Health Disadvantage in US Adults Aged 50 to 74 Years: A Comparison of the Health of Rich and Poor Americans With That of Europeans (2024)

Objectives. We compared the health of older US, English, and other European adults, stratified by wealth.

Methods. Representative samples of adults aged 50 to 74 years were interviewed in 2004 in 10 European countries (n = 17 481), England (n = 6527), and the United States (n = 9940). We calculated prevalence rates of 6 chronic diseases and functional limitations.

Results. American adults reported worse health than did English or European adults. Eighteen percent of Americans reported heart disease, compared with 12% of English and 11% of Europeans. At all wealth levels, Americans were less healthy than were Europeans, but differences were more marked among the poor. Health disparities by wealth were significantly smaller in Europe than in the United States and England. Odds ratios of heart disease in a comparison of the top and bottom wealth tertiles were 1.94 (95% confidence interval [CI] = 1.69, 2.24) in the United States, 2.13 (95% CI = 1.73, 2.62) in England, and 1.38 (95% CI = 1.23, 1.56) in Europe. Smoking, obesity, physical activity levels, and alcohol consumption explained a fraction of health variations.

Conclusions. American adults are less healthy than Europeans at all wealth levels. The poorest Americans experience the greatest disadvantage relative to Europeans.

Wealth, Income, and Education Measurements

We operationalized wealth as household total net worth: the sum of all financial (net stock value, mutual funds, bonds, and savings) and housing wealth (value of primary residence net of mortgage, other real estate value, own business share, and owned cars) minus liabilities. Missing items were imputed with hot-decking procedures, whereby a missing value is replaced by a value randomly selected from an appropriately formed donor pool.^18–20 To account for differences in the number of household members, we divided wealth by the square root of household size.^21,22 We adjusted wealth for purchasing power parity and converted US dollars to EU euros (€ 1.00 = US $1.24) in all countries. For some analyses, we collapsed wealth into country-specific tertiles.

We classified education into 3 categories: in Europe, these corresponded approximately to the International Standard Classification of Education levels 0 to 2 (lower secondary school or lower), 3 (upper secondary school), and 4 to 6 (postsecondary). In the United States, education categories were high school or less (0 to 12 years of schooling), more than high school but not a college graduate (13 to 15 years), and college graduate or more (≥ 16 years). We chose these classifications because they yielded similar educational distributions in Europe and the United States. However, applying International Standard Classification of Education classifications both in Europe and the United States did not alter our results.

Outcome Measurements

We examined health outcomes or conditions assessed with comparable questionnaire items across surveys: chronic diseases, measured by self-reported doctor's diagnosis of heart disease, stroke, hypertension, diabetes or high blood sugar, cancer (excluding skin cancer), and lung disease and disability, measured in 2 ways—limitations on 1 or more of 6 instrumental activities of daily living and 1 or more limitations on 10 mobility and fine-motor control items (mobility).²³

Risk Factors

We categorized smoking as current, former, or never. We categorized alcohol-drinking frequency as drinking any alcoholic beverage daily or almost daily, 5 to 6 days per week, 3 to 4 days per week, 1 to 2 days per week, or less than 1 or 2 days per month. We dichotomized alcohol-drinking intensity based on drinking more than 2 drinks 5 to 6 days per week in HRS and SHARE. In ELSA, questions referred to the past week; therefore, we dichotomized alcohol-drinking intensity based on drinking more than 2 drinks per day combined with drinking 5 to 7 days in the past week. We dichotomized physical activity based on participation in vigorous activities (e.g., sports, heavy housework) more than 1 day per week. We defined body mass index as weight in kilograms divided by height in meters squared. Weight and height were self-reported in HRS and SHARE but measured in ELSA.

