Feelings of hopelessness are prospectively associated with increased risk of death, progression of atherosclerosis and other health outcomes. Places as well as people may promote a sense of hopelessness. We used the Chicago Community Adult Health Study to investigate whether feelings of hopelessness cluster at the neighborhood level. Random-intercept logistic models were used to examine associations of hopelessness with neighborhood conditions (physical disorder and decay, perceived violence and disorder, social cohesion) and census-based measures of neighborhood socioeconomic conditions (poverty, unemployment, % high school dropouts) from 1980–2000. Of the 3074 participants, 459 were categorized as experiencing hopelessness. Greater physical disorder and perceived disorder and a higher unemployment rate were associated with increased odds of hopelessness. Individuals’ reports of hopelessness reflect physical, socioeconomic, and opportunity characteristics of their neighborhoods above and beyond their individual characteristics. Changing opportunity structures in a neighborhood may play a key role in reducing individuals’ feelings of hopelessness.
hopelessness; neighborhood environments; neighborhood socioeconomic conditions; disorder; unemployment
Dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis is hypothesized to be an important pathway linking socioeconomic position and chronic disease.
This paper tests the association between education and the diurnal rhythm of salivary cortisol.
Up to 8 measures of cortisol (mean of 5.38 per respondent) over two days were obtained from 311 respondents, aged 18–70, drawn from the 2001–2002 Chicago Community Adult Health Study. Multi-level models with linear splines were used to estimate waking level, rates of cortisol decline, and area-under-the-curve over the day, by categories of education.
Lower education (0–11 years) was associated with lower waking levels of cortisol, but not the rate of decline of cortisol, resulting in a higher area-under-the-curve for more educated respondents throughout the day.
This study found evidence of lower cortisol exposure among individuals with less education and thus does not support the hypothesis that less education is associated with chronic over-exposure to cortisol.
There is a need for more Comparative Effectiveness Research (CER) to strengthen the evidence base for clinical and policy decision-making. Effectiveness Guidance Documents (EGD) are targeted to clinical researchers. The aim of this EGD is to provide specific recommendations for the design of prospective acupuncture studies to support optimal use of resources for generating evidence that will inform stakeholder decision-making.
Document development based on multiple systematic consensus procedures (written Delphi rounds, interactive consensus workshop, international expert review). To balance aspects of internal and external validity, multiple stakeholders including patients, clinicians and payers were involved.
Recommendations focused mainly on randomized studies and were developed for the following areas: overall research strategy, treatment protocol, expertise and setting, outcomes, study design and statistical analyses, economic evaluation, and publication.
The present EGD, based on an international consensus developed with multiple stakeholder involvement, provides the first systematic methodological guidance for future CER on acupuncture.
Comparative effectiveness research; Effectiveness guidance document; Acupuncture
Biomarkers are an important aspect of research linking psychosocial stress and health. This paper aims to characterize the biological pathways that may mediate the relationship between socioeconomic position (SEP) and cardiovascular disease (CVD) and address opportunities for further research within the Panel Study for Income Dynamics (PSID), with a focus on psychosocial stressors related to SEP. We review the literature on CVD biomarkers, including adhesion and proinflammatory molecules (IL-6, other cytokines, C-reactive proteins, fibrinogen, etc.) and microbial pathogens. The impact of socioeconomic determinants and related psychosocial stressors on CVD biomarkers mediated by behavioral and central nervous system pathways are described. We also address measurement and feasibility issues including: specimen collection methods, processing and storage procedures, laboratory error, and within-person variability. In conclusion, we suggest that PSID consider adding important assessments of specific CVD biomarkers and mediating behavioral measures, health, and medications that will ultimately address many of the gaps in the literature regarding the relationship between socioeconomic position and cardiovascular health.
