A Multilevel Approach to Understanding Mass Incarceration and Health: Key Directions for Research and Practice

After decades of punitive social policies driving up incarceration rates, the number of lives affected by the US criminal legal system is unprecedented. In 2016, 6.7 million people were in jail, in prison, or on probation or parole, more than double the number in 1986.¹ Mounting evidence suggests that this has dire consequences for population health beyond its immediate impact on incarcerated people. Nearly all incarcerated people are taken from family members and communities, but data collection efforts and theoretical frameworks that appreciate the widespread impact this may have are alarmingly lacking. Public health researchers, practitioners, and activists have a responsibility to examine and communicate the widespread health consequences of mass incarceration for all people and communities, as well as how these effects are inequitably distributed.

MULTILEVEL FRAMEWORKS

Sociological and criminological research point to incarceration’s negative consequences for families and communities, including those related to family functioning, parenting capacity, employment markets, and—given the disenfranchisement of many formerly and currently incarcerated people—electoral politics.² Parental and partner incarceration are associated with the increased risk of several mental and physical health problems, including cardiovascular disease, depression, and anxiety.^3,4 Of relevance to many health outcomes, the economic costs of family member incarceration can be burdensome, including legal fees and fines, health care services copayments, and costs associated with visiting and communicating with incarcerated people.³ The stigma associated with family member incarceration can isolate some individuals from their social networks, blocking the social support that might buffer stressors related to family member incarceration.⁴

Less research attention has been given to the health impacts of mass incarceration on incarcerated people’s communities. Several studies indicate that high incarceration rates in an area change sexual networks and increase rates of sexually transmitted infections.² It was also recently reported that living in a zip code with a higher prison admission rate is associated with an increased risk of depression and anxiety among nonincarcerated people.⁵ But large gaps persist in this literature.

Multilevel social epidemiologic frameworks and hierarchal modeling strategies are useful for research on the community health consequences of mass incarceration because they can (1) expand the focus beyond the health of incarcerated people, and (2) draw attention to the additional social, policy, and geographic contexts contributing to how mass incarceration shapes population health. Equally important, multilevel perspectives can center structural racism and inequitable distributions of incarceration across race/ethnicity, gender, and social class and their intersections. For example, Foster and Hagan’s multilevel social exclusion theory proposes that deliberate social policy decisions systematically disconnect children with incarcerated parents from opportunities for healthy development.⁴ Ecosocial theory⁶ can also be applied to understand racial/ethnic inequities in incarceration rates as the product of discriminatory policies and practices (including the war on drugs, “three strikes” laws, and “broken windows” policing) and as determinants of racial/ethnic health inequities.

Multilevel perspectives, too, allow researchers to link incarceration to population dynamics and social policies that collectively affect individual and community health. The criminal legal system, including jails, prisons, and policing practices, is just one of several interlocking institutions that act punitively to poor and, disproportionately, Black and Latinx people. To fully examine the impacts on population health and health inequities, it is critical to understand the relationships between incarceration, employment policies, the Supplemental Nutrition Assistance Program, the Special Supplemental Nutrition Program for Women, Infants, and Children, the foster care system, and other social policies.

Given the way incarceration intersects with many other determinants of health, for both incarcerated and nonincarcerated individuals, it follows that incarceration might matter for critical public health research and practice considerations; these include equitable program and policy implementation, study retention, and treatment adherence. However, the extent of such effects is impossible to know unless researchers and practitioners link health and incarceration data at multiple levels.

DATA GAPS AND OPPORTUNITIES

Despite the importance of incarceration as a social determinant of health and its relevance for public health practice, data resources on incarceration and policing for both individuals and geographic areas are limited. Few national public health surveys ask about jail, prison, or arrest histories or include and retain sufficient numbers of formerly incarcerated people and their families. Data sets that ask retrospectively about incarceration histories rarely ask about the precise timing or duration of multiple incarcerations, making these resources difficult for precise covariate control and consideration of etiologic lags. Questions about individual and family member incarceration and police interactions should be added prospectively.

Recently available data from the Vera Institute of Justice provide, for the first time, national jail and prison incarceration rates at the county level from the 1970s and 1980s to 2015. This resource can be easily linked to individual- or county-level health data. Although counties are a relevant geographic level for the determinants of incarceration rates (e.g., local laws, policing, and judicial practices) and there is marked variation in county-level incarceration rates, aggregation at the county level is too high to reveal neighborhood or social network impacts. Incarceration rate data at geographic levels below the county level (e.g. census tracts) are available for only small areas and only over a few years and are usually acquired through agreements with local departments of correction, which are often hesitant to release it. Expanding national data resources at lower geographic areas is necessary for advancing research on the population health consequences of mass incarceration.

There are also promising policy changes and activist movements aimed at ending mass incarceration under way. Ending cash bail, diversion programs that prevent people from being charged or tried and changing school discipline policies, for instance, could have profound immediate and long-term implications for the health of individuals and communities. Linking health data to these policy changes is essential for their evaluation.

HOLDING GOVERNMENTS ACCOUNTABLE

Evaluating the health impacts of incarceration on families and communities prompts the question of whether mass incarceration is actually improving the public’s well-being, let alone public safety. The literature challenges who counts as “the public” in claims about incarceration as a tool to protect the public, and, when seen through the lens of structural racism, it becomes clear that these claims have too often been invoked for racially unjust ends. Public health perspectives that center racial justice and health equity can be used to imagine alternatives to mass incarceration. Rather than addressing harm after it has happened, they can identify less punitive solutions to preventing crime that consider the poverty, racism, and political disenfranchisement that cause many individuals to turn to crime as a means of survival.

Public health researchers, practitioners, and activists should confront policymakers with the research on population health and health equity impacts of mass incarceration. For example, the American Public Health Association’s policy statement on law enforcement violence⁷ is a research-informed advocacy tool that can be replicated for other dimensions of mass incarceration, including cash bail, construction of new jail or prison facilities, and voting rights for currently or formerly incarcerated people. For research, multilevel analyses and frameworks are useful, although more transparent data sharing at lower geographic levels is needed. Both research and advocacy matter for holding state, local, and federal levels of government accountable for health impacts incurred by policies that contribute to high incarceration rates and racial inequities in these rates and for reckoning with the impact of structural racism on racial inequities in health.