Data Analysis

We initially analyzed men and women separately but pooled results because estimates did not differ by gender. We analyzed data in 5 steps. First, we used logistic regression to model the prevalence of conditions in the United States, England, and Europe, and we adjusted for age and gender, taking the US distributions as standard. Second, we fit smooth nonparametric LOESS function curves to examine health across the distribution of wealth in each region.²⁴ LOESS is a locally weighted regression smoother that fits multiple lines to small parts of the wealth axis and combines their central parts. For clarity, in the figures, we used wealth from –€ 100 000 to € 1 000 000, which comprised 95% of each region's population. Third, to quantify this relation, we fit splines with 3 knots that separated linear segments along which the association between wealth and health was approximately linear. The spline model provides odds ratios (ORs) of each health outcome per € 100 000 increase in wealth separately for 4 wealth segments: less than € 0; € 0 to less than € 199 999; € 200 000 to € 399 999; and € 400 000 or more. Because few individuals reported negative wealth, we only tabulated results for the 3 positive wealth segments. In these models, we considered absolute wealth.

Fourth, we used logistic regression to model prevalence of health outcomes by wealth tertiles to assess wealth relative to countrymen, and tested for differences among England, Europe, and the United States. These models were adjusted for age, gender, and country (in Europe) or region (in the United States: Northeast, Midwest, South, West, or unknown). Coefficients were allowed to differ among countries through interaction terms. Finally, to assess the role of risk factors in explaining health disparities, we modeled the prevalence of health outcomes by wealth tertile, and adjusted for demographics, educational level, and risk factors.

Analyses were performed with appropriate sampling weights in SAS version 8.2 (SAS Institute Inc, Cary, NC) or S-plus version 6.0 (Insightful Corp, Seattle, WA).

Health Disparities by Absolute Wealth

Figure 2 shows predicted health by wealth with adjusted LOESS regression curves. In most cases, the US line never crosses either the English or European lines, indicating that US adults reported worse health than English or other Europeans at every wealth level. Some Americans achieved the same level of health as Europeans but only at very high wealth levels. For example, the risk of reporting heart disease for a European with € 3000 was 11%, which was equivalent to the risk for an American with € 300 000.

Open in a separate window

FIGURE 2

LOESS function of chronic disease and disability by wealth among men and women aged 50 to 74 years in the United States, England, and Europe, for (a) heart disease, (b) stroke, (c) hypertension, (d) diabetes, (e) cancer, (f) lung disease, (g) 1 or more IADL limitation, and (h) 1 or more mobility limitation: HRS, United States, 2004; ELSA, England, 2004; and SHARE, Europe, 2004.

Notes. HRS = US Health and Retirement Survey; ELSA = English Longitudinal Study of Ageing; SHARE = Survey of Health, Ageing and Retirement in Europe; IADL = instrumental activities of daily living. Models adjusted for age, gender, and educational level. Wealth values are in euros (€ 1.00 = US $1.24) and were adjusted by purchasing power parity.

Except for cancer, higher wealth predicted better health in all 3 populations. Curves were much steeper at lower wealth levels, at which point they were steeper in the United States than in Europe. The health disadvantage of Americans was greatest at the bottom of the wealth distribution and diminished with increasing wealth. In supplementary analyses, we investigated LOESS models by country-specific percentile of wealth, rather than absolute wealth in euros, to ameliorate bias from imperfect cross-national wealth purchasing parity adjustments. Key findings were nearly identical (see Supplementary Figure 1, available as a supplement to the online version of this article at http://www.ajph.org).

Table 1 presents ORs from segmented spline models, showing that gradients are steepest at low levels of wealth. For instance, among Americans with wealth € 0 to € 199 999, each € 100 000 wealth increment was associated with a 24% reduction in odds of heart disease (OR = 0.76; 95% confidence interval [CI] = 0.69, 0.84). At wealth between € 200 000 and € 399 999, wealth increments of € 100 000 were associated with reductions in the odds of heart disease of about 12%. At wealth € 400 000 or higher, increasing wealth did not predict heart disease. Patterns were similar for every condition, except cancer. Although CIs usually overlapped, the ORs in the lowest wealth group (€ 0 to € 199 999) were more extreme in the United States than in England or Europe.

Health Disparities Across Wealth Tertiles

Americans reported worse health than English and European adults in all 3 wealth tertiles, but the US disadvantage was most substantial in the lowest tertile. For instance, after we adjusted for demographics, 22% of US adults in the bottom wealth tertile reported heart disease, compared with 17% in the bottom tertile England and 13% in the bottom tertile in Europe (see Supplementary Figure 2, available as a supplement to the online version of this article at http://www.ajph.org).