Background Literature on the socio-economic ‘gradient’ in health often asserts that income is associated with better health not only for the very poor, but also across the entire income distribution. In addition, changes in the shape of the association between incomes during a period of increasing economic inequality have not been previously studied. The goal of the current study was to estimate and compare the shape of the relationship between income and mortality in the USA for the 1970s, the 1980s and the 1990s.
Methods Using income and mortality data from the Panel Study of Income Dynamics for respondents aged 35–64 years, we used a Bayesian Cox Model with regression splines to model the risk of mortality over three 10-year follow-up periods. To identify whether income was more strongly associated with mortality at different parts of the income distribution, we treated income as a linear spline with an unknown knot location.
Results The shape of the association between income and mortality was quite non-linear, with a much stronger association at lower levels of income. The relationship between income and mortality in the USA was also not invariant across time, with the increased risk of death associated with lower income applying to an increasing proportion of the US population over time (9th percentile of income in 1970–79, 20th percentile in 1980–89 and 32nd percentile in 1990–99).
Conclusions Our analyses do not support the claim that income is associated with mortality throughout the income distribution, nor is the association between income and mortality the same across periods. Based on our analyses, a focus on the bottom 30% of the income distribution would seem to return the greatest benefits in reducing socio-economic inequalities in health.
Depression and cardiovascular disease (CVD) are closely associated, but the mechanisms underlying this connection are unclear. Regardless of the low cholesterol levels observed in depression, a small particle size of low-density lipoproteins (LDL), as well as elevated apolipoprotein B (ApoB) levels, are related to increased CVD risk, even when levels of LDL cholesterol are low. We examined the association between elevated depressive symptoms and compositional changes in serum LDL particles in a sample of 2456 middle-aged Finnish men. Depressive symptoms were assessed with the 18-item Human Population Laboratory Depression Scale, and the study population was divided into two groups (elevated depressive symptoms, n=269; non-depressed, n=2187). The levels of serum total cholesterol (TC), low-and high-density lipoprotein cholesterol (LDL-C, HDL-C), triglycerides (TG), and ApoB were determined. The LDL-C/ApoB ratio, a marker of compositional changes in LDL particle size, was calculated. The group with elevated depressive symptoms had lowered levels of serum TC (p=0.028) and LDL-C (p=0.008). No differences were observed in the LDL-C/ApoB ratio. The likelihood for belonging to the group with elevated depressive symptoms increased 10% for each 0.5 mmol/L decrease in the levels of TC (p=0.002) or LDL-C (p=0.001) in regression models adjusted for age, examination years, marital and socioeconomic status, energy expenditure, body mass index, CVD history, alcohol consumption, smoking, and the use of lipid-lowering, antidepressant and antipsychotic medications. Our findings suggest that greater small-particle LDL levels are not associated with depression, and are thus unlikely to underlie the association between cardiovascular risk and depression.
Apolipoprotein B; Cholesterol; Depression; Low-density lipoprotein cholesterol
Socioeconomic position (SEP) has been shown to be related to obesity and weight gain, especially among women. It is unclear how different measures of socioeconomic position may impact weight gain over long periods of time, and whether the effect of different measures vary by gender and age group. We examined the effect of childhood socioeconomic position, education, occupation, and log household income on a measure of weight gain using individual growth mixed regression models and Alameda County Study data collected over thirty four years(1965–1999).
Analyses were performed in four groups stratified by gender and age at baseline: women, 17–30 years (n = 945) and 31–40 years (n = 712); men, 17–30 years (n = 766) and 31–40 years (n = 608).
Low childhood SEP was associated with increased weight gain among women 17–30 (0.13 kg/year, p < 0.001). Low educational status was associated with increased weight gain among women 17–30 (0.14 kg/year, p = 0.030), 31–40 (0.14 kg/year, p = 0.014), and men 17–30 (0.20 kg/year, p = 0.001).
Log household income was inversely associated with weight gain among men 31–40 (−0.10 kg/yr, p = 0.16). Long-term weight gain in adulthood is associated with childhood SEP and education in women and education and income in men.
