Inequities in the Mortality Rates of Older Americans—Race, Sex, Place, and Time

In 2015, after nearly a century of annual increases, US life expectancy began 3 consecutive years of decreases associated with increased mortality rates among young and middle-aged adults of all races.¹ This reversal forms an important context for the analysis by Ferdows and colleagues,² whose study examines county-level mortality trends between 1968 and 2016, focusing specifically on older adults (≥65 years). Their analysis examines differences in sex and race (Black vs White ancestry) overall and across rural and urban areas.² They find that overall age-adjusted mortality among those 65 years and older has decreased substantially for all race and sex groups, with relative gaps changing over time. These gaps increased between Black and White adults after the early 1980s and decreased again after 2010. By looking at urban vs rural differences, which presents a novel analysis, Ferdows et al² report greater mortality decreases in urban areas compared with rural ones and, despite the overall reductions in mortality, an increasing racial gap for older men in rural counties.

But what are we to make of these findings? The authors conclude that public health interventions should focus on addressing the increasing racial disparities in rural regions between older Black and White adults. Yet, by looking primarily at trends in the racial gap for older Americans, adjusted for time-invariant state-level factors and controlled for annual county-level socioeconomic factors, the authors do not fully examine the health advantages of White individuals of all ages. Ferdows et al² do not provide sufficient information on absolute numbers to assess the association between the increasing gap among Black and White individuals living in rural communities and racial inequities overall.² With Black Americans concentrated in urban settings and with a substantially younger age distribution among Black Americans compared with White Americans (indeed, early death means that older people [≥65 years] comprise a smaller proportion of the Black population [8.2%] than they do in the White population [19.0%]),³ a call for increased attention to older Black men living in rural areas may do little to address the racial injustices that exist in the United States. There is value in bringing visibility to the disproportionate mortality rates among older Black people in rural areas, but conditions in these areas are not the most important factors in reduced life expectancy among Black individuals. Indeed, any long-term trend analysis of racial disparities in mortality rates for older Americans should acknowledge that life expectancy for Black men did not reach 65 years until the 1990s.⁴

Ferdows et al² use state-level fixed-effects analyses to account for time-invariant state-level factors and select overall (ie, not disaggregated) county-level socioeconomic factors, but these methods account for neither time-variant state-level associations nor selection by geographic region. The reliance on state-level fixed-effects analyses has implications for the interpretation of the reported findings because most Black Americans (58%) live in the South,⁵ and almost 90% of Black Americans in rural counties or small towns live in the South.⁶ Ferdows et al² did not disentangle nationwide racial gaps in mortality among those 65 years and older from the potential region-specific consequences of time-variant trends in the dismantling of the social safety net in the Deep South. Without accounting for the absolute demographic factors of older Black and White men living in rural areas as well as the geographic concentration of Black Americans, the conclusions drawn from this analysis will likely be incomplete. The racial differences noted in this analysis among men 65 years and older living in rural areas may be associated with temporal changes in the South and the relative segregation, isolation, and lack of visibility of Black people in rural areas rather than the factors listed. The authors acknowledge the importance of assessing US mortality rate trends at the intersection of multiple identity factors (eg, race, age, and sex) without discussing the inequitable policies and systems that underlie this importance.

The authors suggest biomedical associations for the increase in mortality among older Black Americans starting in the 1980s, citing the increase in stroke and heart disease as well as other chronic conditions and health care access, and do not address the potential associations of structural racism, poverty, and inequality. Although it can be argued that health care access becomes more salient for survival among those older than 65 years, a discussion of health care access is insufficient to an understanding of why Black and White differences in mortality among those older than 65 years differ over time and by level of rurality without the historical social, economic, and political context associated with socioeconomic status, access to resources, and health care access. Krieger et al⁷ conducted a study of premature mortality (ie, deaths before age 65 years) trends over a similarly long period (1960-2002) and found similar patterns in changes in the relative extent of racial gaps; racial gaps initially decreased in the 1960s, increased beginning in the 1980s, and decreased again in the 2000s. Unlike Ferdows et al,² Krieger et al⁷ reviewed broader shifts in the national policy, including the reversal of the 1960s Great Society programs, to discuss the stalled mortality decrease beginning in the 1980s and the return to an increased racial mortality gap.⁷ In fact, since the early 1980s, life expectancy in the US began to diverge from the upward trajectory of its peer nations.⁸ If the US had sustained the average mortality rates of other Organization for Economic Co-operation and Development nations, thousands of deaths may not have occurred.

In short, to fully understand temporal trends in the racial gaps in age-adjusted mortality among older Americans, the interpretation and implications of this analysis will need to be historically rooted, and social, economic, and political dynamics must be incorporated into the research design. Furthermore, implications regarding where and how to act must include consideration of both trends and absolute inequities.

Published: August 3, 2020. doi:10.1001/jamanetworkopen.2020.12437

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Bailey Z et al. JAMA Network Open.

Corresponding Author: Mary T Bassett, MD, MPH, Francois-Xavier Bagnoud Center for Health and Human Rights, Harvard University, 651 Huntington Avenue, 7th Floor, Boston, MA 02115 (mbassett@hsph.harvard.edu).

Conflict of Interest Disclosures: None reported.