Showing papers by "George Davey Smith published in 2004"

Is Income Inequality a Determinant of Population Health? Part 1. A Systematic Review

Abstract: This article reviews 98 aggregate and multilevel studies examining the associations between income inequality and health. Overall, there seems to be little support for the idea that income inequality is a major, generalizable determinant of population health differences within or between rich countries. Income inequality may, however, directly influence some health outcomes, such as homicide in some contexts. The strongest evidence for direct health effects is among states in the United States, but even that is somewhat mixed. Despite little support for a direct effect of income inequality on health per se, reducing income inequality by raising the incomes of the most disadvantaged will improve their health, help reduce health inequalities, and generally improve population health.

Mendelian randomization: prospects, potentials, and limitations

Abstract: Mendelian randomization is the term applied to the random assortment of alleles at the time of gamete formation. This results in population distributions of genetic variants that are generally independent of behavioural and environmental factors that typically confound epidemiological associations between putative risk factors and disease. In some circumstances this can provide a study design akin to randomized comparisons. The principles of Mendelian randomization can serve to limit several potential problems in observational epidemiology (Table 1). The avoidance of confounding is clearly a key advantage, and in view of this, Martin Tobin and colleagues6 have suggested that the approach should be termed ‘Mendelian deconfounding’. However, there are several additional and perhaps equally important ways in which Mendelian randomization can strengthen inferences drawn from observational studies. In the example Katan originally presented—that of the association between low serum cholesterol levels and cancer—the most plausible bias would be introduced by reverse causation. The early stages of cancer could lead to a decrease in circulating cholesterol levels, generating an inverse association between cholesterol levels and cancer morbidity or mortality.1 Early stages of cancer will not, however, change inherited genetic variants that are associated with cholesterol levels. Thus if low cholesterol level were a cause of increased cancer risk then individuals with genetic variants associated with lower cholesterol levels should have a higher cancer risk. If, on the other hand, reverse causation is responsible for the association between cholesterol level and cancer, there should be no association between genetic variants related to cholesterol level and cancer risk. Biological forms of reverse causation may influence many epidemiological associations—for example, those between markers of inflammation and coronary heart disease, where existing atherosclerosis may influence the level of factors such as fibrinogen and C-reactive protein.13 Reverse causation can also occur through exposure assignment—for example, people with early stages of coronary heart disease may take vitamin supplements because they believe these will reduce their risk of cardiovascular events. This will tend to generate a positive association between vitamin intake and disease. A form of reverse causation can also occur through reporting bias, with the presence of disease influencing reporting disposition. In casecontrol studies people with the disease under investigation may report on their prior exposure history in a different way than do controls—perhaps because the former will think harder about potential reasons that account for why they have developed the disease. In this situation the association between genetic variants related to the exposure and disease outcome will not usually be biased. In observational studies associations between an exposure and disease will generally be biased if there is selection according to an exposure–disease combination in case-control studies, or according to an exposure–disease risk combination in prospective studies. If, for example, people with an exposure and at low risk of disease for other reasons were differentially excluded from a study the exposure would appear to be positively related to disease outcome, even if there were no such association in the underlying population. This is a form of ‘Berkson’s bias’, well known to epidemiologists.14 A possible example of such associative selection bias relates to the finding in the large American Cancer Society volunteer cohort that high alcohol consumption was associated with a reduced risk of stroke.15 This is somewhat counter-intuitive as the outcome category included haemorrhagic stroke (for which there is no obvious mechanism through which alcohol would reduce risk) and because alcohol is known to increase blood pressure16,17— a major causal factor for stroke.18 Population-based studies have found that alcohol tends to increase stroke risk.19–21 Heavy drinkers who volunteer for a study known to be about the health effects of their lifestyle are likely to be very unrepresentative of all heavy drinkers in the population, in ways that render them to be at low risk of stroke. Moderate and non-drinkers

Childhood Socioeconomic Circumstances and Cause-specific Mortality in Adulthood: Systematic Review and Interpretation

