The U.S. Criminal Legal System and Population Health

Abstract

Purpose of Review:

A growing body of research has now identified the criminal legal system as a major social determinant of population health and health disparities in the United States. The current review provides a description of the U.S. criminal legal landscape, briefly summarizes recent research findings, and identifies new and needed directions for research.

Recent Findings:

Building on prior research first identifying direct contact with the prison system as a social determinant of health, recent research provides evidence of elevated risks for physical and mental morbidity and mortality among those with direct or indirect contact with the criminal legal system. This research has been expanded to include community supervision and contact with police as important drivers of health.

While this evidence base has grown, our understanding of the role of the criminal legal system in population health has remained limited for several reasons: (1) prison and jail incarceration remain the primary forms of contact examined despite the existence of other relevant forms of carceral contact and control; (2) the longitudinal health consequences of contact with the criminal legal system have largely gone undocumented; (3) the majority of the research is descriptive and does not employ causal modeling approaches; and (4) relatedly, the mechanisms that link criminal legal system contact with health are not adequately measured.

Summary:

The criminal legal system has emerged as a significant social determinant of health. While advances have been made in documenting the direct and indirect consequences of contact for population health and health disparities, more work is needed to better ascertain how and why this contact matters.

Introduction

While prisons and jails have long been sites of public health research on infectious disease, mental health, and illicit drug use, incarceration and other forms of criminal legal contact have only recently been studied as social determinants of health [1]. In the past decade, critical advancements have been made to expand the types of contact studied and the populations impacted. As an illustrative example, the killing of unarmed Black men and women at the hands of the police, and the media, public, and political attention to these deaths [2], have further amplified the direct public health consequences of policing as a form of criminal legal contact. We now have new evidence of large racial disparities in the risk of being killed by police, such that 1 in every 1,000 Black men can be expected to be killed by law enforcement over the life course, compared to 1 in every 2,500 White men [3].^1,2 In response to these and other advancements, there have been several high-level reviews documenting the relationship between the criminal legal system and population health [4–7•]. Our review is the latest supplement to these works and their omissions, with a focus on providing a description of the contemporary U.S. criminal legal system landscape, identifying the gaps in the research drawing from examples of recent literature published on this topic (namely within the last five years), the reasons for these gaps, and needed directions to not only better understand the breadth and scale of the criminal legal system but also its impacts on population health and health disparities. In doing so, we organize our review around the following observations: (1) prison and jail incarceration remain the primary forms of contact examined despite the existence of other relevant forms of carceral contact and control; (2) the longitudinal health consequences of contact with the criminal legal system have largely gone undocumented; (3) the majority of the research is descriptive and does not employ causal modeling approaches; and (4) relatedly, the mechanisms that link criminal legal system contact with health are not adequately measured.

Setting the Criminal Legal System Landscape

The U.S. criminal legal system, or “carceral state”, is a vast complex of formal institutions³ that span policing, prosecution, courts, and corrections [10]. Here, we would like to emphasize the point of breadth—that the criminal legal apparatus is not a unitary system per se but rather multiple systems. For example, while much attention has focused on the unprecedented rise in U.S. federal and state mass imprisonment from the 1970s to peaks in the mid-2000s [11, 12], of the 1.9 million people most recently incarcerated in 2024, we estimate 40% were confined in some 5,000 other local jail, juvenile correctional, immigrant detention, Indian country jail, military prison, civil commitment, and psychiatric hospital systems that are often absent from surveys of carceral institutions [13]. It is also important to consider how the criminal legal system “widens its net” through community supervision [14]. Supervisory programs with daily-living stipulations like probation (a type of court-mandated sanction for criminal offenses seen as an alternative to incarceration) and parole (a conditional release from prison into the community) suggest the true population under carceral control occupies an even larger space—approximately 3.7 million adults were on probation or parole on any given day in 2022, compared to 1.2 million adults incarcerated in prison and 660,000 in jail [15–17]. Moreover, factoring how between 25 percent and 40 percent of Americans have been arrested by age 23, and as many as 1 in 3 have a criminal record [18], the scale of the criminal legal system and who it reaches appears staggering. Indeed, recent evidence using data from the Family History of Incarceration Survey (FamHIS) reveals that almost half of Americans have had an immediate family member incarcerated [19].

