS was found, although anxiety and depression subscales were strongly correlated. Homogeneity and test–retest reliability of the total scale and the subscales were good. The dimensional structure and reliability of the HADS was stable across medical settings and age groups. The correlations between HADS scores and age were small. The total HADS scale showed a better balance between sensitivity and positive predictive value (PPV) in identifying cases of psychiatric disorder as defined by the Present State Examination than the depression subscale in identifying cases of unipolar depression as defined by ICD-8.

Conclusions. The moderate PPV suggests that the HADS is best used as a screening questionnaire and not as a ‘case-identifier’ for psychiatric disorder or depression.

Background. A cohort study of a socially homogeneous group of teachers was commenced in 1978 to pursue possible risk factors contributing to the recognized female preponderance of depression.

Methods. Multiple measures of depressive experience included: (i) lifetime rates, duration and number of depressive episodes using two caseness definitions, DSM-III-R major depression and ‘all depression’ (which included a category of minor depression); (ii) self-report measures of state and trait depression, neuroticism, and self-esteem. DSM-III-R anxiety disorder rates are also reported and co-morbidity with major depression examined.

Results. At the 15-year review in 1993, the sample had a mean age of 39 years, there was a trend for a female preponderance in lifetime rates of major depression and ‘all depression’ (and which was more pronounced with the inclusion of data for anxiety disorders), with statistically significant differences in rates of social and simple phobias and combined anxiety disorders. Mean neuroticism scores were consistently higher for women.

Conclusions. The strong association between anxiety and depressive disorders suggests that greater reporting of anxiety and higher neuroticism scores in women may be a key determinant that contributes to any female preponderance in depression rates.

Background. A preliminary report by the authors suggested that the range of affect generated by voices (anger, fear, elation) was linked not to the form, content or topography of voice activity, but to the beliefs patients held about them, in particular their supposed power and authority. We argued that this conformed to a cognitive model; that is, voice beliefs represent an attempt to understand the experience of voices, and cannot be understood by reference to the form/content of voices alone. This study puts this cognitive model to empirical test.

Methods. Sixty-two voice hearers conforming to ICD-10 schizophrenia or schizoaffective diagnoses were interviewed and completed standardized measures of voice activity; beliefs about voices and supporting evidence, coping behaviour; affect and depression.

Results. Beliefs about the power and meaning of voices showed a close relationship with coping behaviour and affect (malevolent voices were associated with fear and anger and were resisted; benevolent voices were associated with positive effect and were engaged) and accounted for the high rate of depression in the sample (53%). Measures of voice form and topography did not show any link with behaviour or affect and in only one-quarter of cases did neutral observers rate voice beliefs as ‘following directly’ from voice content.

Conclusion. The study found support for our cognitive model and therapeutic approach. Factors governing the genesis of these key beliefs remain unknown. A number of hypotheses are discussed, which centre around the possibility that voice beliefs develop as part of an adaptive process to the experience of voices, and are underpinned by core beliefs about the individuals self-worth and interpersonal schemata.

Background. For many years, it has been known that dementia can occur in people with learning disabilities, but there have been few research studies. Studies that do quote rates for dementia show these to be high, but this important fact has received remarkably little attention.

Method. Comprehensive psychiatric and medical assessments were undertaken on the whole population (ascertained as far as is possible) of people with learning disabilities aged 65 years and over living in Leicestershire, UK (N=134), in order to ascertain rates of DCR defined dementia, and associated factors.

Results. Dementia was diagnosed in 21·6%, against an expected prevalence of 5·7%, for a group with this age structure. The rate of dementia increased in successive age cohorts: 15·6% aged 65–74 years; 23·5% aged 65–84 years; and 70·0% aged 85–94 years. People with dementia tended to be older, female, with more poorly controlled epilepsy, a larger number of additional physical disorders, less likely to be smokers and had lower adaptive behaviour scores than did the elderly people without dementia. They were more likely to live in health service accommodation.

Conclusions. Dementia occurs at a much higher rate among elderly people with learning disabilities than it does among the general population; this is independent of the association between dementia and Down's syndrome. Whether this relates aetiologically to genetics, lack of brain ‘reserve’ or history of brain damage is yet to be determined.

Background. Survey data are presented on the associations between retrospectively reported childhood adversities and subsequent onset and persistence of DSM-III-R disorders.

Methods. Data come from the US National Comorbidity Survey, a large survey of the US household population.

Results. Twenty-six adversities were considered, including loss events (e.g. parental divorce), parental psychopathologies (e.g. maternal depression), interpersonal traumas (e.g. rape) and other adversities (e.g. natural disaster). These adversities were consistently associated with onset, but not persistence, of DSM-III-R mood disorders, anxiety disorders, addictive disorders and acting out disorders. Most bivariate associations with onset attenuated in models that controlled for clustering of adversities and for lifetime co-morbidities among psychiatric disorders. Multivariate effects of adversities in logistic models were additive, which means that they have multiplicative effects on probability of disorder onset. Adversities showed little specificity. An analysis of time decay showed that the effects of childhood adversities on disorder onset persist beyond childhood.