This study examines whether the psychological traits of hopelessness and depressive symptoms are related to endothelial dysfunction.
Data come from a subsample of 434 respondents in the 2001–2003 Chicago Community Adult Health Study (CCAHS), a population-based survey designed to study the impact of psychological attributes, neighborhood environment, and socio-economic circumstances on adults age 18 and over. Circulating biomarkers of endothelial dysfunction including e-selectin, p-selectin and s-ICAM1 were obtained from serum samples. Hopelessness was measured by responses to two questions and depressive symptoms were measured by an 11-item version of the CES-D. Multivariate regression models tested whether continuous levels of the biomarkers (natural log transformed) were associated with levels of hopelessness and depressive symptoms separately and concurrently.
In age- and sex-adjusted models, hopelessness showed significant positive linear associations with s-ICAM1. In contrast, there was no significant linear association between hopelessness and e-selectin and p-selectin. Adjustment for clinical risk factors including systolic pressure, chronic health conditions, smoking, and body mass index did not substantively alter these associations. Results from similar models for depressive symptoms did not reveal any association with the three biomarkers of endothelial dysfunction. The associations between hopelessness and e-selectin and s-ICAM1 were robust to the inclusion of adjustments for depressive symptoms.
Negative psychosocial traits may influence cardiovascular outcomes partially through their impact on the early stages of atherosclerosis, and specific psychosocial traits such as hopelessness may play a more direct role in this process than overall depressive symptoms.
Lower socioeconomic position (SEP) is related to higher prevalence of Type 2 diabetes, yet little is known about the relationship of SEP with incident diabetes.
The association between SEP, measured by self-reported education, income, and occupation, and Type 2 diabetes incidence was examined in a community sample of 6147 diabetes-free adults from Alameda County, CA. Cox proportional hazards models estimated the effect of baseline (1965) and time-dependent (value changes over time) measures of SEP on incident diabetes over a 34-year study period (1965–99). Demographic confounders (age, gender, race, and marital status) and potential components of the causal pathway (physical inactivity, smoking, alcohol consumption, body composition, hypertension, depression, and health care access) were included as fixed or time-dependent covariates.
Education, income, and occupation were associated with increased diabetes risk in unadjusted models. In baseline models adjusted for demographics, respondents with <12 years of education had 50% excess risk compared with those with more education [hazard ratio (HR) = 1.5, 95% confidence interval (95% CI) 1.11–2.04], but income and occupation were no longer significantly associated with increased risk. Further adjustment minimized the significance of all associations. Time-dependent effects were consistently elevated for low education and male blue-collar occupation, but non-significant after full adjustment (HR = 1.1, 95% CI 0.79–1.47 and HR = 1.3, 95% CI 0.91–1.89, respectively).
Socioeconomic disadvantage, especially with low educational attainment, is a significant predictor of incident Type 2 diabetes, although associations were largely eliminated after covariate adjustment. Obesity and overweight appear to mediate these associations.
Socioeconomic factors; Type 2 diabetes mellitus; incidence
Self-esteem is considered to be importantly associated with both psychosocial states such as depression as well as physical health. There are no population-based studies that examine the association between self-esteem and mortality. The objective of this study was to assess whether low self-esteem was prospectively associated with increased risk of death in a population-based sample of Finnish men.
A sample of 2682 male residents of Kuopio, Finland were interviewed and followed prospectively as part of the Kuopio Ischemic Heart Disease Risk Factor Study (KIHD). Characteristics of the KIHD sample at baseline included self-esteem, measured by the Rosenberg ten-item scale, socioeconomic factors, behavioral risk factors, other psychosocial characteristics, and prevalent diseases. Mortality was ascertained through linkage to the Finnish national death registry. We assessed the relationship between self-esteem and all-cause mortality using Cox proportional hazards models.