Abstract: There is convincing evidence that exposures acting across the life course influence adult health outcomes (1–3). Lifecourse epidemiology examines a range of potential processes through which exposures acting at different stages of life can, singly or in combination, influence disease risk (table 1) (4). In the critical period model, an exposure acting at a specific time has long-lasting effects on the structure or function of the body. The fetal origins hypothesis, in its original formulation, took this approach (5). Other examples of processes where outcomes appear to depend upon the time window during which an exposure acts are limb development (in relation to maternal thalidomide use); infection with hepatitis B and risk of adulthood liver cancer (with very early postnatal infection being most implicated); and environmental lead exposure, which results in serious neurodevelopmental deficits only if occurring in infancy and childhood (3). However, the influence of exposures acting during critical periods of susceptibility may be modified by later life exposures. This is the case for the associations of birth weight with coronary heart disease, high blood pressure, and insulin resistance, where associations are stronger (or only evident) among those who become obese during adolescence or adulthood (6–8).

“Bodies Count,” and Body Counts: Social Epidemiology and Embodying Inequality

Abstract: INTRODUCTION Bodies count. In epidemiology, this statement would appear to be a core proposition, for it is by counting people—in varying states of health, disease, and disability, the alive and the dead—that we derive our estimates of population rates and risks of morbidity and mortality. But bodies count for more than this, for, in their manifest form—in height, weight, physique, and overall appearance (including posture and disfigurement)—they provide vivid evidence of how we literally embody the world in which we live, thereby producing population patterns of health, disease, disability, and death (1–5). Readily identifiable to the naked eye, these aspects of our being not only are predictive of future health outcomes but also tell of our conjoined social and biologic origins and trajectories.

Associations of parental, birth, and early life characteristics with systolic blood pressure at 5 years of age. Findings from the Mater-University Study of Pregnancy and its Outcomes

Abstract: Background— We examined the associations of a range of parental and early life characteristics with systolic blood pressure at 5 years of age. Methods and Results— Information from 3864 children who were followed up prospectively from their mother’s first antenatal clinic assessment was used. Maternal age, body mass index, and smoking during pregnancy were all positively associated with offspring systolic blood pressure at 5 years of age. The systolic blood pressure of children whose mothers had smoked throughout pregnancy was on average 0.92 mm Hg (95% CI 0.17 to 1.68) greater than that of children whose mothers had never smoked, after full adjustment. Children who had been breast fed until at least 6 months had lower systolic blood pressure than those who were breast fed for a shorter duration. Paternal body mass index and child’s weight, height, and body mass index were all positively associated with blood pressure at age 5. Conclusions— Because childhood blood pressure tracks into adulthood, intervent...

Patterns and distribution of tobacco consumption in India: cross sectional multilevel evidence from the 1998-9 national family health survey

Abstract: Objective To investigate the demographic, socioeconomic, and geographical distribution of tobacco consumption in India. Design Multilevel cross sectional analysis of the 1998-9 Indian national family health survey of 301 984 individuals in 92 447 households in 3215 villages in 440 districts in 26 states. Setting Indian states. Participants 301 984 adults (≥ 18 years). Main outcome measures Dichotomous variable for smoking and chewing tobacco for each respondent (1 if yes, 0 if no) as well as a combined measure of whether an individual smokes, chews tobacco, or both. Results Smoking and chewing tobacco are systematically associated with socioeconomic markers at the individual and household level. Individuals with no education are 2.69 times more likely to smoke and chew tobacco than those with postgraduate education. Households belonging to the lowest fifth of a standard of living index were 2.54 times more likely to consume tobacco than those in the highest fifth. Scheduled tribes (odds ratio 1.23, 95% confidence interval 1.18 to 1.29) and scheduled castes (1.19, 1.16 to 1.23) were more likely to consume tobacco than other caste groups. The socioeconomic differences are more marked for smoking than for chewing tobacco. Socioeconomic markers and demographic characteristics of individuals and households do not account fully for the differences at the level of state, district, and village in smoking and chewing tobacco, with state accounting for the bulk of the variation in tobacco consumption. Conclusion The distribution of tobacco consumption is likely to maintain, and perhaps increase, the current considerable socioeconomic differentials in health in India. Interventions aimed at influencing change in tobacco consumption should consider the socioeconomic and geographical determinants of people9s susceptibility to consume tobacco.

