The recent reports of deaths from police officer shootings have brought the simple but not-often-discussed statistical concepts of rate and burden to national attention. The rate or relative proportion of police shootings resulting in deaths of unarmed black individuals is approximately 5 times higher than the rate among unarmed white individuals. But the absolute number of such deaths among white vs black individuals is approximately the same because the size of the white population in the United States is much greater.1
These concepts should also inform discussions about health equity. Substantial attention has been focused on the higher rates among black individuals in the United States for almost every adverse health outcome, such as mortality and low birth weight, and health determinants such as obesity, poverty, and inadequate insurance coverage. Maps showing these higher rates in parts of the South and West are all too familiar. But the relative burden on different populations that these rates produce is not always discussed and has implications for the development of health policy.
This focus on rates has led to a prevalent view that health equity refers primarily to racial health equity and often only refers to health equity for black individuals. This view is undoubtedly grounded in both the much higher rates of almost every adverse health outcome and health determinant experienced by black individuals and other people of color. These gaps are unacceptable and require policies that ensure higher levels of public- and private-sector investment in health care and social factors for racial and ethnic groups with such high rates of poor health.
However, poor health is not limited to the black population or other people of color. In fact, the total burden of poor health, even within lower socioeconomic groups, is often greater among white than black individuals simply because there are more white than black individuals of lower socioeconomic status. Higher rates of poor health, as defined by race and as unacceptable as they are, do not produce a greater burden of poorer health in the US population because burden is measured in absolute rather than relative terms.
According to data from the Centers for Disease Control and Prevention (CDC),2,3 the 2013 US mortality rate for infants (aged birth to 1 year) of mothers with less than a high school education is 11.6 per 1000 live births among black infants vs 8.8 per 1000 live births among white infants. Yet the total number of black infant deaths in 2013 was 1013, whereas the total number of white infant deaths was 1337. Even with the lowest infant mortality rate of 5.5 per 1000 live births, there were a total of 1495 infant deaths among US Hispanic residents. Similarly, assuming the data are nationally representative, among families with income of less than $25 000 per year in 2015, 25.0% of black individuals and 26.9% of Hispanic persons, but only 19.1% of white individuals reported not being able to visit a physician because of cost. Yet 14 041 low-income white individuals experienced limited access to a physician because of cost compared with 3402 black individuals and 4239 Hispanic individuals. Improving overall outcomes and narrowing or eliminating these gaps in outcomes and determinants is the primary goal of population health policy. This means increased and balanced investments across all determinants of health: health care, behaviors, socioeconomic factors, and the physical environment. Unhealthier individuals and populations need greater investment but not to the exclusion of others; all populations have the potential for better outcomes.
“Proportionate universalism,” an investment concept introduced by Benach and colleagues,4 might help address the rate, burden, and race issues in both mortality and morbidity. This approach proposes universal investment policy across all social groups (addressing burden) and increasing targeted benefits for groups with poorer health (addressing rates). Benach et al describe this approach as “universal policies that explicitly incorporate criteria to increase resource allocation to populations with increasing needs.” In practice, this would mean that a certain minimum level of resources for health determinants, such as tobacco control, reducing child poverty, and removing environmental toxins would be appropriate for the entire population, but the level would need to be increased for subpopulations with higher rates of an outcome above the baseline burden level. For example, a basic universal per capita dollar amount for tobacco control or early childhood programs would involve graduated higher amounts for minority or lower-education individuals or subgroups with poorer rates in these areas. The balance of universal and targeted approaches would vary by health outcome and the specific group with higher rates of concern. Considerable cost-effectiveness research will be needed to move this concept into practice, but it is an absolute underpinning of any realistic population health equity policy effort.5
The effort to reduce health disparities is hindered by viewing health equity only in terms of racial inequities. Such a view can make it too easy for some policy makers to perceive and act as though high rates of poor health by race are not a primary concern to them and their constituents because they represent primarily white citizens. However, individuals who are white and those who are ethnic minorities are both affected by the health equity gap together. This understanding could lead to more common ground for the substantial investment and policy reforms that alone will improve overall health and reduce these unacceptable gaps.6 According to Schroeder,7 “This national health disparity is not simply a factor of the multicultural nature of American society, because it persists when the health of whites only is compared to the more racially homogeneous OECD [Organisation for Economic Co-operation and Development] nations.”
The health burden of poor white individuals derives in part from a different but related social policy history than that for racial minorities. Isenberg8 has suggested that these policies derive “from British colonial policies dedicated to resettling the poor, decisions that conditioned American notions of class and left a permanent imprint” and that “we can no longer ignore the stagnant expendable bottom layers of society in explaining the national identity.” Her book chronicles the underappreciated 400-year history of such policies beginning with the colonists importing British paupers and children as indentured servants, discriminatory land purchasing and voting rights policies, and the early 20th century rebirth of the eugenics movement.
Intentional and unintentional discriminatory social policies that have contributed to the gaps in health outcomes by race are unacceptable and deserve ongoing attention. But universal social policies and investments in education, income, employment, health care access, and public health will help reduce inequities of all races and classes. At the same time, it is important to acknowledge that poorer minority rates occur in all socioeconomic levels and that “race-blind” approaches can increase inequity because they often benefit white individuals more quickly and disproportionally.9
High rates of poor health are critical indicators for policy priority, but equal attention needs to be paid to the relative numbers of individuals experiencing this burden. Seeking support for health policies that help poor people of all races could garner more broad-based support than policies that solely address racial gaps. It should be possible to find common ground for the CDC’s health equity goal of every person having “the opportunity to attain his or her full health potential and no one being disadvantaged from achieving this potential because of social position or other socially determined circumstances.”10
Corresponding Author: David Kindig, MD, PhD, University of Wisconsin-Madison, Population Health Institute, 610 Walnut St, Room 550 WARF, Madison, WI 53705 (firstname.lastname@example.org).
Published Online: January 9, 2017. doi:10.1001/jama.2016.19435
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Additional Contributions: I appreciate helpful comments from several colleagues and friends, as well as data analysis by Nicholas Lardinois, MPPA, and Keith Gennuso, PhD, University of Wisconsin-Madison, Population Health Institute, who were not compensated for their contributions to this article.
Kindig D. Population Health EquityRate and Burden, Race and Class. JAMA. Published online January 09, 2017. doi:10.1001/jama.2016.19435