Table 2 (left panel) shows that differences in nearly every outcome between top and bottom wealth tertiles were similar in England and the United States, but significantly smaller in the rest of Europe. Except for cancer, the prevalence of chronic diseases and functional limitations increased with decreasing wealth tertile in all countries. Within Europe, health differences by wealth tertile tended to be smaller in the southern or Mediterranean countries, Austria, and Switzerland, and relatively large in the Scandinavian countries, Germany, the Netherlands, and France. However, even in the latter populations, health disparities by wealth were generally smaller than in the United States and England (supplemental figure available at http://www.ajph.org).

Incorporating Risk Factors

In each population, lower wealth was associated with higher smoking prevalence, higher body mass index, less physical activity, and lower alcohol consumption (results not shown). Within each region, adjusting for these risk factors attenuated ORs for comparisons of low to high wealth tertiles by about a third (Table 2, left panel). For instance, when we controlled for risk factors, Americans in the bottom wealth tertile had higher odds of reporting heart disease compared with Americans in the top tertile (OR = 1.68; 95% CI = 1.45, 1.94); corresponding ORs were 1.83 (95% CI = 1.48, 2.27) for England and 1.25 (95% CI = 1.11, 1.42) for Europe.

Adults in the United States and England had higher prevalence of obesity and physical inactivity than Europeans, but the latter had higher current smoking prevalence (see table available as a supplement to the online version of this article at http://www.ajph.org). Prevalence of ever smoking was lowest in Europe, but this was largely explained by low ever-smoking rates in women (results not shown). Excessive alcohol use was more prevalent in Europe than in the United States. Adjusting for risk factors only modestly attenuated the American health disadvantage for any health outcome (Table 2, right panel). For example, Europeans had lower odds of reporting heart disease than Americans in both unadjusted (OR = 0.52; 95% CI = 0.48, 0.56) and adjusted models (OR = 0.63; 95% CI = 0.58, 0.68).

Explaining Health Disparities

Our results show that US adults of all socioeconomic levels report worse health than do English or other European adults.^3,4 Notwithstanding possible limitations, several substantial explanations should be considered: survival advantages among US adults with chronic illness, behavioral differences, differences in the health care system, or social policy contexts other than medical care that indirectly impair American health. Our results shed light on the plausibility of several of these.

Higher disease prevalence may stem from higher survival rates in the United States than in Europe.^5–7 Compared with the European approach, the American medical system might be more focused on ameliorating the consequences of disease, with relatively less attention given to prevention.² A combination of lower disease incidence in Europe but higher survival rates in the United States might partly explain our results, but is unlikely to explain discrepancies this large, especially in those younger than 60 years.

Risk factors played a modest role in explaining cross-national health variations. On average, current smoking is higher in Europeans than in Americans, but prevalence of ever smoking is higher among Americans. Obesity prevalence in the United States was higher than in most European countries, yet these factors did not explain cross-national health variations. In part, this may arise from our use of only current measures of obesity instead of life-course measures of accumulated exposure from early life to adulthood.^25–28 Furthermore, other risk factors that we did not examine, such as dietary intake,²⁹ blood lipid levels,³ and hormonal replacement therapy,³⁰ may also have contributed to health disparities across countries.

The policies in Europe, which are more generous and egalitarian when compared to those in the United States, may improve health outcomes and reduce health disparities.^31,32 Compared with the United States, many European countries have a higher provision of social transfers (e.g., social retirement benefits, unemployment compensation, sick pay) and higher levels of public health expenditures.^1,33 In addition, the social policy contexts in the United States and Europe differ substantially in other areas such as working conditions, transportation options, and housing policies. Although there is limited evidence as yet on the causal effect of these contextual factors,^34–36 the pervasive health disadvantage of Americans even among the very wealthy suggests that some of these factors may indeed detrimentally influence health.

