The U.S. has seen substantial improvements in life expectancy over the past century, particularly for those who are better-educated and more affluent.

Our study, out today, looks at the health of older Americans in recent years, using data collected by the U.S. Department of Health and Human Services on more than 50,000 seniors age 65 and older. Seniors in 2014 were 14 percent more likely to report that they were in very good or excellent health, compared to seniors in 2000.

However, a closer look tells a worrisome story: The health divide is widening across socioeconomic groups. Gains in good health primarily went to more advantaged groups.

Our work reveals a health disparity echoed in reports by others. In 1980, a wealthy 50-year-old man could expect to live an additional 5.1 years longer than a poor man of the same age. Thirty years later, the life expectancy of two similar men differs by more than a dozen years.

Measuring older Americans’ health

Health can be measured in many different ways. While physical measures such as weight, blood pressure and cholesterol level are excellent markers, it simply isn’t practical to obtain such information in studies that include many thousands of subjects.

Most research to date looks at trends in older adults who are frail or in otherwise poor health. These studies show that disability rates among older adults have declined by 1 to 3 percent every year since the 1980s.

Since seniors have such complex health needs, it may seem to make intuitive sense to track poor health. But this provides only one perspective. In our opinion, examining only those in poor health neglects to consider how good health – the goal of public health initiatives – is distributed in the population. Using disability trends to evaluate the health of older Americans is analogous to making conclusions about the U.S. economy based solely on the poverty rate.

In fact, when disability is used to examine health disparities, it leads to mixed conclusions. For instance, in comparing whites to blacks, one report showed a decline in the disability gap throughout the 1990s, while another showed an increase starting in the 1990s and extending to 2006.

Our work

It turns out that a single question about health is actually very accurate at estimating an individual’s likelihood of dying: “In general, would you say that your health is excellent, very good, good, fair or poor?”

By focusing on good health rather than poor health, we can think of health as an asset much like wealth, where the goal is to be at higher levels. We found that, by taking this new approach, health disparities among seniors became strikingly clear.

We identified seniors who reported “excellent” or “very good” health from 2000 to 2014. Our results show that seniors are more likely to report being healthy in recent years. In 2014, there were 8.4 million more healthy seniors in the U.S. than in 2000.

However, the gains in health were primarily among non-Hispanic whites and those of higher educational backgrounds or high family income. For instance, between 2000 to 2014, the number of seniors reporting good health increased by 21 percent among non-Hispanic whites. Meanwhile, during the same period, good health decreased 17 percent among non-Hispanic blacks.

Likewise, good health increased by 10 percent among seniors who possess a graduate degree. It also increased by 23 percent among seniors of high family income – that is, whose income was greater than or equal to four times the federal poverty level. Less advantaged counterparts – including those of high school education or less and with family income near or below the federal poverty level – were not so fortunate.

What explains these growing disparities? Given that the individuals in our study were all eligible to participate in Medicare, our results suggest that the influence of social, economic and environmental factors extends beyond access to health insurance.

This suggests that public health initiatives may miss some intended audiences. For instance, the U.S. government’s Healthy People 2020 initiative aims to manage risk factors associated with cardiovascular disease, a leading cause of mortality. However, a recent report shows that public health gains have benefited the wealthy more than the poor.

The idea that higher education may affect an individual’s personal “investment” in themselves by partaking in healthy behaviors can be traced back to the concept of human capital described by Nobel laureate Gary Becker. Income and education are closely related, and their effect on health can last for a lifetime.

Why these disparities matter

By 2050, the population of older adults is expected to nearly double. Older adults’ health will have a significant impact on the national economy, as they will use more health care resources and may stay in the workforce longer.

Furthermore, the growing divide in health suggests that there are at least two different Americas. Depending on where an individual sits on the socioeconomic spectrum, he might expect a different length and quality of life.

Differences in life expectancy are particularly important as policymakers consider potentially raising retirement age for Social Security or the eligibility age for Medicare. In light of this disparity, such efforts to make federal programs financially sustainable would pay out less in the long run to lower income groups.

Indicators point to greater improvements in the length and quality of life among the most privileged groups in the U.S. This raises important questions regarding how we might design better health systems so that all members of society can benefit.

Dr. Davis was supported by a grant from the National Center for Complementary and Integrative Health (R01 AT009003). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Kenneth Langa receives funding from the National Institute on Aging and the US Department of Veterans Affairs.