Tweets by @Stanford_Daily


Research ties death rates, social factors

Internal Medicine Professor Mark Cullen and his interdisciplinary research team published a study last week that illustrates the relationship between death rates and socioeconomic and environmental variables. Cullen and his research team focused the study on 22 socioeconomic and environmental variables that together prove to be better indicators of early death than race or geography.

A few years ago, Victor Fuchs, professor of economics and health research and policy, contacted Cullen to become involved in research assessing trends in mortality rates. As a result, Cullen launched the study with Fuchs.

“The study was a self-initiated idea,” Cullen said. “We wanted to make some sense of the enormous scope of disparity of death in the country and come up with a unified explanation.”

U.S. life expectancy varies widely by geographical location, race and other contextual factors, and Cullen’s team sought to make some sense of the trends.

One way of assessing death rates is to examine mortality statistics. However, Cullen said that these indicators often don’t illustrate specific details about those who died. Instead, studying the lifestyles of different communities and individuals can serve as important information to understand the determinants of mortality rates.

“Interdependence is always a question,” Cullen said. “Variables such as income, education and marital status are huge predictors for accounting for death rates. When you assess the latter factors in correlation to behavior, such as diet and smoking, they demonstrate a heavy dependence.”

Every 10 years, national census data accounts for population numbers, geographical distribution and other statistical factors. About one in 20 families gets a long form with detailed questions about lifestyle factors, such as income, house ownership, marital status and education. With this information, along with statistics about death rates from various counties, the study discovered the extent to which different variables are relevant to death rates.

In addition to census data, the research team looked further into the multidimensional factors that impact populations and correspond to disparities in health outcomes.

“Equally high-quality data can be drawn from publicly available data sets,” Cullen said. “For example, from [the Environmental Protection Agency], you can look at data about air pollution; from the Oceanic Institute, you can get information about climate; from various sources, you can look at ratings about hospitals; or you can use surveys from [the Centers for Disease Control] about population behavior.”

The study incorporated widespread public data, assessing approximately 570 counties that have equal racial distributions of whites and blacks. Despite the diversity of the locations, transparent trends emerged from the data.

“Many social factors are interrelated; however, what’s not interrelated is geography and race,” Cullen said. “Where black people enjoy high levels of education, they do as well as white counterparts. However, for gender there is a difference. Women can have the same education and resources, yet do much better than men. This is something we are still trying to explain.”

The study returned clear trends in its results, which may help to advance public conceptions of welfare and healthcare necessities.

“Think about how different each city is, from the Bronx to Los Angeles, from San Francisco to Atlanta; yet every country falls into same blanket,” Cullen said. “Once you know the major determinants, you can almost always predict the death rates.”

Cullen and his research group plan to expand the study and look at transnational trends of death rates. On a domestic level, however, the research points out important information for national healthcare priorities.

“In the national debate about changing the healthcare system, if you want to level the playing field for health, you should invest in education, housing, job quality and social welfare, because those are where the differences in death rates are made up,” Cullen said.

As these results indicate the influence of social policy on health, the research findings may have implications for the political domain.

“All policy is actually health policy — the environment, making housing better, improving the job market,” Cullen said. “From the point of view of public health, all investments affect public health. If you want to help communities maintain better health, it is important not just to run and get more doctors; rather it is critical to improve the social fabric.”