Seminar on Criminal Justice Involvement and Later Life Health Disparities

Inside the Carceral Setting – Health Care and Aging Populations

Brie Williams, University of California, San Francisco

Decades of mass incarceration have resulted in an aging prison population and an increase in incarcerated people experiencing serious illness. Mass incarceration refers to the uniquely expansive and punitive carceral system in the United States; approximately 1.9 million people are incarcerated in carceral facilities (e.g., prisons, jails, juvenile justice facilities, and immigration detention) and 10.6 million cycle through jails annually. Mass incarceration, in part driven by extremely long sentences, has fueled growth in the number of incarcerated older adults. People who are incarcerated experience accelerated aging, including an earlier onset and higher burden of chronic medical and geriatric conditions Mass incarceration is also rooted in racism and racialized violence. The lifetime risk of incarceration is one in three among Black men and one in six among Latino men, and people of racial and ethnic minority background also receive the longest sentences.

Pathways for Carceral Health Inequities

People aged 55 and older in carceral settings are aging and dying at a greater rate than their peers in the broader U.S. population. Since 1990, this segment of the total U.S. population has doubled while this same group has grown about 8.5 times in U.S. prisons and jails. Additionally, people aged 55 and older account for approximately 12 percent of the state prison population and 62 percent of all state prison deaths.

Accelerated aging is one major pathway to health inequities among older incarcerated people (age 55 and older, as defined by the Bureau of Justice Statistics). Accelerated aging exacerbates risk factors for negative health outcomes, such as substance use disorder and low socioeconomic status, such that the average physiologic age among incarcerated adults is 10 to 15 years older than their chronologic age. As a result, the prevalence of common geriatric conditions in incarcerated adults is similar to the prevalence of these conditions in much older adults outside the carceral setting.

The 8th Amendment to the U.S. Constitution protects incarcerated people’s right to receive access to medical evaluation and the enactment of ordered medical care, which theoretically includes geriatric care for older adults and palliative care for seriously ill individuals. However, too few geriatric care specialists work in carceral settings to provide this care consistently. Incarcerated people can also receive substandard care when transferred to community hospitals. For example, incarcerated patients who are pregnant, comatose, or dying are often required to be shackled, which may lead to shackling ulcers. Standards around incarceration-specific care practices, such as shackling, should be clarified or redressed, especially as the number of incarcerated patients in community hospitals and the age of incarcerated patients dying in community hospitals rise.

Exacerbation of Common Geriatric Problems

Rates of hypertension, heart-related problems, diabetes, asthma, stroke, dementia, mild cognitive impairment, and other chronic conditions are higher among incarcerated persons compared to those in the general U.S. population. Likewise, the prevalence of COVID-19 in older adults is higher in the carceral setting compared to those in the broader community. At the height of the pandemic, older incarcerated adults who developed COVID-19 had worse outcomes compared to younger incarcerated adults, including intensive care unit stays and death.

After release to the community, a history of incarceration is associated with a greater-than-average risk for common geriatric conditions, including cognitive impairment, even when controlling for socioeconomic factors.

Inside carceral settings, poor hearing and urinary incontinence are common problems older incarcerated individuals face. These problems are an understudied but emerging area of concern, especially due to their long-term consequences. For example, poor hearing may lead to an inability to hear staff orders and rule violations, which could lead to longer sentences.

The abilities of older adults to perform activities of daily living (ADL) and instrumental activities of daily living (IADL) are critical geriatric measures because any loss of ADL or IADL could be a harbinger of major medical and psychosocial problems including mortality. However, while ADLs are relevant in the carceral setting, many IADLs are largely irrelevant, including using transportation and cooking. Instead, daily tasks that are necessary for independence in the prison or jail context differ from those in the outside world. For example, daily activities become far more difficult in shackles. Therefore, functional impairment in the incarcerated population should be measured by a different set of ADLs that is more reflective of their daily activities. Activities of daily life in prison, or ADL-P, could measure the ability to:

• Drop and rise from the floor for alarms that are common in carceral facilities
• Get to the dining hall on time for meals before the meal is over
• Hear orders from staff
• Climb on and off an assigned bunk.

To account for differences across carceral settings, a unique ADL-P should be created for each individual housing unit to identify activities needed for independence therein and to determine housing and special assistance needs.

Solitary Confinement

Solitary confinement is generally defined as being housed alone for 22 to 23 hours a day, seven days a week, in a cell roughly the size of a parking spot, often without a window. People who are confined to these spaces are disproportionately racial and ethnic minorities and people who identify as lesbian, gay, bisexual, transgender, and queer (LGBTQ+). On any given day, 45,000 to 122,000 people are estimated to live in solitary confinement in the United States.

