State-level variations in income-related inequality in health and health achievement in the US

doi:10.1016/j.socscimed.2006.01.011

Social Science & Medicine

Volume 63, Issue 2, July 2006, Pages 457-464

https://doi.org/10.1016/j.socscimed.2006.01.011 Get rights and content

Abstract

The objective of this study was to examine state-level variations in income-related inequality in health and overall health achievement in the US. Data that were representative of the US and each state in 2001 were extracted from the Current Population Survey 2001. Income-related inequality in health and health achievement were measured by Health Concentration and Health Achievement Indices, respectively. Significant variations were found across states in income-related inequality in health and health achievement. In particular, states in the south and east regions, on average, experienced a higher degree of health inequality and lower health achievement. About 80% of the state-level variation in health achievement could be explained by demographics, economic structure and performance, and state and local government spending and burden. In contrast, medical care resource indicators were not found to contribute to health achievement in states. States with better health achievement were more urbanized, had lower proportions of minority groups, females and the elderly, fewer individuals below the poverty line, larger primary industry, and lower unemployment rates. Also, per capita state and local government spending, particularly the proportion spent on public health, was positively associated with better health achievement. Because of the direct implications of health level and distribution in resource allocation and social norms, states with a lower level of health achievement need to prioritize efforts in increasing and reallocating resources to diminish health inequality and to improve population health.

Section snippets

Background

The positive correlation between poverty and poor health has been well documented in the literature. In studies of the US, income inequality was used to predict health (measured by several indicators such as mortality and prevalence of diseases) at the state level (Kawachi & Kennedy, 1997; Kennedy, Kawachi, & Prothrow-Stith, 1996). Over half of the variation in age-adjusted total mortality was explained by income inequality. Furthermore, one study using multi-level analysis found that

Measures of income-related inequality in health and health achievement

Although inequality in health can be generally measured by contrasting the health status of individuals who differ by socio-economic characteristics, such as educational achievement (Kunst & Mackenbach, 1994) and race/ethnicity, (Baker, Hoel, Mohr, Lipsitz, & Lackland, 2000; Keppel, Pearcy, & Wagener, 2002; Saftlas, Koonin, & Atrash, 2000; Stewart & Silverstein, 2002) income distribution has been predominantly used in the literature because it is the strongest predictor of the availability of

Results

The average health value of the population in the United States was 0.867, with the average health for the states ranging from 0.845 to 0.887. All states in the US including the District of Columbia had a positive HCI, indicating health inequalities favoring the well offs (Table 1). For the entire US population, the HCI had a value of 0.0166, with the states’ values ranging from 0.0072 to 0.0259. The higher the HCI value, the greater the degree of inequalities. All HCIs were significantly

Discussion

The investigation of health inequality suggests that income-related inequality in health exists in not only the developing countries, but also industrialized countries, among which the United States as a nation often ranks at the low end when income-related inequality in health is measured by HCI as in the current study (Humphries & van Doorslaer, 2000; van Doorslaer et al., 1997). It is possible that the lower rank of the United States compared to the other industrialized countries, in terms

References (33)

P. Baker et al.
Racial, age, and rural/urban disparity in cervical cancer incidence
Annals of Epidemiology
(2000)
P.M. Clarke et al.
On the measurement of relative and absolute income-related health inequality
Social Science and Medicine
(2002)
U.G. Gerdtham et al.
A note on validating Wagstaff and van Doorslaer's health measure in the analysis of inequalities in health
Journal of Health Economics
(1999)
K.H. Humphries et al.
Income-related health inequality in Canada
Social Science and Medicine
(2000)
N. Kakwani et al.
Socioeconomic inequalities in health: Measurement, computation, and statistical inference
Journal of Econometrics
(1997)
I. Kawachi et al.
The relationship of income inequality to mortality: Does the choice of indicator matter?
Social Science and Medicine
(1997)
J.P. Mackenbach et al.
Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe
Social Science and Medicine
(1997)
V. Mor et al.
Journal of Clinical Epidemiology
(1989)
M.J. Soobader et al.
Aggregation and the measurement of income inequality: Effects on morbidity
Social Science and Medicine
(1999)
E. van Doorslaer et al.
Income-related inequalities in health: Some international comparisons
Journal of Health Economics
(1997)

E. van Doorslaer et al.

Equity in the delivery of health care in Europe and the US

Journal of Health Economics

(2000)

A. Wagstaff et al.

On the measurement of inequalities in health

Social Science and Medicine

(1991)

A. Wagstaff

Inequality aversion, health inequalities and health achievement

Journal of Health Economics

(2002)

S. Chatterjee et al.

Regression analysis by example

(2000)

M.M. Farmer et al.

Distress and perceived health: Mechanisms of health decline

Journal of Health and Social Behavior

(1997)

K.F. Ferraro et al.