Body Mass Index and Ischemic and Hemorrhagic Stroke A Prospective Study in Korean Men

Abstract: Background and Purpose— The association between obesity and stroke remains controversial, with earlier studies suggesting that differences might stem from heterogeneous stroke subtype compositions. The association between body mass index (BMI) and stroke subtypes was examined prospectively in a large cohort study. Methods— A total of 234 863 Korean men aged 40 to 64 years without substantial weight loss over 4 years after baseline examination in 1986 were divided into 8 categories of BMI and were followed up between 1991 and 2000 for fatal and nonfatal stroke events. Results— There was a positive association across the whole range of BMI and ischemic stroke, with a confounder-adjusted hazard of 11% (95% CI, 1.09 to 1.12) for 1 kg/m2 higher BMI. A J-shaped association was observed between BMI and hemorrhagic stroke; groups with a higher BMI than the reference category (22 to 23 kg/m2) had significantly increased risks. Full adjustment for confounders and variables potentially on the causal pathway (ie, blo...

Advice to reduce dietary salt for prevention of cardiovascular disease

Abstract: Restricting sodium intake in hypertensive patients over short periods of time reduces blood pressure. Long term effects (on mortality, morbidity or blood pressure) of advice to reduce salt in patients with elevated or normal blood pressure are unclear. Objectives To assess in adults the long term effects (mortality, cardiovascular events, blood pressure, quality of life, weight, urinary sodium excretion, other nutrients and use of anti-hypertensive medications) of advice to restrict dietary sodium using all relevant randomised controlled trials. Inclusion decisions were independently duplicated and based on the following criteria: 1) randomisation was adequate; 2) there was a usual or control diet group; 3) the intervention aimed to reduce sodium intake; 4) the intervention was not multifactorial; 5) the participants were not children, acutely ill, pregnant or institutionalised; 6) follow-up was at least 26 weeks; 7) data on any of the outcomes of interest were available. Three trials in normotensives (n=2326), five in untreated hypertensives (n=387) and three in treated hypertensives (n=801) were included, with follow up from six months to seven years. The large, high quality (and therefore most informative) studies used intensive behavioural interventions. Deaths and cardiovascular events were inconsistently defined and reported; only 17 deaths equally distributed between intervention and control groups occurred. Systolic and diastolic blood pressures were reduced at 13 to 60 months in those given low sodium advice as compared with controls (systolic by 1.1 mm Hg, 95% CI 1.8 to 0.4, diastolic by 0.6 mm hg, 95% CI 1.5 to -0.3), as was urinary 24 hour sodium excretion (by 35.5 mmol/ 24 hours, 95% CI 47.2 to 23.9). Degree of reduction in sodium intake and change in blood pressure were not related. People on anti-hypertensive medications were able to stop their medication more often on a reduced sodium diet as compared with controls, while maintaining similar blood pressure control.

Hyperinsulinaemia and Increased Risk of Breast Cancer: Findings From the British Women's Heart and Health Study

Abstract: Objective: To assess the association between fasting insulin levels and breast cancer. Design: Cross sectional study. Participants: 3868 women aged 60-79 years. Main outcome measure: Prevalent breast cancer (151 cases). Results: Insulin levels were positively associated with breast cancer. The age adjusted odds ratio (95% confidence interval) for a one unit increase in log(e) insulin levels among women without diabetes was 1.34 (1.02, 1.77). This association was not substantively altered by adjustment for potential confounding factors (age of menopause, hysterectomy/oophorectomy, hormone replacement use, oral contraceptive use, parity, adult social class and smoking) or potential mediating factors (body mass index, waist to hip ratio, leg length, age at menarche and childhood social class). Women with both long legs and higher insulin levels were at particularly increased risk, with breast cancer prevalence being 5.7% among women in the highest thirds of both insulin levels and leg-length compared to 1.8% among those in the lowest thirds of both. Positive associations between insulin levels and breast cancer were found for both pre- and post-menopausal breast cancers. Fasting glucose levels, HOMA score, diabetes and a history of gestational glycosuria or diabetes were also positively associated with breast cancer. Conclusions: Hyperinsulinaemia is positively associated with breast cancer in this cohort of older women. This effect may be mediated via a number of hormonal pathways acting at different stages of the life course.