Conclusions. The existence of strong clustering among childhood adversities and lifetime co-morbidity among adult disorders means that caution is needed in interpreting the results of previous single-adversity single-disorder studies as documenting unique effects of specific childhood adversities on specific adult disorders. Future studies need to assess a broader range of adversities and disorders and to explore the existence and effects of commonly occurring adversity clusters. Replication is needed to verify that the effects of childhood adversities are mostly on first onset rather than on the creation of vulnerabilities that lead to increased risk of persistence.

Background. Assess the latent structure of the DSM-III somatization symptoms and the stability of symptom patterns over time.

Methods. Cross-sectional and longitudinal covariation of symptoms of somatization disorder were investigated using structural equation models in a population-based data set from the Epidemiologic Catchment Area study.

Results. Medically unexplained physical complaints were discovered to cluster into three separate factors, consistent with the DSM-IV definition of somatization disorder, but one dominant general factor was defined, consistent with the DSM-III conceptualization. Individual symptom prevalences and factor structures were different for men and women. The factor structures remained stable at 1 year follow-up. Variations in the threshold of number of somatization symptoms required for diagnosis affected prevalence, sex ratio and temporal stability of the diagnosis.

Conclusions. These population-based results support dimensional models of somatization. Implications for changing the threshold of the categorical diagnosis of somatization disorder and providing better care for these patients are given.

Background. Many studies have reported a negative impact of divorce and separation on health although it is still unclear to what extent this is due to early vulnerability, the material and social consequences of divorce or to its direct emotional effects.

Method. Measures of anxiety and depression and potential alcohol abuse at age 43 were compared in 2085 participants from the MRC National Survey of Health and Development who were either married and never divorced or separated, or who had divorced or separated at least once. Analyses were adjusted for sociodemographic features, early vulnerability factors and current stressors.

Results. Divorce and separation were associated with increased anxiety and depression, and increased risk of alcohol abuse. This was the case after adjusting for educational attainment, age at first marriage, parental divorce, childhood aggression and neuroticism, and current financial hardship, lack of a confidante and frequency of social contact with friends or family. The association between divorce and risk of alcohol abuse became non-significant when the latter variable was controlled for. Associations between divorce and psychopathology were observed even though half of those separated or divorced were re-married or reunited with their spouses at the time of the analysis. There was, furthermore, no association between these mental health measures and time since first separation or divorce.

Conclusions. Divorce and separation have a specific and long-term impact on mental health.

Background. In this paper we consider verbal hallucinations as inner speech with pragmatics. The specific pragmatic properties of verbal hallucinations investigated included the number of voices, the characteristics that individuate the voices, the sequential characteristics of the dialogues between voice hearers and their voices, the dialogical positioning of voices hearers, voices and other individuals, and how the voices influence voice hearers' activities.

Methods. These properties were examined in structured interviews with 28 individuals, 14 of whom had a diagnosis of schizophrenia, while 14 were students who did not use psychiatric services.

Results. The analysis showed that voices were most frequently individuated with reference to individuals significant to voice hearers. The talk with voices was typically mundane and related to voice hearers' on-going activities, as is the case for ordinary inner speech. The voices were typically orientated towards the voice hearer, without direct access to each other or to other people. Contrary to received wisdom, the voices typically did not impel actions of voice hearers, rather they influenced voice hearers' decisions on how to act. This was so irrespective of the diagnostic status of informants. Finally, we have found some differences between the voices of informants with, and without, schizophrenia. These concerned the alignment of voices, the type of action required by a voice and the degree of dialogical engagement between voices and voice hearers.

Conclusions. We conclude that verbal hallucinations can be fruitfully considered to be a genus of inner speech. Pragmatics can be used as a framework to distinguish verbal hallucinations in different populations.

Background. We report the outcome of depressive states after 3-4 years in a community sample of the elderly.

Methods. A sample of 1045 persons aged 70+ years in 1990–1 was re-interviewed after 3·6 years.

Results. Mortality (21·7%) and refusal or non-availability (10·4%) were higher in those who initially had had a diagnosis or symptoms of depression. Of those with an ICD-10 depressive episode in 1990–1, 13% retained that diagnosis. Of those who were not depressed initially only 2·5% had become cases. Depression was unrelated to age or apolipoprotein E genotype. The best predictors of the number of depressive symptoms at follow-up was the number at Wave 1, followed by deterioration in health and in activities of daily living, high neuroticism, poor current health, poor social support, low current activity levels and high service use. Depressive symptoms at Wave 1 did not predict subsequent cognitive decline or dementia.

Conclusions. Non-random sample attrition is unavoidable. ICD-10 criteria yield more cases than other systems, while continuous measures of symptoms confer analytical advantages. Risk factors for depressive states in the elderly have been further identified. The prognosis for these states is favourable. At the community level, depressive symptoms do not seem to predict cognitive decline, as they do in referred series.

