The temporal pattern of effects of summertime ozone (O₃) in total, cardiovascular and respiratory mortality were investigated in 21 European cities participating in the APHEA-2 (Air Pollution and Health: a European Approach) project, which is fundamental in determining the importance of the effect in terms of life loss.

Data from each city were analysed separately using distributed lag models with up to 21 lags. City-specific air pollution estimates were regressed on city-specific covariates to obtain overall estimates and to explore sources of possible heterogeneity.

Stronger effects on respiratory mortality that extend to a period of 2 weeks were found. A 10 µg/m³ increase in O₃ was associated with a 0.36% (95% CI –0.21% to 0.94%) increase in respiratory deaths for lag 0 and with 3.35% (95% CI 1.90% to 4.83%) for lags 0–20. Significant adverse health effects were found of summer O₃ (June–August) on total and cardiovascular mortality that persist up to a week, but are counterbalanced by negative effects thereafter.

The results indicate that studies on acute health effects of O₃ using single-day exposures may have overestimated the effects on total and cardiovascular mortality, but underestimated the effects on respiratory mortality.

As a result of increasing urbanisation, people face the prospect of living in environments with few green spaces. There is increasing evidence for a positive relation between green space in people’s living environment and self-reported indicators of physical and mental health. This study investigates whether physician-assessed morbidity is also related to green space in people’s living environment.

Morbidity data were derived from electronic medical records of 195 general practitioners in 96 Dutch practices, serving a population of 345 143 people. Morbidity was classified by the general practitioners according to the International Classification of Primary Care. The percentage of green space within a 1 km and 3 km radius around the postal code coordinates was derived from an existing database and was calculated for each household. Multilevel logistic regression analyses were performed, controlling for demographic and socioeconomic characteristics.

The annual prevalence rate of 15 of the 24 disease clusters was lower in living environments with more green space in a 1 km radius. The relation was strongest for anxiety disorder and depression. The relation was stronger for children and people with a lower socioeconomic status. Furthermore, the relation was strongest in slightly urban areas and not apparent in very strongly urban areas.

This study indicates that the previously established relation between green space and a number of self-reported general indicators of physical and mental health can also be found for clusters of specific physician-assessed morbidity. The study stresses the importance of green space close to home for children and lower socioeconomic groups.

Although trouble sleeping is a common problem among women at mid-life, patterns in trouble sleeping relating to social and health-related risk factors are unclear. This analysis describes the dynamics of trouble sleeping among women at mid-life.

The National Survey of Health and Development is a nationally representative study of births in 1946 in England, Scotland, and Wales followed up through mid-life. Multistate life table analysis utilised 893 women interviewed annually between ages 48 to 54 years.

Women spent an average of 2.6 years with trouble sleeping, and the average length of a continuous episode of trouble sleeping was 2.4 years. Among women who reported at least one episode, the average number of episodes was 1.5. Health-related risk factors at age 43 of number of physical conditions, anxiety and depression symptoms, use of prescription medication, and current or past trouble sleeping were related to increased total and per episode duration of trouble sleeping over the 7-year study interval and increased duration per episode. Differences associated with these risk factors ranged from 1.2 to 1.8 years for duration over the study interval and 0.5 to 0.8 years per episode. There was no association between average number of episodes per woman reporting at least one episode and these health-related risk factors at age 43.

This study provides support for association between increased duration of trouble sleeping, in total and per episode, and health risk factors at age 43, suggesting a long-term relationship between risk factors and sleep.

High job strain has been linked with cardiovascular outcomes. This study aimed to examine whether job strain is associated with angina pectoris symptoms among British and Finnish non-manual employees.

Postal questionnaire survey data among 40–60-year-old employees of the British Whitehall II Study (n = 4551, 27% women) and the Finnish Helsinki Health Study (n = 7605, 83% women) cohort were analysed. Angina pectoris symptoms were the outcome in logistic regression analysis. Karasek’s job strain was examined. Models were adjusted first for age, second for occupational class and finally for smoking, heavy drinking, physical inactivity, unhealthy food habits and obesity.

Angina pectoris symptoms were reported by 5% of women and 3% of men in Britain, and by 6% of women and 4% of men in Finland. High job strain was associated with angina pectoris symptoms among men in Britain (OR 2.08; CI 95% 1.07 to 4.02) and women in Finland (OR 1.90; CI 95% 1.36 to 2.63) independent of age, occupational class, and behavioural risk factors. However, similar associations between job strain and angina pectoris symptoms were not observed among men in Finland or women in Britain.