Is Income Inequality a Determinant of Population Health? Part 2. U.S. National and Regional Trends in Income Inequality and Age- and Cause-Specific Mortality

Abstract: This article describes US income inequality and 100-year national and 30-year regional trends in age- and cause-specific mortality There is little congruence between national trends in income inequality and age- or cause-specific mortality except perhaps for suicide and homicide The variable trends in some causes of mortality may be associated regionally with income inequality However, between 1978 and 2000 those regions experiencing the largest increases in income inequality had the largest declines in mortality (r= 081, p < 0001) Understanding the social determinants of population health requires appreciating how broad indicators of social and economic conditions are related, at different times and places, to the levels and social distribution of major risk factors for particular health outcomes

Does breast-feeding in infancy lower blood pressure in childhood? The Avon Longitudinal Study of Parents and Children (ALSPAC).

Abstract: Background— Breast-feeding in infancy has been associated with decreased coronary heart disease mortality, but the underlying mechanisms are unclear. We investigated the association of breast-feedi...

Advanced Paternal Age and Risk of Fetal Death: A Cohort Study

Abstract: A possible detrimental paternal age effect on offspring health due to mutations of paternal origin should be reflected in an association between paternal age and fetal loss. The authors used data from a prospective study of 23,821 pregnant women recruited consecutively to the Danish National Birth Cohort from 1997 to 1999 to assess the association between paternal age and fetal death. Fathers of the pregnancies were identified by record linkage to population registers. The paternal age-related risks of fetal death and its components, early and late fetal loss, were estimated using survival analysis. Pregnancies fathered by a man aged 50 or more years (n = 124) had almost twice the risk of ending in a fetal loss compared with pregnancies with younger fathers (hazard ratio = 1.88, 95% confidence interval: 0.93, 3.82), after adjustment for maternal age, reproductive history, and maternal lifestyle during pregnancy. Various approaches to adjustment for potential residual confounding of the relation by maternal age did not affect the relative risk estimates. The paternal age-related risk of late fetal death was higher than the risk of early fetal death and started to increase from the age of 45 years. It should, however, be interpreted cautiously because of the restricted number of fetal deaths.

Measurement of the electric form-factor of the neutron at $Q^2$ = 0.5 and 1.0 $GeV^2/c^2$

Abstract: The electric form factor of the neutron was determined from measurements of the (d) over right arrow((e) over right arrow ,e(`)n)p reaction for quasielastic kinematics. Polarized electrons were scattered off a polarized deuterated ammonia ((ND3)-N-15) target in which the deuteron polarization was perpendicular to the momentum transfer. The scattered electrons were detected in a magnetic spectrometer in coincidence with neutrons in a large solid angle detector. We find G(E)(n)=0.0526+/-0.0033(stat)+/-0.0026(sys) and 0.0454+/-0.0054+/-0.0037 at Q(2)=0.5 and 1.0 (GeV/c)(2), respectively.

Diabetes status and post-load plasma glucose concentration in relation to site-specific cancer mortality: findings from the original Whitehall study.

Abstract: ObjectiveWhile several studies have reported on the relation of diabetes status with pancreatic cancer risk, the predictive value of this disorder for other malignancies is unclear. Methods: The Whitehall study, a 25year follow-up for mortality experience of 18,006 men with data on post-challenge blood glucose and self-reported diabetes, allowed us to address these issues. Results: There were 2158 cancer deaths at follow-up. Of the 15 cancer outcomes, diabetes status was positively associated with mortality from carcinoma of the pancreas and liver, while the relationship with lung cancer was inverse, after controlling for a range of potential covariates and mediators which included obesity and socioeconomic position. After excluding deaths occurring in the first 10years of follow-up to examine the effect of reverse causality, the magnitude of the relationships for carcinoma of the pancreas and lung was little altered, while for liver cancer it was markedly attenuated. Conclusions: In the present study, diabetes status was related to pancreatic, liver, and lung cancer risk. Cohorts with serially collected data on blood glucose and covariates are required to further examine this area.

Taking folate in pregnancy and risk of maternal breast cancer.

