Many studies have been published which have examined the relationship between paid employment and women's health. As employment outside the home is likely to have differential effects for women with different family commitments, further analysis taking account of the association between paid employment and household circumstances is necessary.

Using data from a large, representative British sample, this paper examines the effects of interactions between paid employment, social class, and parental status on women's health. The results show differential effects of these variables on physical and mental health. The most important influence on women's mental health (as measured by the 30-item General Health Questionnaire) is the age of their youngest child; women with children under five are most likely to show signs of psychological disturbance. With respect to physical health, age of the youngest child has no significant effect, but there is an interaction between employment status and social class. Paid employment, particularly full-time work, is associated with good physical health for middle-class women but not for working-class women.

This paper investigates factors affecting the distribution of psychiatric morbidity in the community. It identifies a close relationship between mean Chronic General Health Questionnaire (CGHQ) scores in subdivisions of a large random sample of the community (the Health and Lifestyle Survey, N = 6317) and the prevalence in these groups of abnormal, above-threshold CGHQ scores. The frequency distributions of CGHQ scores in these different populations move up and down as a whole: like other physiological and behavioural attributes, these mental health outcomes in individuals are associated with characteristics of the populations in which they arise. Populations thus carry a collective responsibility for their own mental health and well-being. This implies that explanations for the differing prevalence rates of psychiatric morbidity must be sought in the characteristics of their parent populations; and control measures are unlikely to succeed if they do not involve population-wide changes.

The paper of Anderson et al. (1993), based on cross-sectional data, showed that minor psychiatric disorder in a population is linearly related to the mean number of psychiatric symptoms in the population. The present investigation asks whether the same relationship holds longitudinally as well as cross-sectionally. Data from a 7-year follow-up of a general population sample demonstrate, for the first time, that a relationship exists between changes in prevalence of psychiatric disorder and changes in the mean number of psychiatric symptoms in a given population. Moreover, the relationship is linear; a one-point decrease in mean scores on the GHQ-30 is associated with a 6% decrease in prevalence of disorder.

As part of the Health and Lifestyle Survey (HALS), 6096 adults from a representative sample of community residents completed a standardized questionnaire which elicited psychiatric symptoms – the General Health Questionnaire (GHQ-30) of Goldberg (1972). Their survival status at 7 years was determined, and 474 had died. Respondents whose GHQ score was above the conventional threshold (4/5), used when screening for psychiatric disorder, were 60% more likely to die over the 7-year interval than respondents whose GHQ score was below threshold (risk ratio 1·64 for men and 1·58 for women). The differences were statistically significant for both sexes when the effects of age, physical disorder, social class and smoking behaviour were taken into account. Personality traits (neuroticism, extraversion) were not related to mortality, neither was self-reported difficulty in sleeping. However, a low subjective rating of one's general health was associated with increased mortality. We explored a range of cut-points and found reliable effects for both sexes at all cut-points from 3/4 to 6/7. Possible differences related to natural versus unnatural causes of death were explored and found not to explain the results. The findings from this general population study confirm that it is not only psychiatric disorders, but also psychiatric symptoms, which are associated with subsequent mortality, and thereby reinforce the importance of appropriate intervention to alleviate these symptoms.

The value of a new method of scoring the General Health Questionnaire (CGHQ) has been assessed in a representative sample of 6317 British community residents aged 18 +. The CGHQ, which takes account of the chronicity of psychiatric symptoms, is found to be superior to the conventional scoring method in yielding a wider range of scores, a more normal distribution and in its close association to a well validated measure of neurotic illness.

On the other hand, the conventional GHQ and the CGHQ score appear equally sensitive to a number of socio-demographic variables, including social class, marital status and employment. An important finding is that the two methods identify different individuals as cases. Compared with the GHQ, the CGHQ identifies a higher proportion of the middle-aged. The implications of these findings with respect to the optimal method of scoring the GHQ are discussed.

Background. Common mental disorder prevalence decreases substantially around the conventional retirement age for men in the UK, but trends for older women are more continuous. Prevalence changes in depression and anxiety around retirement are less clear, as is the role of risk factors. The aim of this study was to establish whether work status, age or other known risk factors account for the reduced prevalence of depressive episode and anxiety disorder around retirement ages for men and for women.

Method. The British Psychiatric Morbidity Survey (BPMS) 2000 was analysed, including 1875 men and 2253 women aged 45–75 years. Diagnoses were from the Revised Clinical Interview Schedule (CIS-R). Logistic models were adjusted for sociodemographic factors, social network, work status, life events, physical illness and disability.

Results. There are marked reductions in the prevalence of depressive episode after 60 years for women [60% lower prevalence, 95% confidence interval (CI) 40–80] and 65 years for men (90% lower prevalence, 95% CI 70–100), compared to the youngest age groups. For anxiety disorder, the reduction in prevalence was 80% (95% CI 60–90) for men and 40% (95% CI 20–60) for women. In fully adjusted multivariate models, the strong association between diagnoses and age groups remained, for both genders. Work status was a significant factor for men but not for women.

Conclusion. There is a discontinuity in the prevalence of depressive episode for both men and women, coinciding with statutory retirement ages. No studied risk factor reduced the associations between age group and disorders. This population scale recovery may provide a model for understanding non-genetic factors.

Attitudes to death were explored in 125 community residents aged 81 and older. Those people who thought about dying had less frequent contact with their interview informant; those who thought about death more frequently showed less severe cognitive impairment, a greater severity of depressive symptoms, and were more likely to be unmarried and were more commonly reported to be depressed by their interview informant. Worries about dying showed no association with sociodemographic or clinical variables. The commonest category of response from spontaneous comments was concerning the circumstances of dying. The results are discussed in light of other research findings, and emphasize the influence of low mood and social factors on death attitudes. This may have implications for closer examination of such attitudes in the assessment of depression and suicidal risk as well as in the care of the dying.

An individual's responses to Goldberg's 30-item General Health Questionnaire are usually represented as a single score which provides a measure of the number of psychiatric symptoms reported. No account is taken of the nature of the symptoms. Factor analyses of the GHQ-30 were undertaken in ten randomly selected samples of 600 adults each, and also on 12 age-sex groupings covering the age range 18–98. The results indicate an impressive degree of consistency of the factor structure, and the identification of five distinct factors corresponding to anxiety, feelings of incompetence, depression, difficulty in coping, and social dysfunction.