Death and disappearance: Measuring racial disparities in mortality and life expectancy among people in state prisons, United States 2000-2014.

backgroundResearch on carceral institutions and mortality finds that people in prisons and jails have a high risk of death immediately following release from custody and that while incarcerated, racial disparities in prisoner mortality counter observed death patterns among similarly situated non-incarcerated, demographic groups. Yet, many of these studies rely on data prior to the millennium, during the COVID-19 pandemic, or are relegated to a small number or select group of states. In this paper, we explore changes in mortality and life-expectancy among different demographic groups, before and after the Great Recession, across forty-four states that reported deaths in custody to the federal government between 2000 and 2014.

methodsDrawing on a novel dataset created and curated, we calculate standard, age- specific quantities (death rates and life-expectancy) using period lifetable methods, disaggregated by race and sex, across three different periods (2000-2004, 2005-2009, and 2010-2014) for each state. Ordinary least squares regression models with state and year fixed-effects are included to examine state-level factors that may explain differences in prisoner mortality rates between 2000 and 2014. We also benchmark death counts reported to federal agencies with official state reports to cross-validate general mortality patterns.

resultsAmong imprisoned men, age-specific trends in mortality have shifted across the three periods. Following the Great Recession and the push for criminal justice reforms, prisoner mortality dropped significantly and is concentrated at older ages among men during 2010-2014; the shifting pattern of mortality means that men age 30 in 2010-2014 had similar death rates as men in their early 20s during 2000-2004, representing a 7.5 year shift in age-specific mortality rates. Gains in the mortality decline were disproportionately experienced by Non-Hispanic White and Non-Hispanic Black men, with the latter experiencing the greatest gains in life-expectancy of any demographic group. State-level violent crime rates are strongly and positively associated with prison mortality rates across states, net of socioeconomic and political factors. The large and significant disappearance of deaths in prisons from official data reported to federal agencies calls into question the narrowing gap in racial disparities among people in carceral facilities.

conclusionsLegal decisions and social policies aimed at reducing mortality may be most effective in the short-run; however, the effects of these policy changes may fadeout over time. Research should clearly discern whether changes in mortality rates across states are due to diminished gains in social policies or increases in the disappearance (or underreporting) of deaths in custody. Understanding how and why gains in survivorship may stall is important for aligning health initiatives with social policy to facilitate maximal and consistent mortality declines for all demographic groups.