Abstract: Taking folate before conception and then for the first three months of pregnancy reduces the risk of recurrence of neural tube defects,1 and fortification of food has been proposed. The effects of long term exposure to high concentrations of supplemental folate are unknown, and antimetabolite effects are theoretically possible.2 Data on the long term effects of increased folate intake in pregnancy are limited. We followed up a large trial of folate supplementation in pregnancy from the 1960s.3 4 We examined the association between folate status and death, and we also analysed the effects of folate supplementation. From June 1966 to June 1967, 3187 women were identified as potentially eligible for a trial of folate supplementation.3 4 At her booking visit, the mother's age, gestation, parity, weight, and blood pressure were recorded, and blood was taken to …

Mis)use of Factor Analysis in the Study of Insulin Resistance Syndrome

Abstract: Over the last decade, factor analysis has been used increasingly to describe patterns of simultaneous occurrence of the central components of the insulin resistance syndrome. In this paper, the authors describe factor analysis, review studies that have used factor analysis to examine the insulin resistance syndrome, and explore how factor analysis might be used to increase our understanding of this syndrome. Most studies that they reviewed gave vague reasons for using factor analysis and did not demonstrate an understanding of the use and limitations of this statistical method. Confirmatory factor analysis based on sound theoretical concepts and a clear understanding of the statistical methods may provide some insights into the pathophysiology of the syndrome. However, to date none of the studies has adopted this approach, and other statistical approaches and study designs are likely to provide greater understanding of the syndrome.

Commentary: The hormone replacement–coronary heart disease conundrum: is this the death of observational epidemiology?

Abstract: Under its definition for the word ‘hindsight’ the Oxford English Dictionary includes the following statement ‘hindsight is always better than foresight’ (http://dictionary.oed.com/), and the slogan of a private survey and evaluation company, ingeniously called Hindsight, is ‘remember hindsight is always 20/20!’ (http://www. hndsight.com/). We have the benefit of the ‘hindsight’ from randomized controlled trials (RCT) when we comment on this meta-analysis of observational studies, but whether the conflicting results between the trial and observational evidence on the association between hormone replacement therapy (HRT) use and coronary heart disease (CHD) will lead to 20/20 vision remains to be seen. The disparity between findings from observational studies and RCT of the effects of HRT on CHD,1–4 has created considerable debate among researchers, practitioners and postmenopausal women. The authors of the meta-analysis reprinted in this issue of the International Journal of Epidemiology concluded that the pooled estimate of effect from the best quality observational studies (internally controlled prospective and angiographic studies) inferred a relative reduction of 50% with ever use of HRT and stated that ‘overall, the bulk of the evidence strongly supports a protective effect of estrogens that is unlikely to be explained by confounding factors’.4 By contrast, recent randomized trials among both women with established CHD and healthy women have found HRT to be associated with slightly increased risk of CHD or null effects.1,2 For example, the large Women’s Health Initiative (WHI) randomized trial found that the hazards ratio for CHD associated with being allocated to combined HRT was 1.29 (95% CI: 1.02, 1.63), after 5.2 years of follow-up.1 These marked differences between observational findings and trials are important for two reasons. First, and foremost, is the clinical impact. As another commentator on the same subject remarked: ‘Does HRT decrease or increase the risk of heart disease? At least every woman, every gynecologist and every primary care doctor want to know the “correct” answer.’5 Second, is the broader implication for observational epidemiology. Prior to the publication of the WHI it was suggested that well conducted observational studies produced similar estimates of treatment effects as RCT, and that the notion of a hierarchy of evidence with the RCT on top could not be supported.6,7 The differing results between observational studies and RCT in the association between HRT and CHD throw this idea into question and may signify the death of observational epidemiology.8 It is important, therefore, to determine why the results from the trials and observational studies are so different. A number of explanations have been suggested for these disparities. Whilst some have suggested that the results of the trials were biased because of contamination, and in the case of the WHI, early termination of the arm assessing the effect of combined HRT, the consistency across a number of trials of a null effect make these explanations unlikely. More plausible explanations are that women who participated in the trials were importantly different from those who participated in the observational studies, or that the observational study results were confounded.

Childhood mental ability and blood pressure at midlife: Linking the Scottish Mental Survey 1932 and the Midspan studies

Abstract: Objectives To establish the relationship between childhood mental ability and adult hypertension.Design Retrospective cohort study.Setting Community.Participants Non-clinical sample of people born in 1921 who participated in both the Scottish Mental Survey 1932 and the Midspan studies. Nine hundred

Inverse Association Between Birth Weight and C-Reactive Protein Concentrations in the MIDSPAN Family Study

Abstract: Objective— Inflammation markers and low birth weight each predict elevated risk of cardiovascular events and type 2 diabetes. However, potential associations between the low-grade inflammatory resp...