The results yielded partial support for the association between job strain and angina pectoris symptoms across national contexts.

South Asian people in the UK and other western countries have elevated rates of coronary heart disease (CHD). Psychosocial factors contribute to CHD risk, but information about psychosocial risk profiles in UK South Asians is limited. This study aimed to examine the profile of conventional and novel psychosocial risk factors in South Asian compared with white men and women.

Using a cross-sectional population study design, psychosocial profiles were assessed in 1130 South Asian and 818 white European healthy men and women aged between 35 and 75 years, who had previously participated in a cardiovascular risk assessment programme in West London. Psychosocial factors potentially contributing to CHD risk were assessed using standardised questionnaires.

UK South Asians reported significantly higher psychosocial adversity compared with UK whites. South Asian men and women experienced greater chronic stress, in the form of financial strain, residential crowding, family conflict, social deprivation and discrimination, than white Europeans. They had larger social networks, but reported lower social support and greater depression and hostility. These effects were largely independent of socioeconomic status.

UK South Asians experience significant psychosocial adversity compared with UK white Europeans. This is consistent with the heightened vulnerability to CHD observed in this population.

Dimensions of the individual socioeconomic position—education, social class, social status and income—are associated with mortality. Inequalities in death also related to the social position of the household. It is, however, less clear how the socioeconomic position of one marital/cohabiting partner influences the mortality of the other partner. We examine the independent effect on mortality of own and partner’s positions regarding these four socioeconomic factors.

Register data on education, social class, social status and income of both marital/cohabiting partners were collected from the 1990 Census of the employed Swedish population aged 30–59 (N = 1 502 148). Data on all-cause mortality and deaths from cancer and circulatory disease for the subsequent period 1991–2003 were collected from the Cause of Death Register. Relative mortality risks for different socioeconomic groups were estimated by Cox regression.

All-cause mortality of both men and women differs by women’s education and status and by men’s social class and income. For men, the wife’s education is more important for the mortality risk than his own education when the man’s social class is included in the model. For women, the husband’s social class yields larger mortality differences than own occupational measures. Women’s education and men’s social class and income are particularly important for women’s deaths from circulatory diseases.

The partner’s social position has a clear independent association with individual mortality, and women’s education and men’s social class seem to be particularly important. Suggested explanations of health inequality are not always compatible with the observed relationship between partners’ social and economic resources and mortality.

There is little known about potential mechanisms underlying the association between maternal employment and childhood obesity. The relationships between maternal hours worked per week (none, 1–20 hours, 21+ hours) and children’s dietary and physical activity/inactivity habits were examined. Where mothers were employed, the relationships between flexible work arrangements and these health behaviours were also examined.

Data from 12 576 singleton children age 5 years in the UK Millennium Cohort Study were analysed. Mothers reported information about their employment patterns. Mothers also reported on indicators of their child’s dietary (crisps/sweets, fruit/vegetables, sweetened beverage, fruit consumption), physical activity (participation in organised exercise, transport to school) and inactivity (television/computer use) habits at age 5.

After adjustment for potential confounding and mediating factors, children whose mothers worked part-time or full-time were more likely to primarily drink sweetened beverages between meals (compared to other beverages), use the television/computer at least 2 hours daily (compared to 0–2) or be driven to school (compared to walk/cycle) than children whose mothers had never been employed. Children whose mothers worked full-time were less likely to primarily eat fruit/vegetables between meals (compared to other snacks) or eat three or more portions of fruit daily (compared to two or fewer). Although in unadjusted analyses children whose mothers used flexible work arrangements engaged in healthier behaviours, relationships were no longer significant after adjustment.

For many families the only parent or both parents are working. This may limit parents’ capacity to provide their children with healthy foods and opportunities for physical activity. Policies and programmes are needed to help support parents and create a health-promoting environment.

Among adults, a stronger sense of coherence (SOC) seems to promote a better coping with strain and is associated with healthier behaviours, including eating patterns. The parents’ SOC may also have a wider effect on the health behaviour of their children. The aim of this study is to determine whether there are associations between parents’ SOC and the eating patterns of their children and if this can be explained by mediating factors.

