Premature Death in the United States as a Factor of Lifestyle, Healthcare, and Socieconomic Factors

Using the 2019 County Health Rankings

Motivations

Developing a better understanding of what factors cause premature death is a priority in the United States. By identifying the largest, most significant causes of premature death, American policymakers can focus on improving the well-being of constituents through new social programs, incentives, and regulations.

2019 County Health Rankings

The County Health Rankings & Roadmaps program is a joint partnership between the Robert Wood Johnson Foundation and the University of Wisconsing Population Health Institute. The annual County Health Rankings measures health indicators across America at the county level. While the initiative provides its own analysis and exploration of the data, this analysis uses a comma-separated value (CSV) export of the 2019 County Health Rankings raw datasheet. The data can be found here.

Years of Potential Life Lost

Years of Potential Life Lost (YPLL) is the amount of years an individual died prior to the mean life expectancy, set for the 2019 County Health Rankings at 75. For example, an individual dying at age 50 have 15 years of potential life lost. The benefit of using YPLL as the response variable is that it accounts for age discrepancies in a population. Given the same amount of deaths, a region with an older population will account for fewer years of potential life lost. This allows for a better assessment of which predictors lead to a premature death.

Predictors of Premature Death

In an effort to model the reality many Americans face, this analysis attempts to identify the primary socioeconomic, lifestyle, and healthcare indicators that could contribute to premature death at the county level. These variables were selected out of the extensive list of the 2019 County Health Rankings as the most representative and versatile indicators while accounting for multicollinearity. As will be discussed in the causal model, the data is log transformed prior to the multivariate analysis. For a detailed explanation of the variables, please visit the County Health Rankings webpage on 2019 Measures. The variables used are:

Years of Potential Life Lost

Percent Smokers

Percent Physically Inactive

Percent Excessive Drinking

Primary Care Physician (PCP) Rate

Mental Health Rate

Preventable Hospital Stays

High School Graduation Rate

Percent Unemployed

Percent Children in Poverty

Social Association Rate

Injury Death Rate

Average Daily Pollution (2.5Pm)

As explained in the causal model analysis, the relationship between independent variables and the Years of Potential Life Lost per Capita is unclear. Caution is suggested in making policy recommendations of this data.

For More Information

JoeBank values discourse and critical responses to their work to improve the lives of Americans. Please contact the lead researcher, Joseph Conran with any questions or comments. He can be reached at conran[at]adnumerant.com