The Aberdeen Children of the 1950s cohort study: background, methods and follow-up information on a new resource for the study of life course and intergenerational influences on health

Abstract: In this paper we introduce and describe in detail an addition to the UK's population-based resources for the investigation of biological and social influences on health across the life course and between generations: the Aberdeen Children of the 1950s study. We also provide an account of postwar Aberdeen when study members were growing up, report on findings of analyses of data from the original survey on which this study is based and its follow-up, assess the strengths and limitations of the study, and outline current and future research directions. This cohort comprises individuals born in Aberdeen, Scotland (UK) between 1950 and 1956, and is derived from 15 thousand subjects who took part in the Aberdeen Child Development Survey, a cross-sectional study of 'mental subnormality' (learning disability) in a population of all children who were attending Aberdeen primary schools in December 1962. Data collection included information on birthweight, gestational age, childhood height and weight, tests of cognition and behavioural disorder, and a range of multilevel socio-economic indicators. In 1998 we began the process of revitalising this cohort (now termed the Aberdeen Children of the 1950s study). We have been successful in ascertaining the current vital status and whereabouts of 98.5% of a target population of 12 150 subjects (6276 males, 5874 females) with full baseline data. The large majority (81%) of study participants still reside in Scotland and many (73%) have remained in the Grampian region which incorporates Aberdeen. At the present time, a total of almost 500 subjects are known to have died. Linkages to hospital admissions and other health endpoints captured through the Scottish Morbidity Records system have been completed. This includes an intergenerational linkage to approximately eight thousand deliveries in Scotland occurring to female members of the study population. A postal questionnaire to all traced surviving cohort members has also been distributed.

Maternal and personal cigarette smoking synergize to increase airflow limitation in adults.

Abstract: Susceptibility of the lungs to cigarette smoke is poorly understood. It is not known whether maternal smoking increases chronic obstructive pulmonary disease (COPD) risk. In 1998 we reported an inverse association between maternal smoking (prerecorded) and FEV(1) in adults. Because FEV(1) and FVC are strongly correlated, it is unclear whether the association in question reflects a link with lung volume, airflow limitation, or both. We extended our original analysis to investigate whether maternal and personal smoking synergize to increase airflow limitation. We estimated residual FEV(1) to express FEV(1) variation that was not associated with FVC. Maternal smoking was inversely associated with FVC and FEV(1) irrespective of personal smoking. It was inversely associated with FEV(1)/FVC, forced midexpiratory flow rates (FEF(25-75) [mean forced expiratory flow during the middle half of the FVC], FEF(25-75)/FVC), and residual FEV(1) in current smokers but not in never or former smokers (heterogeneity p = 0.016, 0.024, 0.021, and 0.016, respectively). We tested the clinical relevance of findings in ever smokers without asthma: 10 cigarettes/day maternal smoking increased prevalent COPD by 1.7 (95% confidence interval: 1.2-2.5) after adjustment for covariates. Maternal smoking impairs lung volume irrespective of personal smoking and appears to synergize with personal smoking to increase airflow limitation and COPD.

Parents' growth in childhood and the birth weight of their offspring.

Abstract: Background A person's birth weight is inversely related to both their own and their parents' cardiovascular disease mortality risk, but mechanisms underlying such transgenerational associations are unclear. We investigated the influence of the childhood growth of the mother or father on the birth weight of their first-born offspring. Methods We used data from the long-term follow up (in 1997-1998) of 4999 children from 1352 families who participated in the Boyd Orr Survey of Diet and Health in Pre-War Britain (1937-1939). Complete information on childhood height, potential confounding variables, and the birth weight of first-born offspring was available for 637 subjects. Results Mother's height in childhood was positively associated with her offspring's birth weight. Leg length, but not trunk length, was the component of maternal height associated with offspring birth weight. For each unit increase in z-score for maternal childhood leg length, there was a 96-g (95% confidence interval = 6-186) increase in offspring birth weight after controlling for childhood socioeconomic variables and adult height. There were weaker positive associations of paternal height and leg length in childhood with offspring birth weight. Associations were not confounded by maternal birth weight or midgrandparental height. Conclusions Our findings are consistent with the hypothesis that maternal growth during childhood influences offspring birth weight, independently of maternal birth weight, final attained height, or midgrandparental height. Because leg length is a sensitive marker of adverse nutritional and social exposures during childhood, these results suggest a key role for a mother's early environmental exposures as a determinant of her child's subsequent health.