Cross-sectional. In 2006, 1268 (response rate 79%) 10- and 11-year-old children in southern Finland, in a classroom situation, filled in a questionnaire assessing meal pattern and food frequency intake. Parents, 816 (response rate 64%), filled in a questionnaire assessing SOC (13 items) and eating patterns. Matching data were found from 772 child–parent pairs. ² tests, factor analysis, univariate analysis of variance and logistic regression analyses were used as statistical methods.

A weaker parental SOC was associated with children’s irregular meal pattern (p = 0.004), more frequent intake of energy-rich foods (p = 0.002) and less frequent intake of nutrient-rich foods (p = 0.051). Mediating factors, such as availability and accessibility of fruit and vegetables at home, parent’s nutrition knowledge, parents’ own fruit and vegetable intake and an irregular meal pattern, explained the association between parents’ SOC and children’s intake of nutrient-dense foods, but not the association with energy-rich foods.

Parents’ weaker SOC was associated with children’s unhealthier eating patterns. More research is needed on the associations between parents’ SOC and other health-related behaviours in their children.

Studies of human papillomavirus (HPV) awareness and HPV vaccine acceptability have included few non-white participants, making it difficult to explore ethnic differences. This study assessed HPV awareness and HPV vaccine acceptability in a sample of women representing the major UK ethnic minority groups.

A cross-sectional study design was used to assess awareness of HPV and acceptability of HPV vaccination. Participants were recruited using quota sampling to ensure adequate representation of ethnic minority women: Indian, Pakistani, Bangladeshi, Caribbean, African and Chinese women (n = 750). A comparison sample of white British women (n = 200) was also recruited.

Awareness of HPV was lower among ethnic minority women than among white women (6–18% vs 39% in white women), and this was not explained by generational status or language spoken at home. In a subsample who were mothers (n = 601), ethnicity and religion were strongly associated with acceptability of HPV vaccination. Acceptability was highest among white mothers (63%) and lowest among South Asians (11–25%). Those from non-Christian religions were also less accepting of the vaccine (17–34%). The most common barriers to giving HPV vaccination were a need for more information, sex-related concerns and concern about side-effects. South Asian women were the most likely to cite sex-related concerns, and were also least likely to believe the vaccine would offer their daughters protection.

These findings suggest some cultural barriers that could be addressed in tailored information aimed at ethnic minority groups. They also highlight the importance of recording ethnicity as part of HPV vaccine uptake data.

Some small studies have reported high risk of suicide after surgical treatment for peptic ulcer. The aim of the present study was to explore the risk of suicide in hospitalised gastric ulcer and duodenal ulcer patients separately among operated and non-operated cohorts.

Retrospective cohorts of 163 579 non-operated patients with gastric ulcer or duodenal ulcer and 28 112 patients with surgical treatment for ulcer, recorded in the Swedish Inpatient Register since 1965, were followed from the first hospitalisation, or operation for the surgery cohort, until death, any cancer, emigration, or 31 December 2003. Standardised mortality ratios (SMRs) were calculated, and Poisson regression produced adjusted relative risk estimates among operated and non-operated patients.

Non-operated patients hospitalised for peptic ulcer showed a 70% excess risk of suicide (SMR 1.7, 95% CI 1.6 to 1.9) and those who underwent operation had a 60% increased risk (SMR 1.6, 95% CI 1.4 to 1.8). The risk of suicide was very high during the first year after hospitalisation (SMR 4.0, 95% CI 3.4 to 4.7) and more marked among women, patients under 70 and patients hospitalised without complications of ulcer. Both gastric ulcer and duodenal ulcer patients had high risk of suicide completion.

Hospitalised patients with gastric ulcer or duodenal ulcer have an increased risk of suicide regardless of surgical treatment. These patients, especially women, are at very high risk during the first year after first hospitalisation/operation. The evaluation and management of suicidal thoughts in patients in medical settings should be further considered.

Ethnic/racial inequalities in access to and quality of healthcare have been repeatedly documented in the USA. Although there is some evidence of inequalities in England, research is not so extensive. Ethnic inequalities in use of primary and secondary health services, and in outcomes of care, were examined in England.

Four waves of the Health Survey for England were analysed, a representative population survey with ethnic minority oversamples. Outcome measures included use of primary and secondary healthcare services and clinical outcomes of care (controlled, uncontrolled and undiagnosed) for three conditions – hypertension, raised cholesterol and diabetes.