Low self-esteem was associated with a two-fold [hazard ratio (HR) = 2.0, 95% confidence interval (CI) = 1.3–3.2] increase in age-adjusted mortality. This relationship was partially explained by behavioral and socioeconomic factors, and prevalent diseases, and fully explained by other psychosocial characteristics (hopelessness, depression, cynical hostility, and sullenness). When adjusted for hopelessness alone there was no increased risk associated with low self-esteem (HR = 1.3, 95% CI = 0.8–2.2).
This study found no association between self-esteem and all-cause mortality after adjustment for other psychosocial characteristics, primarily hopelessness. Our understanding of the observed relationships between some psychosocial factors and mortality may be improved by simultaneous measurement of multiple psychosocial domains, thus diminishing the potential for residual confounding.
Self-esteem; Mortality; Psychosocial Factors; Socioeconomic Factors
This study examined associations between several lifecourse socioeconomic position (SEP) measures (childhood SEP, education, income, occupation) and diabetes incidence from 1965–1999 in a sample of 5,422 diabetes-free black and white participants in the Alameda County Study.
Race-specific Cox proportional hazard models estimated diabetes risk associated with each SEP measure. Demographic confounders (age, gender, marital status) and potential pathway components (physical inactivity, body composition, smoking, alcohol consumption, hypertension, depression, health care access) were included as covariates.
Diabetes incidence was 2-fold greater for blacks than whites. Diabetes risk factors independently increased risk, but effect sizes were greater among whites. Low childhood SEP elevated risk for both racial groups. Protective effects were suggested for low education and blue-collar occupation among blacks, but these factors increased risk for whites. Income was protective for whites, but not blacks. Covariate adjustment had negligible effects on associations between each SEP measure and diabetes incidence for both racial groups.
These findings suggest an important role for lifecourse SEP measures in determining risk of diabetes, regardless of race, and net of factors that may confound or mediate these associations.
Identifying biological and behavioural causes of diseases has been one of the central concerns of epidemiology for the past half century. This has led to the development of increasingly sophisticated conceptual and analytical approaches focused on the isolation of single causes of disease states. However, the growing recognition that (i) factors at multiple levels, including biological, behavioural and group levels may influence health and disease, and (ii) that the interrelation among these factors often includes dynamic feedback and changes over time challenges this dominant epidemiological paradigm. Using obesity as an example, we discuss how the adoption of complex systems dynamic models allows us to take into account the causes of disease at multiple levels, reciprocal relations and interrelation between causes that characterize the causation of obesity. We also discuss some of the key difficulties that the discipline faces in incorporating these methods into non-infectious disease epidemiology. We conclude with a discussion of a potential way forward.
Agent-based modelling; dynamic systems modelling; epidemiology; regression
To measure the childhood and life course socioeconomic exposures of people born between 1871 and 1949, and then to estimate the probability of death between 1965 and 1994, the probability of functional limitation in 1994, and the combined probability of dying or experiencing functional limitation during this period.
Setting, participants and design
Data were from the Alameda County Study (California) and pertained to people aged 17–94 years (n = 6627) in 1965 (baseline). Socioeconomic position (SEP) in childhood was based on respondent's reports of their father's occupation, and life course disadvantage was measured by cross‐classifying childhood SEP and the respondent's education and household income in 1965. The health outcomes were all‐cause mortality (n = 2420) and functional limitation measured using the Nagi index (n = 453, 17.4% of those alive in 1994). Relationships were examined before and after adjustment for changed socioeconomic circumstances after 1965.
Those from a low SEP in childhood, and those exposed to a greater number of episodes of disadvantage over the life course before 1965, were subsequently more likely to die, to report functional limitation and to experience the greatest health‐related burden.
All‐cause mortality, functional limitation and overall health‐related burden in middle and late adulthood are shaped by socioeconomic conditions experienced during childhood and cumulative disadvantage over the life course. The contributions made to adult health by childhood SEP and accumulated disadvantage suggest that each constitutes a distinct socioeconomic influence that may require different policy responses and intervention options.