The association of the PON1 Q192R polymorphism with coronary heart disease: findings from the British Women's Heart and Health cohort study and a meta-analysis

Abstract: There have been inconsistent results from case-control studies assessing the association of the PON1 Q192R polymorphism with coronary heart disease (CHD). Most studies have included predominantly men and the association in women is unclear. Since lipid levels vary between the sexes the antioxidant effect of PON1 and any genes associated with it may also vary by sex. We have examined the association of the PON1 Q192R polymorphism with CHD in a large cohort of British women and combined the results from our cohort study with those from all other published studies. The distribution of genotypes was the same among women with CHD and those without disease. The odds ratio (95% confidence interval) of having CHD comparing those with either the QR or RR genotype to those with QQ genotype (dominant model of association) was 1.03 (0.89, 1.21) and the per allele odds ratio was 0.98 (0.95, 1.01). In a meta-analysis of this and 38 other published studies (10,738 cases and 17,068 controls) the pooled odds ratio for the dominant effect was 1.14 (1.08, 1.20) and for the per allele effect was 1.10 (1.06, 1.13). There was evidence of small study bias in the meta-analyses and the dominant effect among those studies with 500 or more cases was 1.05 (0.96, 1.15). Ethnicity and reporting of whether the genotyping was done blind to the participants clinical status also contributed to heterogeneity between studies, but there was no difference in effect between studies with 50% or more women compared to those with fewer women and no difference between studies of healthy populations compared to those at high risk (with diabetes, renal disease of familial hypercholesterolaemia). There is no robust evidence that the PON1 Q192R polymorphism is associated with CHD risk in Caucasian women or men.

Breastfeeding and cardiovascular mortality: the Boyd Orr cohort and a systematic review with meta-analysis

Abstract: Aims To investigate the association of breastfeeding with all-cause, cardiovascular, and ischaemic heart disease mortality. Methods and results A long-term follow-up of 4999 children originally surveyed from 1937 to 1939 was undertaken (Boyd Orr cohort). Four thousand three hundred and seventy-nine subjects (88%) were traced in adulthood and 3555 (71%) had complete data on all covariates. The results were combined with a meta-analysis of the published literature. In the Boyd Orr study, there was little evidence that breastfeeding was associated with all-cause (hazard ratio: 1.04 [95% CI: 0.90–1.20]), cardiovascular (1.04 [0.83–1.30]), or ischaemic heart disease (1.02 [0.77–1.36]) mortality, compared with bottle-feeding. Meta-analyses of observational studies showed little evidence of an association of breastfeeding with all-cause (pooled rate ratio: 1.01 [95% CI: 0.91–1.13]) or cardiovascular (1.06 [0.94–1.20]) mortality. There was a moderate-to-high degree of between-study heterogeneity for the association between breastfeeding and ischaemic heart disease mortality (\batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(I^{2}\) \end{document} value—indicating the degree of between-study variation attributable to heterogeneity—66%), and estimates were consistent with both an important beneficial or adverse effect of breastfeeding. Conclusion There is little consistent evidence that breastfeeding influences subsequent all-cause or cardiovascular disease mortality. Results from other well-designed cohorts may clarify residual uncertainty.

Association Between Childhood Socioeconomic Status and Coronary Heart Disease Risk Among Postmenopausal Women: Findings From the British Women's Heart and Health Study

Abstract: Objectives We assessed the association between childhood socioeconomic status (SES) and coronary heart disease among postmenopausal women Methods We conducted a cross-sectional analysis of 3444 women aged 60 to 79 years Results There was an independent linear association between childhood and adult SES and coronary heart disease The association between childhood SES and coronary heart disease was attenuated when we adjusted for insulin resistance syndrome, adult smoking, physical activity, biomarkers of childhood nutrition, and passive smoking Conclusions The association between adverse childhood SES and coronary heart disease is in part mediated through insulin resistance, which may be influenced by poor childhood nutrition, and in part through the association between childhood SES and adult behavioral risk factors