Features of the US health care system may contribute to the worse health of Americans compared with Europeans.³⁷ In particular, most European countries have a stronger primary care orientation than does the United States.^12–14 Previous evidence suggests that a strong primary care system is associated with better health outcomes,^14,37,38 partly because it entails a stronger focus on primary prevention,³⁹ a more equitable distribution of resources,¹² and a higher efficacy of the health system.³⁸ Thus, part of the US health disadvantage may be explained by a weak focus on primary care compared with most European health systems.^12,14,37,40 However, the fact that even wealthy Americans had worse health than their European counterparts suggests that cross-national disparities may not be fully attributable to health care.

Wealth may influence health through different mechanisms in the United States and Europe. Lower wealth limits access to care, particularly among uninsured or partially insured Americans,¹¹ whereas in Europe medical care is accessible to all. Furthermore, a stronger primary care orientation in Europe may lead to smaller health disparities than are found in the United States.^10,12,37,41 However, the fact that health disparities were relatively large in England as well suggests that mechanisms outside the health care system may also be involved. Wealth enhances access to material resources such as housing, and is a source of immediate consumption in periods of economic strain.⁴² Wealth may also increase sense of control over life and other psychosocial resources that can enhance health.^43,44 Examining whether these pathways operate differently in the United States and Europe is an important subject for further research.

Differences between the United States and Europe in how health influences wealth might also have contributed to steeper wealth gradients in health in the former. Major health events lead to wealth depletion among Americans,⁴⁵ partly because the uninsured and underinsured use up wealth to cover medical expenses. By contrast, universal healthcare access in Europe may substantially ameliorate this problem.⁴⁶ The onset of chronic diseases also reduces work and income,^4,45–47 but further research is needed to examine whether these effects differ between Europe and the United States.

Study Limitations

An important limitation of our study was the reliance on self-reports. Adults in the United States may be diagnosed more often with certain diseases than are Europeans. Thus, we cannot discard the possibility that higher disease prevalence may stem from higher doctor consultation or screening rates in the United States, e.g., for cancer or hypertension.^48,49 However, several lines of evidence indicate that differential reporting may not fully explain the cross-country differences. First, previous studies have indicated that a health disadvantage between England and the United States was also observed for objectively measured biological markers such as plasma fibrinogen, glycosylated hemoglobin A_1c, C-reactive protein, and high-density lipoprotein cholesterol.³ Similarly, when reporting disability, Americans appear to have higher response thresholds than do Europeans.⁵⁰ Furthermore, poor Americans are unlikely to receive closer medical surveillance than are European adults, who have access to universal health care.^10,12

As a secondary validation approach, we compared World Health Organization data on life expectancy and healthy life expectancy between the United States and Europe.⁵¹ Except for women in Denmark, both life expectancy and the number of years a person can expect to live in good health were substantially lower in the United States than in the European countries we studied.⁵¹ Cross-nationally, comparable data on disease prevalence is limited and further confirmatory research is warranted. Yet, these results generally support our findings of a health gap between the United States and many European countries.

We did not explicitly examine the role of race, because data on race were not available for Europe. Because we excluded US minorities, the worse health of Americans compared with Europeans cannot be attributed to racial disparities within the United States. In supplementary analysis, we found that, consistent with a large body of research, US Blacks had generally worse health than did US Whites. For instance, 4% of Whites reported a stroke as opposed to 9% of Blacks, and 14% of Whites reported diabetes as opposed to 26% of Blacks. ORs for a comparison of health across wealth tertiles remained almost unchanged after we incorporated US minorities in the analysis. For instance, ORs for a comparison of the high and low tertiles of wealth were 2.42 (95% CI = 2.12, 2.77) for diabetes and 2.36 (95% CI = 1.88, 2.95) for stroke.

A limitation of our study is that some European countries had relatively low response rates. The lower prevalence of health problems in some countries such as Switzerland may indeed be partly because of selective response. Despite this, a health disadvantage in the United States was also observed compared with countries with relatively high response rates such as France. Furthermore, the fact that our findings are consistent with data on mortality and healthy life expectancy indirectly suggests that response rate differences do not fully account for our main results.