Our point on the breadth and scale of the criminal legal system is reinforced by how it interacts with (non-criminal) institutions. Civil and administrative institutions, such as family and immigration courts, and civil detention facilities, though not necessarily acting on criminal law, can impose punitive sanctions and compulsory detentions onto individuals—even in lieu of criminal convictions [20]. A good example is how the unauthorized migration of non-criminal immigrants into the U.S. is classified as an administrative violation (of immigration law), yet can result in incarceration (in an immigration detention facility) as punishment. On this, Beckett and Murakawa (2013) use the term institutional annexation to describe how sites and actors become extensions of the overall criminal legal system (i.e., in their words forming “the shadow carceral state”) and the construction of punishment in the United States. Institutional offloading represents another parallel process. Sirois (2023) for instance, analyzing ethnographic data from a California juvenile court and interviews with court actors, describes how state institutions, such as “Social Services” on one end and “Probation” on the other, attempt to offload responsibility for unwanted crossover clients to each other [21]. This closeness of dealings the criminal legal system has with other social institutions like the welfare state, and how its activities are perceived as fungible, is finally suggestive of a third process we call institutional infiltration. We see this visibly with the increased presence of police patrolling the hallways of educational institutions (i.e., poor-performing, racially segregated urban schools), contributing to what has been dubbed the “school-to-prison pipeline”, and early life course exposures to criminal justice [22]. Together, we find these institutional processes illustrate the pervasiveness of the criminal legal system today. Rather than operating in isolation, it now frequently exchanges with, spills into, or replaces surrounding social institutions [23•].

Observations on Criminal Legal Contact and Population Health

So, what does the current state of the criminal legal system as we have presented mean for population health? Ideally, it means research treating criminal legal contact as a social determinant of health should be amenable to the broad reach of the criminal legal system itself, and its interconnectedness to other institutions—all of which implicate more people, and more mechanisms influencing health, than it would initially seem. Likewise, we make three more observations about broadening the substantive and methodological horizon of research linking criminal legal contact and health, which we describe in the remaining sections.

Beyond Prison and Jail Incarceration to Other Forms of Carceral Control

The prevailing body of research investigating the criminal legal system and population health focuses on prison and jail incarceration,⁴ as well as short-term postrelease effects⁵ following confinements in these carceral systems. Because of this, we can say rather definitively, that compared to the general population,⁶ the prison-and-jail incarcerated are susceptible to higher rates of communicable diseases (e.g., sexually transmitted diseases, HIV, and hepatitis C), chronic conditions (e.g., hypertension, diabetes, and asthma), substance use, mental health disorders, and mortality (including suicide) [1, 4, 5, 7•, 26•]. New streams of research, moreover, continue to address gendered [27], racial [28], and age-related [29] disparities systemic in our prisons and jails, and, how these carceral institutions and their internal conditions drive poor health outcomes. Curran et al. (2023), for example, found use of prescription medication for chronic conditions was consistently lower in prisons and jails relative to community settings, reporting a 2.9-fold relative disparity for type 2 diabetes, 3.0 for HIV, and 4.1-fold for severe mental disorder [30], while a longitudinal study by Leibowitz et al. (2021) on Massachusetts state prisons operating amid the 2020 COVID-19 pandemic demonstrated an association between prison crowding and elevated COVID-19 incidence rates [31]. Here, we draw on Leibowitz et al. (2021) and also research by Marquez et al. (2021), who calculated the standardized COVID-19 mortality rate for the U.S. prison population to be 2.5 times greater than the general population [32], particularly as they signal the utility of the criminal legal system in epidemiological research. For example, when national health shocks (like pandemics) affect the U.S., we can use carceral institutions to better study disease etiology and transmission, as well as identify public health risk factors attributable to modifiable confinement conditions—which can motivate reform.