Background. The aims of this study were to examine the associations between exposure to unemployment following school leaving and rates of psychiatric disorder using data gathered on a birth cohort of New Zealand young people studied up to the age of 18.

Methods. At age 18 cohort members were assessed on: (a) duration of exposure to unemployment from age 16; (b) DSM-IV diagnostic criteria for major depression, anxiety disorders, conduct disorder, nicotine dependence, other substance abuse/dependence and attempted suicide. This information was integrated into longitudinal data gathered on the social circumstances, family background and adjustment of the cohort up to the age of 18.

Results. Increasing exposure to unemployment was associated with increasing risks of psychiatric disorder in adolescence. Those exposed to 6 months or more unemployment had rates of disorder that were 1·5 to 5·4 times higher than those not exposed to unemployment. However, most of the elevated risk of disorder among those unemployed was explained by family and personal factors that were present prior to school leaving age. Nonetheless, even after control for these factors those exposed to unemployment had significantly higher rates of anxiety disorder and substance use disorders.

Conclusions. To a large extent the relationships between unemployment and psychiatric disorder seen in this cohort were explained by social, family and personal factors that were present before school leaving age. Nonetheless, young people exposed to unemployment had higher rates of substance use and anxiety disorder.

Background. It is widely believed that for the severely mentally ill continuity of care is essential to ensure a better outcome and prevent long-term hospitalization. However, not much progress has been made in the operationalization and measurement of this concept. We used two indicators to compare continuity of care of schizophrenic patients in two kinds of mental health systems. One is a community mental health system without the back-up of a mental hospital (South-Verona, Italy). The other is an institution-based system in which mental hospitals are still predominant (Groningen, The Netherlands).

Methods. The first indicator of continuity of care, readiness of aftercare, is the time from discharge from hospital to the first day- or out-patient contact. Survival analysis was applied to correct for censored observations. The second indicator, flexibility of care, is the use of combinations of in-, day- and out-patient care during 2-year follow-up.

Results. More patients in South-Verona received community care within 2 weeks after discharge (71·5%), than in the Groningen register area (54·6%). The survival functions differed significantly. Cox regression analysis revealed that in both systems a contact before admission, the time between this contact and admission and the duration of the admission are predictors for aftercare. A higher percentage of patients made multiple service use (combinations of in-, day- and out-patient care) in South-Verona than in Groningen (62 v. 45%).

Conclusions. Both indicators showed a higher continuity of care in the South-Verona system.

Background. Many social and demographic correlates of psychiatric disorder have shown marked secular changes in recent decades. This study was designed to explore some of the implications of these trends, focusing on the illustrative case of teenage motherhood.

Method. Prospective data from two British birth cohort studies (the 1946 and 1958 cohorts) were used to examine the social, educational and behavioural precursors of teenage versus older age at motherhood, and the implications of teenage motherhood for women's later marital and social circumstances and risks of psychiatric morbidity, in samples born 12 years apart.

Results. Educational and social disadvantage were associated with similarly increased risks of teenage motherhood in both cohorts, but the findings suggested an additional association with teacher-rated adolescent conduct problems in the more recent sample. Rates of teacher-rated emotional problems were not elevated among teenage mothers in either cohort. In adult life, teenage motherhood was associated with a range of adverse social outcomes, including partnership breakdowns, large family size, and poorer housing conditions. Relative risks of these adult adversities were similar for teenage mothers in the two cohorts, but absolute levels of adversity were higher in the more recent sample, reflecting general secular changes in many of the indicators involved. In the later, but not the earlier, cohort, teenage motherhood was also associated with increased risks for psychiatric morbidity in adulthood.

Conclusions. The findings underline the importance of taking account of secular trends in examining the impact of psychosocial risks.

Background. An initial prevalence survey of mental disorders among 993 subjects aged 15 and above randomly drawn from four major Taiwanese aboriginal groups (the Atayal, Ami, Bunun and Paiwan) was conducted from 1986 to 1988. The incidence of alcoholism was investigated in a follow-up survey from 1990 to 1992.

Methods. Both surveys employed a semi-structured clinical interview with satisfactory reliability for case identification, and DSM-III-R as the diagnostic criteria for alcohol use disorders. The estimation of incidence rates of first onset alcoholism (alcohol abuse or dependence) was based on person-years at risk of 499 subjects who did not have any lifetime diagnosis of such morbidity at phase I.

Results. The follow-up rate was 99·6% and only four subjects among the survivors were not found. The age-standardized annual incidence rates of alcoholism ranged from 2·8 to 4·9% among the four aboriginal groups, and the rank order of rates was consistent with that of prevalences among these groups. The incidence rates of alcoholism were the highest among adolescents and young adults in men, and among the middle-aged in women.

Conclusions. High rates of first onset alcoholism among the Taiwanese aborigines indicate an interaction of sociocultural and biological factors in the development of such morbidity.