A social epidemiologic perspective considers factors at multiple levels of influence (e.g., social networks, neighborhoods, states) that may individually or jointly affect health and health behaviour. This provides a useful lens through which to understand the production of health behaviours in general, and drug use in particular. However, the analytic models that are commonly applied in population health sciences limit the inference we are able to draw about the determination of health behaviour by factors, likely interrelated, across levels of influence. Complex system dynamic modelling techniques may be useful in enabling the adoption of a social epidemiologic approach in health behaviour and drug use research. We provide an example of a model that aims to incorporate factors at multiple levels of influence in understanding drug dependence. We conclude with suggestions about future directions in the field and how such models may serve as virtual laboratories for policy experiments aimed at improving health behaviour.
population health; multilevel; regression modeling; social epidemiology
Considerable evidence indicates that income and other measures of socioeconomic position are associated with a wide variety of health outcomes, there are few studies that prospectively examine the association between socioeconomic position over decades and health outcomes. The present study, covering almost three decades of the life course, examined the cumulative impact of different income measures on psychological well-being among adults.
We used data collected over 29 (1965–94) years from Alameda County Study participants to study the association between average income, income changes, profit and benefit incomes—and five scales of psychological well-being—Purpose in Life, Self-acceptance, Personal Growth, Environmental Mastery, and Autonomy. In age-adjusted models, the psychological well-being measures were each regressed on each of the income measures. Potential confounders (sex, education, race/ethnicity, social isolation, depression and perceived health) were also examined.
Mean income over almost three decades was strongly associated with all five scales of psychological well-being. Psychological well-being increased with the number of waves in which profit income was reported and with income increases over time. For all scales except Autonomy, psychological well-being decreased with the number of waves receiving need-based benefit and with decreasing income over time.
Psychological well-being may reflect the accumulation of socioeconomic advantage and disadvantage over decades.
Psychological well-being; socioeconomic factors; quality of life; life course
The spatial segregation of the U.S. population by socioeconomic position and especially race-ethnicity suggests that the social contexts or “neighborhoods” in which people live may substantially contribute to social disparities in hypertension. The Chicago Community Adult Health Study did face-to-face interviews, including direct measurement of blood pressure, with a representative probability sample of adults in Chicago. These data were used to estimate socioeconomic and racial-ethnic disparities in the prevalence, awareness, treatment, and control of hypertension, and to analyze how these disparities are related to the areas in which people live. Hypertension was significantly negatively associated with neighborhood affluence/gentrification, and adjustments for context eliminated the highly significant disparity between blacks/African-Americans and whites, and reduced the significant educational disparity by 10–15% to borderline statistical significance. Awareness of hypertension was significantly higher in more disadvantaged neighborhoods and in places with higher concentrations of blacks (and lower concentrations of Hispanics and immigrants). Adjustment for context completely eliminated blacks’ greater awareness, but slightly accentuated the lesser awareness of Hispanics and the greater levels of awareness among the less educated. There was no consistent evidence of either social disparities in or contextual associations with treatment of hypertension, given awareness. Among those on medication, blacks were only 40–50% as likely as whites to have their hypertension controlled, but context played little or no role in either the level of or disparities in control of hypertension. In sum, residential contexts potentially play a large role in accounting for racial-ethnic, and to a lesser degree, socioeconomic disparities in hypertension prevalence and, in a different way, awareness, but not in treatment or control of diagnosed hypertension.
Neighborhoods; social disparities; multi-level modeling; health inequalities; hypertension; blood pressure
Studies of the relationship between work time and health have been inconclusive. Consequently, we sought to examine the effect of work time on progression of atherosclerosis.
This prospective study of 621 middle-aged Finnish men evaluated effects of baseline and repeat measures of work time on 11-year progression of ultrasonographically assessed carotid intima-media thickness (IMT) and interactions with cardiovascular disease. Multiple linear regression models adjusted for 21 biological, behavioral, and psychosocial risk factors.