Health trajectories: Long-term dynamics among black and white adults

Journal of Health and Social Behavior

(1997)

Cited by (30)

Environmental pollution and socioeconomic health inequality: Evidence from China
2023, Sustainable Cities and Society
Recent work on socioeconomic health inequality has explored the possible determinants, but there is a lack of empirical evidence examining the associations between environmental pollution and socioeconomic health inequality. Most of studies employed the one-dimensional decomposition method, focusing on health but ignoring rank, to analyze the factors of health inequality. Hence, we fill this gap by using nationally representative individual-level data, employing the recentered influence function regressions method (a two-dimensional decomposition method), and estimating the marginal effect of environmental pollution on socioeconomic inequality in three health outcomes in China straightforward: physical discomfort, chronic disease, and self-rated health. Findings show that environmental pollution increases the income related health inequality of people, and the effect is larger for the middle-aged. The evidence of the possible channel through which environmental pollution is correlated with socioeconomic health inequality is also provided. In the face of environmental pollution, the rich pay more attention to their health condition and increase beneficial health behaviors, such as increasing sleeping hours, decreasing smoking, or having a more balanced diet, compared with the poor. It partly explains why environmental pollution widens the income related health inequality. These findings imply that reducing environmental pollution is an effective strategy for improving environmental quality and narrowing the income related health inequality.
Revealing mental health status in Iran's capital: Putting equity and efficiency together
2012, Social Science and Medicine
Citation Excerpt :
Adam Wagstaff first proposed the AI method (Wagstaff, 2002) in order to assess the status of 44 developing countries with regard to the three health variables of mortality in the under-five age group, child malnutrition and adult fertility rates. Following on from Wagstaff's seminal work, this method has been applied in several studies, including Ke Xu's study of general health in the US (Xu, 2006), Hernandez-Quevedo's investigation into physical disabilities across Europe (Hernandez-Quevedo,Jones, Lopez-Nicolas, & Rice, 2006), Olalekan Uthman's study of chronic childhood malnutrition in Nigeria (Uthman, 2009) and finally, Arokiasamy and Pradhan's analysis of children's health in India (Arokiasamy & Pradhan, 2010). In this paper, we applied the AI approach in an attempt to reveal mental health status in Iran's capital, Tehran.
Information about urban health is often based on averages, while to better understand health status in urban areas, inequality should also be included. In this paper, we applied an achievement index approach in order to surmount this defect and to examine mental health status in Iran's capital, Tehran. The data we required for this study were taken from the Urban Health Equity Assessment and Response Tool (Urban HEART) survey which was conducted in Tehran in 2007, covering people aged 15 and above. The concentration index, which is a commonly used measure of socioeconomic inequalities in health, was extended to enable the combination of inequality and averages and the formation of a mental health achievement index. Values from the standard concentration indices showed that mental disorders are concentrated disproportionately among the poor in Tehran. An extension of the standard concentration indices revealed that, in most of Tehran's districts, the mental health of populations in the poorest quintile is much worse than that of other groups. In addition, when we computed the achievement index and ranked districts according to this index, the ranking was different from the ranking by averages. These findings imply that mental health varies significantly across the economic groups of the population in Tehran and that efficiency-oriented strategies which target average level of mental health alone are not sufficient to improve mental health of all people especially mental health of the poor. Equity-oriented strategies which target the mental health inequalities should be considered as well.
Measuring inequalities in health: What do we know? What do we need to know?
2012, Health Policy
Citation Excerpt :
If ill-health declines monotonically with income, the greater the degree of inequality aversion, the greater the gap between the mean μ and the value of the index IHA(v). Several applications of the IHA have been provided in the literature; Hernández Quevedo et al. [13], who exploit the European Community Household Panel (ECHP) database, Meheus and van Doorslaer [32] study based on the Demographic and Health Surveys and Xu [33] using the US Current Population Survey, are some examples. Wagstaff et al. [34] show that the health concentration index can be decomposed into the contributions of individual factors to income-related health inequality, in which each contribution is the product of the sensitivity of health with respect to that factor and the degree of income-related inequality in that particular factor.
We argue that policy analysis aiming at curving inequalities in health calls for a better understanding of what we know about its measurement pathways. Assuming that health is a good that individuals trade off against other goods, unavoidable health inequalities result when after controlling for unavoidable factors (e.g., age and gender), differences in socioeconomic status of an individual systemically engender differences in health outcomes. However, the measurement of such inequality and underpinning reasons behind are not suggestive of a clear picture. In reviewing the literature, we conclude that it is unclear what the evidence suggests about the reasons for health inequalities as well as the best possible instruments to measure both inequality and socioeconomic health gradients. We provide an evaluation of the different sources of health inequity and we draw upon measurement issues and their policy significance.