Our purchasing power parity adjustment may not fully capture price differences across countries. Furthermore, we omitted pension wealth because of the considerable international variation in pension systems. Overall, public pensions are bigger and distributed more equally in Europe than they are in the United States and England.³³ Thus, European households with a certain level of nonpension wealth may actually be wealthier than a household with the same level of nonpension wealth in the United States or England. However, the finding that LOESS curves that used wealth percentile were similar to those that used absolute wealth is reassuring (see Supplementary Figure 1, available at http://www.ajph.org). This suggests that, assuming the ranking of total wealth is preserved, our 2 main findings are robust to price differences and the omission of pension wealth.

Our study focused on middle-aged adults only. Because cross-country health variations might be larger at young ages,^13,37 we may have underestimated the total magnitude of health differences between Europe and the United States.

Conclusions

Americans face a health disadvantage such that no matter what their wealth, their health lags behind that achieved by comparable Europeans. The disadvantage is remarkably pervasive and affects even the wealthy, but is largest for the poor. Tackling risk factors such as obesity and improving the financing and delivery of health services particularly among the poor might partly contribute to eliminating health disparities. Yet, changes in broader social policies and determinants may also be required. Identifying the conditions that result in worse health even among wealthy Americans is an important topic for future research.

M. Avendano was supported by the Netherlands Organization for Scientific Research (grant 451-07-001) and a fellowship from Erasmus University. M. M. Glymour is a Robert Wood Johnson Foundation Health and Society Scholar at Columbia University. Funding for the Health and Retirement Study was provided by the US National Institute on Aging (NIA; grant U01AG009740). The English Longitudinal Study on Ageing has been supported by the NIA (grants 2RO1AG7644-01A1 and 2RO1AG017644) and several British government departments—the Department for Education and Skills; Department for Environment, Food, and Rural Affairs; Department of Health; Department of Trade and Industry; Department for Work and Pensions; Her Majesty's Treasury Inland Revenue; the Office of the Deputy Prime Minister; and the Office for National Statistics. We used data from the early release 1 of SHARE 2004. The SHARE data collection was primarily funded by the European Commission through the 5th framework program (project QLK6-CT-2001-00360 in the thematic program Quality of Life). Additional funding came from the NIA (grants U01 AG09740-13S2, P01 AG005842, P01 AG08291, P30 AG12815, Y1-AG-4553-01, and OGHA 04-064). Data collection in Austria (through the Austrian Science Fund) and Switzerland (through the Swiss Federal Office for Education and Science [BBW/OFES/UFES]) was nationally funded. For methodological details see Börsch-Supan et al.¹⁸

We acknowledge the support of Caspar Looman in the analysis of the data.

1. Health Data 2006. Paris, France: Organisation for Economic Co-operation and Development; 2006 [Google Scholar]

2. Anderson G, Hussey PS. Comparing health system performance in OECD countries. Organisation for Economic Co-operation and Development. Health Aff (Millwood) 2001;20:219–232 [PubMed] [Google Scholar]

3. Banks J, Marmot M, Oldfield Z, Smith JP. Disease and disadvantage in the United States and in England. JAMA 2006;295:2037–2045 [PubMed] [Google Scholar]

4. Sacker A, Wiggins RD, Bartley M, McDonough P. Self-rated health trajectories in the United States and the United Kingdom: a comparative study. Am J Public Health 2007;97:812–818 [PMC free article] [PubMed] [Google Scholar]

5. Hanson GS. Comparison of health status between the United States and England. JAMA 2006;296:2312; author reply p2313 [PubMed] [Google Scholar]

6. Thorn CF. Comparison of health status between the United States and England. JAMA 2006;296:2312; author reply p2313 [PubMed] [Google Scholar]

7. Whitney E. Comparison of health status between the United States and England. JAMA 2006;296:2312–2313; author reply p2313 [PubMed] [Google Scholar]

8. Thorpe KE, Howard DH, Galactionova K. Differences in disease prevalence as a source of the US-European health care spending gap. Health Aff (Millwood) 2007;26:w678–w686 [PubMed] [Google Scholar]

9. Silventoinen K, Sans S, Tolonen H, et al. Trends in obesity and energy supply in the WHO MONICA Project. Int J Obes Relat Metab Disord 2004;28:710–718 [PubMed] [Google Scholar]