Working 3 (minimum), 5 (medium), or 7 (maximum) days per week at baseline was associated with 23%, 31%, and 40% 11-year increases in IMT, respectively. The relative change ratio (RCR) at maximum vs minimum was 1.14 for baseline days worked per week and 1.10 for hours worked per year of follow-up. Significant interactions existed between cardiovascular disease and work time. Men with ischemic heart disease (IHD) who worked the maximum of 14.5 hours per day experienced a 69% increase in IMT compared with a 29% increase in men without IHD. The RCR ratio for IHD (RCRIHD/RCRno IHD) was 1.44 for hours per day. Similarly, the RCR ratio for baseline carotid artery stenosis was 1.29 for hours per day and 1.22 for hours per year.
Increases in work time are positively associated with progression of carotid atherosclerosis in middle-aged men, especially in those with preexisting cardiovascular disease. Our findings are consistent with the hemodynamic theory of atherosclerosis.
Little is known about population determinants of short sleep duration. The authors examined associations between short sleep duration and income, education and race/ethnicity, and assessed changes over time in relative disparities.
Questionnaire data from the Alameda County Health and Ways of Living Study (ACS) was obtained at five time-points (1965, 1974, 1983, 1994, and 1999) for short sleep duration (<7 hours sleep per night). Household income, education level and race/ethnicity were assessed at baseline (n=6928). Odds ratios were computed to examine short sleep duration across income, education and race/ethnicity, adjusting for age, sex and time-varying covariates, and to assess changes over time.
Prevalence of short sleep at baseline was 15.2%.The (age-adjusted) odds of short sleep was increased for the lowest household income quintile (OR = 1.62, 95% CI: 1.34, 1.94), those with less than high school education (OR = 1.51, 95% CI: 1.30, 1.75), and among African Americans (odds ratio (OR) = 1.97, 95% confidence interval (CI): 1.68, 2.30). Relative disparities increased over time for African Americans and Hispanics compared to whites.
Socioeconomic position is a robust determinant of short sleep duration, even after adjusting for health-related characteristics linked to short sleep duration.
sleep; sleep deprivation; social class; socioeconomic factors; health behavior
The role of circulating levels of total homocysteine tHcy in the development of coronary heart disease (CHD) is still under debate. One reason for conflicting results between previous studies on homocysteine and heart diseases could be consequence of different interactions between homocysteine and genes in different study populations. Many genetic factors play a role in folate-homocysteine metabolism, like functional polymorphism (Val108Met) in the Catechol-O-methyltransferase (COMT) gene.
Methodology and Findings
Our aim was to examine the role of COMT Val158Met polymorphism and interaction of this polymorphism with serum tHcy and folate concentration on the risk of acute coronary and events in middle-aged men from eastern Finland. A population-based prospective cohort of 792 men aged 46–64 years was examined as part of the Kuopio Ischaemic Heart Disease Risk Factor Study. During an average follow-up of 9.3 years, there were 69 acute coronary events in men with no previous history of CHD. When comparing the COMT low activity genotype with the others, we found an age and examination year adjusted hazard rate ratio (HRR) of 1.73 (95% confidence interval (CI), 1.07–2.79), and an age, examination year, serum LDL and HDL cholesterol, and triglyceride concentration, systolic blood pressure and smoking adjusted HRR of 1.77 (95% CI, 1.05–2.77). Although serum tHcy concentration was not statistically significantly associated with acute coronary events (HRR for the highest third versus others 1.52, 95% CI, 0.93–2.49), subjects with both high serum tHcy and the COMT low activity genotype had an additionally increased adjusted risk of HRR 2.94 (95% CI 1.50–5.76) as compared with other men.
This prospective cohort study suggests that the functional COMT Val158Met polymorphism is associated with increased risk of acute coronary events and it may interact with high serum tHcy levels.