Measuring achievement: Changes in risk factors for cardiovascular disease in Australia
2009, Social Science and Medicine
Citation Excerpt :
The AI combines both a measure of average morbidity (or health) and an absolute measure of inequality in morbidity (or health) into a single measure. The AI has been used to compare measures of child health across developing countries (Wagstaff, 2002), measures of self-reported health across states within a country (Xu, 2006), and indicators of health limitations within eight European countries across time (Hernandez-Quevedo, Jones, Lopez Nicolas, & Rice, 2006). To date, comparisons conducted using the AI, have been based on rankings of point estimates and have not taken into account uncertainty, or undertaken hypothesis testing using statistical methods.
A possible measure for evaluating health system performance is the achievement index, which can be calculated using prevalence and distribution of a health measure across different socioeconomic groups. This study extends this approach by examining how achievement can be represented on a two-dimensional plane with the x-axis being the difference in mean ill-health and the y-axis being the difference in an absolute measure of inequality based on the generalised concentration index. The achievement plane is an easily understandable visual aid which provides a method of tracking changes in health and inequality over time, as well as uncertainty around these measures. We also demonstrate how comparisons over time and at different levels of inequality aversion can be undertaken using measures of net achievement. To illustrate the use of the achievement plane, we compared changes in prevalence of various cardiovascular risk factors and absolute inequality in the distribution of these factors, using data from four successive Australian National Health Surveys conducted between 1989 and 2005. While self-reported rates of smoking and high cholesterol have been declining, inequalities have been rising as the greatest reductions in these risk factors have been among higher income groups. Conversely for risk factors where the prevalence has been increasing, health inequalities are either not changing (i.e. diabetes and obesity), or diminishing over time (overweight/obese). All of these changes can be summarized using an achievement plane and graphs of net achievement to examine changes in prevalence and distribution of these risk factors over time.
Calculating concentration index with repetitive values of indicators of economic welfare
2009, Journal of Health Economics
Citation Excerpt :
To compare the two variance estimates, we need a baseline variance estimate. Because there are no well-established analytical variance estimators in such cases, we use bootstrapped standard errors as an approximate baseline estimate, as suggested in the literature (e.g., Xu, 2006). The first exercise involves the use of the NHIS 2004.
Repetitive values of the ranking indicators of economic welfare are often introduced due to incidental ties or censoring in the welfare variable, or the categorical nature of welfare variables used in numerous national surveys. In calculating concentration index (CI), assigning different fractional ranks to observations that have same values of the welfare measure leads to unstable and inconsistent CI estimates when the welfare variable is categorical or censored. In this paper, we establish an interval within which the CI estimates lie, and propose a solution, which is an extension of (Kakwani, N.C., Wagstaff, A., van Doorslaer, E., 1997. Socioeconomic inequalities in health: measurement, computation, and statistical inference. Journal of Econometrics 77, 87–103), for consistent and replicable estimates of CI when there are a substantial number of ties of the welfare indicator.
The SEEDs of two gastrointestinal diseases: Socioeconomic, environmental, and demographic factors related to cryptosporidiosis and giardiasis in Massachusetts
2008, Environmental Research
We assessed associations between community-level socioeconomic, demographic, and environmental characteristics, and the presence of two potentially waterborne infectious diseases, cryptosporidiosis and giardiasis, as reported to the Massachusetts Department of Public Health.
We created a series of maps showing the spatial distribution of cryptosporidiosis and giardiasis in Massachusetts (1993–2002) overall and by age, using logistic regression to analyze associations between community-level characteristics and the presence of at least one reported case of each disease. This analysis was repeated for communities with predominantly private water supplies.
After adjusting for population size, higher population density and larger than average household sizes were associated with increased odds of reported cases of cryptosporidiosis. Giardiasis was also associated with high population density, but was not associated with household size. In the elderly, income was positively associated with the presence of giardiasis.
These findings suggest that greater population density and larger household sizes may increase the likelihood of protozoan gastrointestinal infection. The results emphasize the necessity to account for distal factors, such as demographic characteristics, that may ultimately play a role in the transmission or reporting of disease.

View all citing articles on Scopus

View full text

State-level variations in income-related inequality in health and health achievement in the US

Abstract

Section snippets

Background

Measures of income-related inequality in health and health achievement

Results

Discussion

Annals of Epidemiology

Social Science and Medicine

Journal of Health Economics

Social Science and Medicine

Journal of Econometrics

Social Science and Medicine

Social Science and Medicine

Journal of Clinical Epidemiology

Social Science and Medicine

Journal of Health Economics

Journal of Health Economics

Social Science and Medicine

Journal of Health Economics

Regression analysis by example

Distress and perceived health: Mechanisms of health decline

Journal of Health and Social Behavior

Health trajectories: Long-term dynamics among black and white adults

Journal of Health and Social Behavior