10. Dahl E, Fritzell J, Lahelma E, Martikainen P, Kunst A, Mackenbach JP. Welfare state regimes and health inequalities. In: Siegrist J, Marmot M, eds. Social Inequalities in Health: New Evidence and Policy Implications London, England: Oxford University Press; 2006 [Google Scholar]

11. Adams PF, Dey AN, Vickerie JL. Summary health statistics for the US population: National Health Interview Survey, 2005. Vital Health Stat 10 2007;(233):1–104 [PubMed] [Google Scholar]

12. Starfield B, Shi L. Policy relevant determinants of health: an international perspective. Health Policy 2002;60:201–218 [PubMed] [Google Scholar]

13. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q 2005;83:457–502 [PMC free article] [PubMed] [Google Scholar]

14. Starfield B. Is primary care essential? Lancet 1994;344:1129–1133 [PubMed] [Google Scholar]

15. Heeringa S, Connor J. Technical Description of the Health and Retirement Study Sample Design. Ann Arbor, MI: Survey Research Center, University of Michigan; 1995 [Google Scholar]

16. Juster F, Suzman R. An overview of the health and retirement study. J Hum Resour 1995;30(suppl):S7–S56 [Google Scholar]

17. Ofstedal MB, Fisher GG, Herzog AR. Documentation of Cognitive Functioning Measures in the Health and Retirement Study. Ann Arbor, Michigan: Survey Research Center, University of Michigan; 2002 [Google Scholar]

18. Börsch-Supan A, Brugiavini A, Jürges H, Mackenbach J, Siegrist J, Weber G. Health, Ageing and Retirement in Europe. Morlenbach, Germany: Strauss GmbH; 2005 [Google Scholar]

19. Marmot M, Banks J, Blundell R, Lessof C, Nazroo J. Health, Wealth and Lifestyles of the Older Population in England: The 2002 English Longitudinal Study of Ageing London, England: University College London; 2002 [Google Scholar]

20. St Clair P, Bugliari D, Chien S, et al. RAND HRS Data Documentation Version E Santa Monica, CA: Labor & Population Program; 2005 [Google Scholar]

21. Huisman M, Kunst AE, Mackenbach JP. Socioeconomic inequalities in morbidity among the elderly; a European overview. Soc Sci Med 2003;57:861–873 [PubMed] [Google Scholar]

22. Buhmann B, Rainwater L, Schmaus G, Smeeding TJ. Equivalence scales, well-being, inequality and poverty: sensitivity estimates across ten countries using the Luxembourg income study (LIS) data base. Rev Income Wealth 1998;34:115–142 [Google Scholar]

23. Nicholas S, Huppert F, McWilliams B, Melzer D. Physical and cognitive function. In: Marmot M, James B, Blundell R, Lessof C, Nazroo J, eds. Health, Wealth and Lifestyles of the Older Population in England: The 2002 English Longitudinal Study of Ageing London, England: Institute for Fiscal Studies; 2003 [Google Scholar]

24. Cleveland W. Robust locally weighted regression and smoothing scatterplots. J Am Stat Assoc 1979;74:829–836 [Google Scholar]

25. Galobardes B, Smith GD, Lynch JW. Systematic review of the influence of childhood socioeconomic circ*mstances on risk for cardiovascular disease in adulthood. Ann Epidemiol 2006;16:91–104 [PubMed] [Google Scholar]

26. Singh-Manoux A, Ferrie JE, Chandola T, Marmot M. Socioeconomic trajectories across the life course and health outcomes in midlife: evidence for the accumulation hypothesis? Int J Epidemiol 2004;33:1072–1079 [PubMed] [Google Scholar]

27. Power C, Matthews S, Manor O. Inequalities in self-rated health: explanations from different stages of life. Lancet 1998;351:1009–1014 [PubMed] [Google Scholar]

28. Smith GD, Hart C, Blane D, Gillis C, Hawthorne V. Lifetime socioeconomic position and mortality: prospective observational study. BMJ 1997;314:547–552 [PMC free article] [PubMed] [Google Scholar]

29. Girois SB, Kumanyika SK, Morabia A, Mauger E. A comparison of knowledge and attitudes about diet and health among 35- to 75-year-old adults in the United States and Geneva, Switzerland. Am J Public Health 2001;91:418–424 [PMC free article] [PubMed] [Google Scholar]