Ethnic minority respondents were not less likely to use GP services. For example, the adjusted odds ratios for Indian, Pakistani and Bangladeshi versus white respondents were 1.29 (95% confidence intervals 1.07 to 1.54), 1.32 (1.10 to 1.58) and 1.35 (1.10 to 1.65) respectively. Similarly, there were no ethnic inequalities for the clinical outcomes of care for hypertension and raised cholesterol, and, on the whole, no inequalities in outcomes of care for diabetes. There were ethnic inequalities in access to hospital services, and marked inequalities in use of dental care.

Ethnic inequalities in access to healthcare and the outcomes of care for three conditions (hypertension, raised cholesterol and diabetes), for which treatment is largely provided in primary care, appear to be minimal in England. Although inequalities may exist for other conditions and other healthcare settings, particularly internationally, the implication is that ethnic inequalities in healthcare are minimal within NHS primary care.

Children have a social status position of their own, apart from that of the family, that may have an impact on short-term and long-term health. The aim of the present study was to analyse the associations between childhood social status in school (ie, peer status) and disease-specific morbidity in adulthood.

Data were derived from a longitudinal study using a 1953 cohort born in Stockholm, Sweden: The Stockholm Birth Cohort Study (1953–2003). Peer status was sociometrically assessed in sixth grade (1966). Hazard ratios for adult disease-specific morbidity based on information on inpatient care (1973–2003) were calculated by peer status category for men and women separately, using Cox regression.

The results indicate that the lower the childhood peer status, the higher the overall adult disease risk. There were, however, differences in the degree and magnitude to which disease-specific inpatient care varied with peer status. Some of the steepest gradients were found for mental and behavioural disorders (eg, alcohol abuse and drug dependence), external causes (eg, suicide) and various lifestyle-related diseases (eg, ischaemic heart disease and diabetes). The results were not explained by childhood social class.

The present study underlines the importance of recognising children’s social position, apart from that of their family, for later health. Not only psychologically related diseases but also those related to behavioural risk factors demonstrate some of the largest relative differences by peer status, suggesting that health-related behaviour may be one important mechanism in the association between peer status and morbidity.

Studies of the association between neighbourhood fragmentation and suicide have yielded varied results, and none has simultaneously adjusted for neighbourhood fragmentation, neighbourhood deprivation and individual-level factors.

A multilevel analysis of a 3-year cohort study was carried out using probabilistic linkage of census and mortality records, and two measures of neighbourhood fragmentation. A total of 2.8 million respondents to the 1996 New Zealand census were followed up for 3 years for mortality (1101 suicide deaths in the analysis), aged from 20 to 74 years at follow-up.

No consistent association was observed between either measure of neighbourhood social fragmentation and suicide, after controlling for individual-level confounders and neighbourhood deprivation. There was some evidence of a U-shaped relationship between neighbourhood fragmentation and suicide, especially for the Congdon(NZ) index. There was no evidence of an association for a nine-variable index that captured family-type variables as well as the usual attachment-type variables. Neighbourhood deprivation remained as an important predictor of suicide rates.

This paper highlights the importance of understanding both the measure and the wider national context when considering neighbourhood effects on suicide.

Social epidemiology investigates both individuals and their collectives. Although the limits that define the individual bodies are very apparent, the collective body’s geographical or cultural limits (eg "neighbourhood") are more difficult to discern. Also, epidemiologists normally investigate causation as changes in group means. However, many variables of interest in epidemiology may cause a change in the variance of the distribution of the dependent variable. In spite of that, variance is normally considered a measure of uncertainty or a nuisance rather than a source of substantive information. This reasoning is also true in many multilevel investigations, whereas understanding the distribution of variance across levels should be fundamental. This means-centric reductionism is mostly concerned with risk factors and creates a paradoxical situation, as social medicine is not only interested in increasing the (mean) health of the population, but also in understanding and decreasing inappropriate health and health care inequalities (variance).

Critical essay and literature review.

The present study promotes (a) the application of measures of variance and clustering to evaluate the boundaries one uses in defining collective levels of analysis (eg neighbourhoods), (b) the combined use of measures of variance and means-centric measures of association, and (c) the investigation of causes of health variation (variance-altering causation).

Both measures of variance and means-centric measures of association need to be included when performing contextual analyses. The variance approach, a new aspect of contextual analysis that cannot be interpreted in means-centric terms, allows perspectives to be expanded.