Low Income, Not Race or Lifestyle, Is the Greatest Threat to Health
David R. Williams, Ph.D., M.P.H., professor of sociology at the University of Michigan Institute for Social Research, Ann Arbor, Mich., studied the relationship between socioeconomic status (SES) and health.
Using the Panel Study of Income Dynamics (PSID), an ongoing longitudinal survey of U.S. households begun in 1968 with mortality follow-up through 1992, Williams and his colleagues examined how aspects of SES such as income, occupation, gender, and race affect health status and predict mortality.
His research team also analyzed data from the Detroit Area Study (DAS) conducted in 1995 and the Americans' Changing Lives (ACL) longitudinal survey begun in 1986 and repeated in 1989 and 1994.
The research resulted in the publication of a series of articles in which Williams observes:
- Racial differences in health are markedly, but not wholly, explained by income.
- Perceived discrimination and race-related stress also play a role.
- Persistent low income can be an especially good predictor of mortality regardless of race.
- Contrary to popular opinion, when income is taken into account, behavioral risk factors such as smoking, alcohol drinking, sedentary lifestyle, and obesity accounted for only a small proportion of differences in mortality across sex, race, and age groups.
In an article in the Journal of Health Psychology, Williams and colleagues state that:
- Education and especially income matter most to health status, but discrimination and stress play an incremental role.
In an article in the American Journal of Public Health, Williams and colleagues write that:
- While income loss had the greatest impact on mortality among middle-income individuals, persistent poverty was the strongest predictor among low-income persons.
In a study published in the Journal of the American Medical Association (JAMA), the researchers found that:
- Persons with incomes of less than $10,000 were 3.22 times more likely to die of any cause than were those with incomes over $30,000.
- Because income is so important a determinant of health disparities among people of different races, Williams concludes that policies to reduce income differentials, especially to raise incomes in the lowest tier, would provide the most help in dealing with these differences.
RWJF supported the research from January 1995 through December 1996 with a $242,334 grant.
This grant from RWJF supported studies conducted by David R. Williams, Ph.D., M.P.H., professor of sociology at the University of Michigan, Ann Arbor, Mich., on the relationship between socioeconomic status (SES) and health.
Using the Panel Study of Income Dynamics (PSID), an ongoing longitudinal survey of a nationally representative sample of US households begun in 1968 with mortality follow-up through 1992, Williams examined how aspects of SES such as income, occupation, race, and gender affect health status and predict mortality.
Specifically, Williams aimed to:
- Analyze the extent to which SES differentials in mortality have been widening in recent decades.
- Understand the relative contribution of multiple indicators within SES to mortality.
- Identify the health consequences of stability versus change in the history and patterning of SES over time.
- Examine how the association between SES and mortality varies by race and gender.
RESULTS AND FINDINGS
In a series of published articles, Williams and colleagues presented the following findings and conclusions.
- Education and especially income matter most in health status, but discrimination and stress play an incremental role. In a 1997 article in the Journal of Health Psychology, Williams and colleagues presented an analysis of data from a personal-interview survey of adults in the Detroit area conducted in 1995: the Detroit Area Study, or DAS. They examined the extent to which differences in socioeconomic status, perceived discrimination, and stress explained racial disparities in self-reported physical and mental health. African-Americans reported higher levels of ill health than whites. The authors found that "the race effect is dramatically reduced when economic status is considered, but it remains significant." The impact of everyday discrimination or chronic stress was more important than major events. For stress among the general population, however, life events were more important than chronic stress.
- While income loss had the greatest impact on mortality among middle-income individuals, persistent poverty was the strongest predictor among low-income persons. In an article in the September 1997 issue of the American Journal of Public Health, Williams and colleagues used PSID data to examine the effects of income level and change on mortality. Among adults aged 45 through 64, those with five-year average incomes under $15,000 were 3.89 times more likely to die of any cause during the subsequent five-year period than were those with incomes over $70,000. Those whose incomes were $20,000 to $70,000 had little to no increased risk of dying in the next five years unless they had experienced one or more income losses of at least 50 percent, in which case the odds were 3.22 times as great that they were likely to die. These differentials did not appear to be related to sex or race. However, the authors caution that "African-Americans may be more likely to have unstable incomes of a nature not reflected by the measures used in this study."
- Contrary to popular opinion, behavioral risk accounts for only a small proportion of differences in mortality across age, sex, and race groups when taking income into account. Controlling for other demographic factors, persons with incomes of less than $10,000 were 3.22 times more likely to die of any cause than those with incomes over $30,000. When factors such as smoking, drinking, obesity, or physical inactivity were considered, the risk of dying of any cause for the lowest income group was still 2.77 times as great. These findings were presented in an article in JAMA (June 3, 1998 issue). Williams and his colleagues based the article on an examination of data on personal characteristics and behavioral risks from the Americans' Changing Lives (ACL) longitudinal survey conducted by the University of Michigan Survey Research Center in 1986, 1989, and 1994.
- The researchers concluded that although reducing the prevalence of health-risk behaviors in low-income populations is an important public health goal, socioeconomic differences in mortality are due to a wider array of factors and therefore would persist even with improved health behaviors among the disadvantaged.
Williams drew from his study findings to make recommendations or observations regarding several policy issues.
- Because income is so important a determinant of health disparities among people of different races, policies to reduce income differentials especially to raise incomes in the lowest tier would provide the most help to deal with these differences. Williams and colleagues found repeatedly that "it's deadly to be poor." Given the results of this and similar studies, they concluded that policies to help equalize income offer the greatest promise to improve health and longevity and should be considered in future research and policy formulation.
