It is now widely recognized that health outcomes are deeply influenced by a variety of social factors outside of health care. The dramatic differences in morbidity, mortality, and risk factors that researchers have documented within and between countries are patterned after classic social determinants of health, such as education and income (Link and Phelan, 1995; CSDH, 2008), as well as placed-based characteristics of the physical and social environment in which people live-and the macrostructural policies that shape them. A 2013 report from the National Research Council and the Institute of Medicine cited these socioecological factors, along with unhealthy behaviors and deficiencies in the health care system, as leading explanations for the "health disadvantage" of the United States. In a comparison of 17 high-income countries, age-adjusted all-cause mortality rates for 2008 ranged from 378.0 per 100,000 in Australia to 504.9 in the United States. The report found a pervasive pattern of health disadvantages across diverse categories of illness and injury that existed across age groups, sexes, racial and ethnic groups, and social classes (NRC and IOM, 2013). Recent attention has focused on the substantial health disparities that exist within the United States, where life expectancy varies at the state level by 7.0 years for males and 6.7 years for females (NRC and IOM, 2013) but mortality and life expectancy vary even more substantially across smaller geographic areas such as counties (University of Wisconsin Population Health Institute, 2013; Kulkarni et al., 2011) and census tracts. In many U.S. cities, life expectancy can vary by as much as 25 years across neighborhoods (Evans et al., 2012). The same dramatic geographic disparities can be seen for other outcomes, such as infant mortality, obesity, and the prevalence of diabetes and other chronic diseases. Of the various social determinants of health that explain health disparities by geography or demographic characteristics (e.g., age, gender, race-ethnicity), the literature has always pointed prominently to education. Research based on decades of experience in the developing world has identified educational status (especially of the mother) as a major predictor of health outcomes, and economic trends in the industrialized world have intensified the relationship between education and health. In the United States, the gradient in health outcomes by educational attainment has steepened over the last four decades (Goldman and Smith, 2011; Olshansky et al., 2012) in all regions of the United States (Montez and Berkman, 2014), producing a larger gap in health status between Americans with high and low education. Among white Americans without a high school diploma, especially women, life expectancy has decreased since the 1990s, 1 The authors are participants in the activities of the IOM Roundtable on Population Health Improvement.