To compare the relation between mortality and income inequality in Canada with that in the United States.
The degree of income inequality, defined as the percentage of total household income received by the less well off 50% of households, was calculated and these measures were examined in relation to all cause mortality, grouped by and adjusted for age.
The 10 Canadian provinces, the 50 US states, and 53 Canadian and 282 US metropolitan areas.
Canadian provinces and metropolitan areas generally had both lower income inequality and lower mortality than US states and metropolitan areas. In age grouped regression models that combined Canadian and US metropolitan areas, income inequality was a significant explanatory variable for all age groupings except for elderly people. The effect was largest for working age populations, in which a hypothetical 1% increase in the share of income to the poorer half of households would reduce mortality by 21 deaths per 100 000. Within Canada, however, income inequality was not significantly associated with mortality.
Canada seems to counter the increasingly noted association at the societal level between income inequality and mortality. The lack of a significant association between income inequality and mortality in Canada may indicate that the effects of income inequality on health are not automatic and may be blunted by the different ways in which social and economic resources are distributed in Canada and in the United States.
Objective To assess the extent to which observed associations between income inequality and mortality at population level are statistical artifacts. Design Indirect “what if” simulation using observed risks of mortality at individual level as a function of income to construct hypothetical state-level mortality specific for age and sex as if the statistical artifact argument were 100% correct. Method Data from the 1990 census for the 50 US states plus Washington, DC, were used for population distributions by age, sex, state, and income range; data disaggregated by age, sex, and state from the Centers for Disease Control and Prevention were used for mortality; and regressions from the national longitudinal mortality study were used for the individual-level relation between income and risk of mortality. Results Hypothetical mortality, although correlated with inequality (as implied by the logic of the statistical artifact argument), showed a weaker association with the level of income inequality in each state than the observed mortality. Conclusions The observed associations in the United States at the state level between income inequality and mortality cannot be entirely or substantially explained as statistical artifacts of an underlying individual-level relation between income and mortality. There remains an important association between income inequality and mortality at state level above anything that could be accounted for by any statistical artifact. This result reinforces the need to consider a broad range of factors, including the social milieu, as fundamental determinants of health.
To assess the extent to which observed associations at population level between income inequality and mortality are statistical artefacts.
Indirect “what if” simulation by using observed risks of mortality at individual level as a function of income to construct hypothetical state level mortality specific for age and sex as if the statistical artefact argument were 100% correct.
Data from the 1990 census for the 50 US states plus Washington, DC, were used for population distributions by age, sex, state, and income range; data disaggregated by age, sex, and state from the Centers for Disease Control and Prevention were used for mortality; and regressions from the national longitudinal mortality study were used for the individual level relation between income and risk of mortality.
Hypothetical mortality, while correlated with inequality (as implied by the logic of the statistical artefact argument), showed a weaker association with states’ levels of income inequality than the observed mortality.
The observed associations in the United States at the state level between income inequality and mortality cannot be entirely or substantially explained as statistical artefacts of an underlying individual level relation between income and mortality. There remains an important association between income inequality and mortality at state level over and above anything that could be accounted for by any statistical artefact. This result reinforces the need to consider a broad range of factors, including the social milieu, as fundamental determinants of health.
Key messagesEvidence is accumulating that living in a society with higher inequality in income predisposes its members to higher mortality; at the same time, there is widespread evidence that, for individuals, higher income is protectiveThis individual level relation could “explain” the former societal level relationThe strength of observed levels of association between income inequality and mortality, however, may go well beyond what can be explained as a statistical artefact of an individual level relation between income and mortalityThe empirical analysis reported here, based on 1990 data for US states, suggests that the association between income inequality and mortality is considerably stronger than can be accounted for by any statistical artefactResearch underpinning public health policy should therefore take a broad view of the importance of the social milieu as a fundamental determinant of health