But as we would like to rectify, the criminal legal system is much greater than the sum of prison and jail incarceration. Often, this is not an error of oversight by researchers but rather a result of data scarcity and reliance on secondary data analysis. Popular sources of data equipped to link criminal legal contact to health, such as the Fragile Families and Child Wellbeing Study (FFCWS), the National Longitudinal Study of Adolescent Health (Add Health), the National Longitudinal Survey of Youth 1979 (NLSY79), offer slim pickings when it comes to both treatment and outcome variables, meaning researchers often have no choice but to operationalize incarceration and health in limited ways.⁷

However, restricting the pool of valid data sources to study health because criminal legal contact is conceived as only prison or jail incarceration is a conceptual error that researchers must avoid. When it is apparent felony convictions and imprisonment, specifically, are no longer the most common outcome of criminal legal encounters [33], amid ongoing decarceration efforts, we suspect the environment has changed into the tough-on-crime surveillance and punishment of minor deviance⁸ (e.g., misdemeanors and other low-level offenses) and of the formerly incarcerated—particularly in minority communities [35]. This is where studying probation, parole, and other community supervision routines become instrumental as more people are processed by these aspects of the criminal legal system.⁹ New research should describe health outcomes for populations under probation-parole just as vigorously for those under incarceration, and draw proper comparisons to reference groups. Wildeman et al. (2019), for example, found adults on probation experienced 2.1 times increased risk of mortality compared with an age-matched group and faced a higher risk of dying compared to those in prison and jail [37], while a separate retrospective cohort study found the overall mortality risk of a Michigan cohort under probation and parole was 7.0 times greater compared to the general state population [38]. To this end, research should decompose why alternative criminal legal contact can be so deadly despite its intended leniency, such as work examining the role of overdose deaths in mediating probation-related mortality [39] or flaws of state social support and monitoring policies that lead vulnerable populations to fail their probation or parole terms and recidivate [40•].

Additionally, modern regimes of “zero-tolerance” or “aggressive” policing [41], mean even more citizens are being exposed to involuntary contact with criminal legal authority [42]—and suffering from a range of health interactions that should prompt research. Evidence, for example, shows that residing in neighborhoods with high rates of pedestrian stops and lethal police violence is associated with higher odds of hypertension and other disease risk factors for all residents [43, 44•], with indirect exposures also affecting the likelihood of preterm birth for pregnant women [45]. Even the anticipatory stress of police violence, the worry that one might become a victim, is associated with depression and anxiety [46], and is indicative of how far the criminal legal system “gets under the skin”¹⁰ [47, 48], specifically of Black individuals for which racially-targeted criminal legal contact is now a normative life course event [49, 50•]. Recent work, for example, demonstrates how indirect events, such as news on police killings of unarmed Black Americans, still increase emergency room visits for depression [51] and result in poorer mental health days among Black adults [52•].

Other systems of law enforcement, such as Immigration and Customs Enforcement (ICE) also create spillover health effects. Undocumented Latine immigrant families experience heightened anxiety and adverse mental health outcomes from fears of deportation [53], but it may also be that immigration enforcement policies now affect the health of legally-immigrated and USA-born Latines too. Eastus et al. (2020) found Latine adults living in areas with the highest quartile of ICE-issued detainer requests (to detain individuals suspect of being undocumented immigrants) have greater odds of poor self-rated health relative to those in the lowest quartile [54], while another study, using a quasi-experimental design, found infants of USA-born and immigrant Latina mothers faced increased risk of low birthweight following one of largest ICE immigration raids in history [55].

In all, community supervision and policing are areas where research remains nascent, but given their outsized effects on population health, treating them as legitimate forms of carceral control and determinants of health is imperative. However, shifting research towards more community-based settings (as opposed to just prisons or jails), means the roles of other social institutions become more pronounced. Probation and parole, for example, are often described as revolving doors as people transition between criminal legal, welfare, housing, and employment systems [56]. In particular, the healthcare system inherits a prominent role in this network when individuals released on parole, for example, must transition their health needs from correctional facilities to the community healthcare system, creating discontinuities in care [57], or when increased police presence in hospitals deepens medical mistrust and lowers healthcare utilization [58]. Resultantly, thinking more about how populations with criminal legal contact interact with a plurality of social institutions can help isolate relationships between key independent variables and health outcomes.

