The Outsized Impact of Behavioral Causes of Death on Longevity in the United States: Comparing Total and per Death Years of Life Lost

Deaths Due to Suicide, Homicide, Accidental Poisoning, and Traffic Accidents in America

Long life involves more than avoiding disease. Americans also die of suicide, homicide, drug overdoses, and traffic accidents. In the pre-COVID-19 census year 2010, for example, these four behavioral causes were responsible for 3.5 percent of all deaths for non-Hispanic Whites and 3.2 percent of all deaths for non-Hispanic Blacks (hereafter “Whites” and “Blacks”).

Although deaths from behavioral causes represent only a small fraction of all U.S. deaths, there is more to the story, as victims of behavioral causes tend to be younger than victims of other causes (Case and Deaton 2015; Firebaugh et al. 2014). To adjust for those age differences, we examine years of life lost, a complement of life expectancy. Years of life lost is derived from age- and cause-specific mortality rates for some age range (Andersen, Canudas-Romo and Keiding 2013). For example, average years of life lost per death from birth to age 90 is 15 years when individuals live an average of 75 years between birth and age 90. For a specific cause of death, average years of life lost per death is 90 minus the average age at death for victims of that cause.

In this analysis we use cause- and age-specific multiple decrement life tables for the United States in 2010 to contrast total and per death years of life lost for the four leading behavioral causes of death (homicide, suicide, drug overdose, traffic accidents) with the two leading overall causes of death (heart disease and cancer). We report results for the age span 0 to 90 years; similar conclusions obtain for ages 0 to 85 years and 0 to 95 years. We chose a census year to provide the most accurate population estimates, and we selected 2010 because it is more representative of U.S. mortality than the COVID-19 census year 2020.

Findings

For each cause of death, we express years of life lost as the proportion of cancer’s years of life lost. Although heart disease is responsible for more deaths in the United States, cancer is the more appropriate yardstick here, as cancer is responsible for more years of life lost and thus has a greater impact on longevity, our focus here.

Figure 1 highlights the differences between total and per death years of life lost because of the major behavioral causes of death in the United States. For both Blacks and Whites, total years of life lost because of cancer and heart disease greatly exceed years of life lost because of homicide and other behavioral causes. But the average life years lost for a single death from a behavioral cause is nearly three times greater than a single death from cancer and three to four times greater than a single death from heart disease. Moreover, Blacks and Whites differ very little in the per death impact of homicide, suicide, and accidental poisoning on longevity, despite substantial Black-White differences in the total impact of those causes of death on longevity.

OPEN IN VIEWER

Figure 1.

Impact of behavioral causes of death on longevity in the United States in 2010: total versus per death.

Conclusion

A death averted is a life prolonged. With respect to longevity in a population, however, not all averted deaths are equal. Averting death at age 35 years has a greater impact than averting death at age 85 years. By employing a method that adjusts for this fact, we present visualization of the extent to which homicide, suicide, traffic accidents, and drug overdose have an outsized effect on longevity in the United States. Because these causes of death are within the purview of social science research (Acciai and Firebaugh 2017; Case and Deaton 2015; Harper, Riddell, and King 2021; Woolf and Schoomaker 2019), social scientists have a central role to play in pinpointing the determinants of longevity in America, and in how these determinants shift over time.

Supplemental Material