Commentary: Social capital, social epidemiology and disease aetiology

Abstract: The role of social capital in the production of health has developed over recent years into a major academic concern, and is now beginning to feed through into policy discussions concerning the determinants of population health. Social capital has, of course, had greater resonance in fields such as development economics than it has so far had in health, but the confluence of these two threads is now marked. This is made clear by the work of the leading popularizer of social capital—Robert Putnam—who in his seminal 1993 book Making Democracy Work 1 explicitly states that health should not be considered an outcome of social capital, saying that: we must be careful not to give governments credit (or blame) for matters beyond their control. In the language of policy analysis, we want to measure ‘outputs’ rather than ‘outcomes’—health care rather than mortality rates … Health depends on factors like diet and lifestyle that are beyond the control of any democratic government. 1 Only 7 years later he had dramatically reversed his opinion and decided that: Of all the domains in which I have traced the consequences of social capital, in none is the importance of social connectedness so well established as in the case of health and well-being’. 2 The explosion of interest in social capital has not, as yet, led to greater clarity in the conceptualization of exactly what the term social capital refers to and how the supposed connections with health are generated and maintained. 3,4 We therefore welcome Simon Szreter and Michael Woolcock’s 5 clear and persuasive formulation and think that it will—rightly—become a touchstone for ongoing debates regarding social capital in the health field. Other commentators in this issue of the International Journal of Epidemiology 6‐10 have raised a variety of important points and we will not duplicate these here. Instead we would like to elaborate some specific aspects of the epidemiological interpretation of the historical evidence. This leads to more general considerations of how to interpret social influences on population health. In the central section of their article Szreter and Woolcock discuss mortality crises in 19th century Britain and date improvements in mortality to the 1870s and 1880s, as Simon Szreter has done previously. 11 They develop one specific case, that of Birmingham and the role of Joseph Chamberlain in preaching the ‘civic gospel’ of gas and water socialism over this period. In particular they consider that Chamberlain’s activities were central to the development of linking and bridging social capital which protected civil society in Birmingham from the usual nepotism and corruption that sank other British cities in the central decades of the 19th century. Without Chamberlain’s activities the rapid urbanization that translated itself into Simon Szreter’s four Ds—disruption, deprivation disease and death 12 —would have continued to bear its consequences. We will argue that consideration of age-specific mortality trends during the second half of 19th century shows that the social capital mechanism suggested by Szreter and Woolcock as being crucial, is an incomplete explanation. We will also address the same specific historical case raised by Szreter and Woolcock and illustrate the negative externalities—in this case international—that can result from the deployment of particular forms of social capital. Our main point is to highlight the need to consider specific, biologically plausible and wellsupported aetiological mechanisms when attempting to map aspects of the social environment onto health outcomes. 13

Parental growth at different life stages and offspring birthweight: an intergenerational cohort study.

Abstract: Using three generations of the 1958 British national birth cohort we investigated ways in which parental size is related to offspring birthweight. By age 41 years, 4566 singleton female and 4050 male cohort members (born 3-9 March, 1958) had become parents and provided information on their singleton offspring. Mother's birthweight (standardised for gestational age and sex) was the strongest determinant of offspring birthweight (effect size [ES] per SDS 112 g [95% CI 97, 128]), which was little affected by adjustment for maternal height or BMI (ES 95 g and 105 g respectively). The intergenerational birthweight association was not observed for mothers born very small or large. Mother's childhood height at age 7 (ES 46 g [24, 67]), but not BMI (ES 3 g [-18, 23]), was associated with offspring birthweight after adjustment for grandparental size, own birthweight, and adult size. Controlling for other growth measures strongly attenuated the association between mother's adult height and offspring birthweight: (ES 90 g, unadjusted, and 25 g, adjusted), while the association between adult BMI and offspring birthweight was little affected (ES 55 g and 51 g respectively). Father's BMI did not affect offspring birthweight, while the associations for height were similar, albeit weaker, than those observed for the mother. Our results suggest that intergenerational associations in birthweights are largely independent of postnatal size. Maternal height in childhood was positively related to offspring birthweight, while the effect of her BMI was restricted to adulthood.