Addressing the Long Arm of the Criminal Legal System

As we describe widening effects (i.e., the latitude) of the criminal legal system, it is also necessary to address the longitudinal effects of criminal legal contact, meaning across time, age, and the life course. We know from bedrock research analyzing the NLSY79 dataset that the fact of imprisonment itself is more correlated with negative changes in health status than the length of incarceration [59], but new research should test this finding for jail, other forms of detention, probation and parole, and policing exposures to see if the timing, frequency, and duration of these contacts matter for health. To accomplish this, research should quantify within-person¹¹ or within-cohort changes in health following these periods of broader criminal legal contact, just as studies have done for the ends of imprisonment (see e.g., Wallace and Wang (2020) [60]), but in the long as well as the short term. As Neil and Sampson (2021) have demonstrated by analyzing arrest record data over the life course [61], capturing also between-birth cohort effects can reveal how rates of criminal legal contact differ across the age distribution, and how macrosocial change, as in local, state, or national health and social policies, might drive unequal rates of contact. Such analyses can uncover if probation, parole, and policing are new stages in the life course of at-risk groups, as incarceration has previously been argued for older generations of Black men [62].

Studying the criminal legal system through a longitudinal perspective would further account for health disparities in aging and the shifting age structure of incarcerated populations. For example, 20 percent of Black men born in the late 1960s had served prison time by their early 30s, compared to only 3 percent of White men born in the same years [62].¹² Consequently, as large cohorts of Black (and Latino) men who have endured high rates of incarceration now enter their 50s and 60s, it becomes increasingly important to understand how criminal legal contact impacts health later in life,¹³ and moreover, how early life racial health disparities increase, decrease, or neutralize as populations age. Studying aging populations with criminal legal contact then also matters for those with contact now. Older adults (aged 55 and up) accounted for 8 percent of all adult arrests in 2021, up 5 percent from 2000, while it is estimated from 1991 to 2021, the national prison population aged 55 and older increased from 3 percent to 15 percent [63]. The aging demographics of these populations is a public health issue, as exposure to carceral environments is reported to accelerate biological aging.¹⁴ In an age-adjusted analysis, for example, jail-incarcerated adults were found to have the same rate of geriatric morbidity, at age 59, as community-dwelling adults at age 75 [65]. All told, researchers and practitioners should anticipate these earlier presentations and higher rates [66] of geriatric conditions, such as declines in functional mobility, cognitive impairments (e.g., Alzheimer’s disease and related dementias), or multiple chronic conditions—not only among older populations, but also as predicted health outcomes for youth with early and severe criminal legal contact.

Filling the Scarcity of Causal Modeling Approaches

Although, to this point, we have seen increasing investment into research examining the links between the criminal legal system and health, most studies are descriptive, and few in this growing body of work have explicitly focused on establishing causality. Some aspect of this is due to a lack of need—e.g., descriptive evidence of racially disproportionate police shootings is sufficient to understand that the criminal legal system plays a role in generating mortality disparities [42]—but the other major driving force is simply a lack of data. As noted above, even basic information describing the many arms and operations of the criminal legal system is lacking, and drawing causal insights from such thin data is often a considerable challenge. Take the case of probation and health outcomes, for instance. To parse the discrete causal effects of this experience on health overtime, one might need a dataset containing information on an individual’s exposure history, their health, and a large vector of confounding forces that might alternatively explain the observed association (e.g., household income prior to exposure). To our knowledge, such data is unavailable—obstructing the clean identification of causal effects.

Population datasets offer some leverage, providing measurements of individuals’ health and social characteristics pre- and post-criminal legal exposure that can be used to condition models and better isolate effects from other explanatory factors [67, 68]. Still, these data are often misaligned in their goals. Health datasets often contain a limited set of criminal legal exposures, likely measured with some error (e.g., only incarceration history; or a self-reported number of times a person has been stopped by police), while data with richer descriptions of the various forms of criminal legal contact that one has faced often lack robust and objective health measurement (e.g., self-reported health). Some researchers have taken advantage of so-called natural experiments to make progress in other areas—such as Harding et al. (2019), who examined a random assignment of judges to gain traction on the effects of incarceration [69]—but, given the nature of the exposure, which is deeply tied to other social conditions that generate health vulnerability and are ethically dubious to purposely randomize, these sorts of empirical designs are few and far between.

