The rapid increase in the criminal justice population has important health implications for justice-
involved individuals, their families and communities with the most vulnerable populations at greater risk. A
recent review found elevated mortality among those released from prison relative to the general population,
suggesting that incarceration may have lasting consequences even after individuals return to the community.
The implications for health disparities are crucial given that some groups (e.g. African Americans, Native
Americans) have higher rates of incarceration and experience excess mortality relative to Whites. Yet, we have
little understanding of the degree to which criminal justice exposure elevates mortality risk and whether this
relationship is stronger among minority groups. This study will provide rigorous quasi-experimental
evidence on how criminal justice exposure influences mortality and disparities using longitudinal data from
a statewide census of arrestees assigned to prison, jail, community supervision, or released.
Our study leverages a natural policy experiment in a difference-in-difference design with propensity score
weighting and instrumental variables combined with rare comprehensive data on criminal history and
mortality. South Dakota's Public Safety Improvement Act of 2013 (hereafter, SB70) reduced prison sentences,
increased assignment to probation, re-classified felony offenses to lower severity classes, and allowed those on
community supervision to earn reductions in time on supervision. Essentially, those arrested post-SB70 were
subject to different penalties and supervision relative to those arrested pre-SB70 for the same offenses, based
only on timing. Our study utilizes this policy experiment for the exogenous variation needed to assess the
relationship between criminal justice exposure and mortality. An intrinsic aspect is to assess disparities by
race/ethnicity and sex with a unique opportunity to provide evidence on Native Americans. For this project, we
have obtained criminal history data on all 202,594 arrestees (2000-2019) as well as matched mortality records.
In summary, this study will provide much-needed evidence on excess mortality among arrestees
sentenced to prison, jail, community supervision or released overall and by demographic group; assess whether
SB70 influenced the risk, timing and causes of mortality overall and by demographic group; and finally,
estimate how criminal justice exposure influenced the risk, timing and causes of mortality overall and by
demographic group. Understanding how exposure to the criminal justice system influences mortality is
particularly policy-relevant given the mass incarceration in the US and states' efforts to reduce its impact on
budgets and society. Disparities in criminal justice exposure mirror disparities in mortality, suggesting that the
two phenomena may be intertwined. Thus, this study is not simply an evaluation of SB70, but rather a quasi-
experimental assessment of the impact of criminal justice exposure on public health and health disparities.