30. Lundberg V, Tolonen H, Stegmayr B, Kuulasmaa K, Asplund K. Use of oral contraceptives and hormone replacement therapy in the WHO MONICA project. Maturitas 2004;48:39–49 [PubMed] [Google Scholar]

31. Dahl E, Fritzell J, Lahelma E, Martikainen P, Kunst A, Mackenbach J. Welfare state regimes and health inequalities. In: Siegrist J, Marmot M, eds. Social Inequalities in Health: New Evidence and Policy Implications Oxford, England: Oxford University Press; 2006 [Google Scholar]

32. Navarro V, Muntaner C, Borrell C, et al. Politics and health outcomes. Lancet 2006;368:1033–1037 [PubMed] [Google Scholar]

33. Disney RJ, Johnson P, ed. Pension Systems and Retirement Incomes Across OECD Countries Cheltenham, England: Edward Elgar; 2001 [Google Scholar]

34. Kennelly B, O'Shea E, Garvey E. Social capital, life expectancy and mortality: a cross-national examination. Soc Sci Med 2003;56:2367–2377 [PubMed] [Google Scholar]

36. Islam MK, Merlo J, Kawachi I, Lindström M, Gerdtham UG. Social capital and health: does egalitarianism matter? A literature review. Int J Equity Health 2006;5:3. [PMC free article] [PubMed] [Google Scholar]

37. Macinko J, Starfield B, Shi L. The contribution of primary care systems to health outcomes within Organization for Economic Cooperation and Development (OECD) countries, 1970-1998. Health Serv Res 2003;38:831–865 [PMC free article] [PubMed] [Google Scholar]

38. Macinko J, Starfield B, Shi L. Quantifying the health benefits of primary care physician supply in the United States. Int J Health Serv 2007;37:111–126 [PubMed] [Google Scholar]

39. Shi L, Starfield B. Primary care, income inequality, and self-rated health in the United States: a mixed-level analysis. Int J Health Serv 2000;30:541–555 [PubMed] [Google Scholar]

40. Starfield B. Insurance and the US health care system. N Engl J Med 2005;353:418–419 [PubMed] [Google Scholar]

41. Shi L, Starfield B, Politzer R, Regan J. Primary care, self-rated health, and reductions in social disparities in health. Health Serv Res 2002;37:529–550 [PMC free article] [PubMed] [Google Scholar]

42. Wolf E. International comparisons of wealth inequalities. Rev Income Wealth 1996;42:433–451 [Google Scholar]

43. Steptoe A, Marmot M. The role of psychobiological pathways in socio-economic inequalities in cardiovascular disease risk. Eur Heart J 2002;23:13–25 [PubMed] [Google Scholar]

44. Marmot MG, Bosma H, Hemingway H, Brunner E, Stansfeld S. Contribution of job control and other risk factors to social variations in coronary heart disease incidence. Lancet 1997;350:235–239 [PubMed] [Google Scholar]

45. Lee J, Kim H. An examination of the impact of health on wealth depletion in elderly individuals. J Gerontol B Psychol Sci Soc Sci 2003;58:S120–S126 [PubMed] [Google Scholar]

46. Hurd M. Health, wealth, and the role of institutions. J Hum Resour 2003;38:386–415 [Google Scholar]

47. Smith J. Unraveling the SES-health connection. In: Waite L, ed. Aging, Health, and Public Policy: Demographic and Economic Perspectives New York, NY: Population Council; 2005:108–132 [Google Scholar]

48. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74–108 [PubMed] [Google Scholar]

49. Wolf-Maier K, Cooper RS, Banegas JR, et al. Hypertension prevalence and blood pressure levels in 6 European countries, Canada, and the United States. JAMA 2003;289:2363–2369 [PubMed] [Google Scholar]

50. Kapteyn A, Smith JP, van Soest A. Vignettes and self-reports of work disability in the United States and the Netherlands. Am Econ Rev 2007;97:461–473 [Google Scholar]

51. WHO Statistical Information System Geneva, Switzerland: World Health Organization; 2007 [Google Scholar]