In some cases, establishing causality is key to supporting effective policy and practice. For instance, as discussed in more detail below, a precise understanding of the mechanisms linking current manifestations of the criminal legal system to poor health would be important for any effort seeking reform—i.e., interventions seeking to shed light on the aspects of incarceration that are most responsible for generating massive and lifelong vulnerabilities to premature death. Still, questions about mechanisms are fundamentally causal in nature—requiring first the identification of a treatment effect to decompose into mediating pathways. As such, the lack of high quality data allowing for identification of causal effects is a prevailing issue that must be addressed.

Identifying Mechanisms

Finally, part of the methodological strategies we have advocated thus far is the identification of mechanisms that create health disparities in the first place. Better understanding the explanatory mechanisms mediating the association between criminal legal contact and worsening health are critical for informing interventions and policies aimed at reducing the negative health consequences of criminal legal contact [70]. But mechanisms can be illusive. With a few exceptions (see e.g., recent work by Widdowson and Fisher (2020) [71]), the majority of empirical work linking any form of criminal legal contact to health do not model mechanisms. As discussed in our previous section, this is largely due to the scarcity of data necessary for modeling mechanisms, as surveys and other available secondary data used to examine specific hypotheses were not originally designed to do so [72]. Consequently, this is where research would benefit from new projects, ideally incorporating original data collection tailored for and research designs capable of modeling (direct, indirect, or conditional) mechanisms through which incarceration affects individual, family, and population health—and moreover, tests of how various mechanisms interact and which have more explanatory power than others, under what settings [4, 73].

Drawing from social science research, including a rich body of qualitative and ethnographic work, which has documented the social and economic costs of criminal legal contact ([74–76]), several interrelated mechanisms have nevertheless been theorized. First, incarceration environments themselves, such as exposures to resource and sensory deprivations (e.g., loss of freedom), hostile norms (e.g., abiding by prison codes or neglectful medical care), and contagion effects (e.g., living in overcrowded or poorly ventilated quarters) may explain declines in health [73].¹⁵ Social integration is another pathway, as individuals facing incarceration, as well as arrests and community supervision, experience disruptions to support systems seen as protective of health, such as family, schooling or jobs ([78]). (Re)integration barriers are also significant. Known as “invisible punishments”, those with criminal records face socioeconomic discrimination, such as labor market, insurance, housing, and voting disenfranchisement [74], as well as struggles with stigma and re-establishing family ties that all may increase morbidity and mortality ([59]). Lastly, more generally, criminal legal contact may affect health by acting as both an acute and chronic stressor ([78–80]). Adjusting to sudden detainment in ICE facilities, unexpected escalation by police, or restrictive drug testing patterns on probation, for a few examples, can be immediately traumatic [81–83]. Stress from criminal legal contact can also become chronic, as aging incarcerated populations, for instance, spend longer and longer periods enduring harsh living conditions or separation from loved ones [84]. However, assuming the same mechanisms affect all populations involved in the criminal legal system is inconclusive. More work is still needed to identify unique and shared mechanisms across all forms of contact and well as variation in mechanisms by key demographic characteristics and levels of aggregation (e.g., at the family and community level where research is sparse [85]). And most importantly, future research must investigate the macrosociological mechanisms¹⁶ that overexpose some sub-populations (such as low-income, urban Black men and women) to the criminal legal system over others, and result in the health disparities we still study today [50•].

Conclusion

Criminal legal contact is now a fixture in many conventional social determinants of health frameworks, when decades ago its role appeared negligible [86]. Though we have made tangible progress in our understanding of the role of the criminal legal system in health, how do we usher it on? We believe the next era of research should continue to progress through work employing broader forms of criminal legal contact, longitudinal and causal methodological approaches, and models of mechanisms. New measures, incorporating other institutions connected to the criminal legal system and understudied health outcomes like mental health, are also needed. However, research endeavors that implement these ideas will be time-intensive. They will require the imputing of empirical and theoretical approaches from disciplines outside of public health. But, if done, they will lead to profound advancements in our understanding of the role of the criminal legal system in population health and health disparities.