- Remedies for race-related discrimination and stress should also be sought through change in society's economic, legal, political, and social structures.
- Researchers and policy analysts need to expand knowledge about what works to increase income among the poor including granting tax credits, legislating increases in the minimum wage, and offering direct income supplements. Williams acknowledged waning public support for and commitment to improving the social and economic conditions of vulnerable populations. But he and the co-authors of the American Journal of Public Health article pointed to the possibility that Medicare and Social Security may have reduced the risk of mortality for persons older than 65 years despite prior and current income differentials.
- Health care does not play a central role in the disparity drama. Williams and colleagues did not dismiss the importance of access to health care because of its potential for reducing pain and suffering and improving the quality of life. But the researchers were interested primarily in the root causes underlying persistent gaps in health status among low-income individuals and people of color.
- Today's focus on behavioral-risk reduction should be examined to see whether it is prudent policy or victim blaming, especially of racial/ethnic populations and low-income groups. The conventional wisdom surrounding most health promotion and disease prevention efforts places heavy emphasis on personal responsibility. One of the most provocative findings in the JAMA article co-authored by Williams was that high-risk behaviors had only a minor impact on differences in mortality when income is taken into account. The policy implication is that we may be addressing the easy and relatively inexpensive questions and ignoring those that are hardest to answer but matter the most.
- Many unanswered questions remain about the complex reasons for health disparities among people of different racial, ethnic and cultural backgrounds. Williams' research pointed to challenges for future research. For example, in the Journal of Health Psychology study, the mental health of whites appeared to be more severely affected by stress than was the mental health of African-Americans. Because that finding is at odds with findings about physical health, Williams and colleagues hypothesized that stress takes a cumulative physical toll on African-Americans in a different way than it does on whites. In addition, in the American Journal of Public Health article, middle-income persons appeared to be more adversely affected by income loss than those with low incomes. The authors suggested further work to determine whether this could be due to the been-down-so-long, it-looks-like-up-to-me syndrome or the cushioning effect of public aid or community support for low-income persons.
Findings were published in several journals, including the Journal of the American Medical Association, Journal of Health Psychology, and American Journal of Public Health. Williams and his colleagues presented the findings at a number of conferences, including the United Nations University/World Institute for Development Economics Research Conference in Helsinki, Finland, two sections of the American Sociological Association annual meeting in New York, N.Y., and the annual meeting of the American Public Health Association in New York, N.Y. (See the Bibliography for details.)
GRANT DETAILS & CONTACT INFORMATION
Inequality and Health: Patterns and Dynamics
University of Michigan Institute for Social Research (Ann Arbor, MI)
Dates: January 1995 to December 1996
David R. Williams, Ph.D., M.P.H.
(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)
McDonough P, Duncan GJ, Williams DR and House JS. "The Impact of Income Dynamics on Mortality in the United States," in The Mortality Crisis in Transitional Economies. Cornia, G and Paniccia R (eds.) New York: Oxford University Press, pp. 392419, 2000.
House JS and Williams DR. Article in the NHLBI Report of the Conference on Socioeconomic Status and Cardiovascular Health and Disease: Psychosocial Pathway Linking SES and CVD. Bethesda, MD: National Institutes of Health; 124199, 1995.
Krieger N, Williams DR and Moss NE. "Measuring Social Class in US Public Health Research: Concepts, Methodologies, and Guidelines." Annual Review of Public Health, 18: 341378, 1997. Abstract available online.
Lantz PM, House JS, Lepowski JM, Williams DR, Mero RP and Chen J. "Socioeconomic Factors, Health Behaviors, and Mortality." Journal of the American Medical Association, 279(21): 17031708, 1998. Abstract available online.
McDonough P, Duncan GJ, Williams DR and House JS. "Income Dynamics and Adult Mortality in the US, 19721999." American Journal of Public Health, 87(9): 14761483, 1997. Abstract available online.
McDonough P, Williams DR, House JS and Duncan GJ. "Gender and the Socieconomic Gradient in Mortality." Journal of Health and Social Behavior, 40(1): 1731, 1999. Abstract available online.
Williams DR and Collins C. "US Socioeconomic and Racial Differences in Health: Patterns and Explanations." Annual Review of Sociology, 21: 349386, 1995.
Williams DR, Yu Y, Jackson JS and Anderson NB. "Racial Differences in Physical and Mental Health: Socioeconomic Status, Stress and Discrimination." Journal of Health Psychology, 2(3): 335351, 1997.
Presentations and Testimony
Nancy Krieger, David R. Williams and NE Moss. "Measuring Social Class in US Public Health Research: Concepts, Methodologies, and Guidelines," at the Annual Meeting of the American Sociological Association, August 16, 1996, New York.
David R. Williams, Peggy McDonough, James S. House and Greg Duncan. "Incomes Dynamics and Mortality for Black and White Adults," at the Annual Meeting of the American Sociological Association, August 19, 1996, New York.
Peggy McDonough, David R. Williams, James S. House and Greg Duncan. "Gender, Social Class and Mortality," at the Annual Meeting of the American Public Health Association, November 20, 1996, New York.
Peggy McDonough, Greg Duncan, David R. Williams and James S. House, "Income Dynamics in the U.S.: Macroeconomic Trends, Microeconomic Patterns and Individual Health Consequences," at United National University/World Institute for Development, Economics Research Conference on Economic Shocks, Social Stress, and the Demographic Impact on Mortality, April 1997, Helsinki, Finland.
Report prepared by: David Kales
Report prepared by: Janet Heroux
Reviewed by: Molly McKaughan
Reviewed by: Marian Bass
Program Officer: David Colby