Solitary confinement causes profound harm to mental and physical health, including increased rates of self-injury, post-release suicide, and overall mortality. People incarcerated in solitary confinement units report sensory deprivation, deep loneliness, and disturbances to visual depth perception from the inability to look more than 10 feet away for months or years. Exercise is a common treatment and preventive measure for many common geriatric conditions, but solitary confinement prevents older incarcerated individuals from accessing this community standard of wellness care. Staff in these units also experience negative health outcomes, including physical and mental trauma, moral injury, and burnout. Researchers and policymakers should work together to determine how solitary confinement affects common conditions among older adults and advocate for the development of evidence-based approaches to eliminating (or at least greatly reducing) the use of this dangerous practice.

Policy Considerations

A significant barrier to research has been the poor roll out of the Death in Custody Reporting Act that requires state prisons to report on the number and causes of death during custody. Beginning in 2020, the Department of Justice switched the implementing agency from a statistical unit (Bureau of Justice Statistics) to a grants and technical assistance unit (Bureau of Justice Assistance). That handoff was poorly coordinated and, as a result, the Department of Justice has not published any analysis or datasets on deaths occurring in carceral facilities since 2019.

Cognitive impairment is a risk factor for reincarceration. Training police, correctional officers, legal professionals (including judges) and health care professionals to consider whether cognitive impairment is present and how to better interact with people who are cognitively impaired is an important area for investigation, as is the capacity for interventions to reduce the likelihood of incarceration.

With the aging of the carceral population, health care professionals practicing in carceral settings and in community hospitals and specialty clinics that serve incarcerate patients should understand the rights of incarcerated patients, including their rights to choose medical interventions, appoint a decision-making proxy, or say goodbye to loved ones at the end of life. Patients and health care professionals alike report a lack of knowledge around compassionate release, an umbrella term for a range of policies in state and federal prisons that allow incarcerated individuals who are too ill or cognitively impaired to understand their imprisonment or participate in rehabilitation to die outside of prison. Lack of knowledge around compassionate release, including how to petition and how to prognosticate, contributes to its infrequent use.

Funding is also essential to carceral health care. Carceral centers face a major cost crisis, due in large part to the aging of incarcerated populations. The cost to incarcerate older adults is about four to nine times higher than the cost to incarcerate younger adults; much of this cost differential is due to health care, which comprises about $12.5 billion of the total $82 billion in annual combined state correctional spending. A more nuanced, evidence-based understanding of the health and functional ability of older incarcerated adults would provide insight into this cost crisis by allowing carceral centers and researchers to predict long-term health care needs, hospitalization costs, and predicted mortality rates.

Outside the Carceral Setting – Criminal Justice Involvement and Later Life Health: Mechanisms and Consequences

Alexander Testa, University of Texas Health Science Center at Houston

The era of mass incarceration in the United States began with policies established in the 1970s, so many previously incarcerated people are now back in the community. Dr. Testa reviewed the 10 sources of data that allow researchers to study health outcomes after incarceration for individuals 45 and older:

• Health and Retirement Study (HRS)
• National Longitudinal Survey of Youth 1979 (NLSY-79)
• Community-specific samples
• Medicare administrative claims data
• Veterans Aging Cohort Study
• Add Health Parent Study
• Transition Clinic Network Data
• National Survey on Drug Use and Health (NSDUH)
• Criminal Justice Administrative Records System (CJARS)
• South Carolina Alzheimer’s Disease Registry

HRS and NLSY-79 account for 21 of the 39 publications on criminal justice encounters among older adults since 2013.

Data Collection

Many of the data sources researchers use to study health outcomes in aging populations of formerly incarcerated individuals include few questions about incarceration itself. HRS had one question on childhood police encounters in 2008 and 2012 and two questions on incarceration in 2012 and 2014. Similarly, only one question from NLSY-79 addressed incarceration by listing “jail” as an option for a survey respondent’s type of residence.

Because major datasets have minimal information on incarceration, many unanswered questions remain about the differences in health outcomes related to various types of criminal justice contact and the differential effects of incarceration at various times in the life course. For example, one study using Veterans Affairs data found that premature deaths among formerly incarcerated people may lead to survivorship bias in the data because estimates of the median age of death for incarcerated or previously incarcerated individuals range from ages 47 to 49, and studies of aging populations often begin at age 50.

Similarly, more data are needed to draw conclusions about age-related health issues as a pathway to incarceration; for example, 3.5 percent of individuals who have contact with the criminal justice system have dementia, but more research must be done to establish a causal relationship.

Findings

Survey data has linked prior incarceration to many negative health outcomes for individuals aged 45 and older. Cognitively, prior incarceration correlates to higher rates of depression, disability, and AD/ADRD as well as to lower verbal recall and worse cognitive ability. Behaviorally, prior incarceration correlates to higher rates of hospital and emergency department use, smoking, drug use, overdose, and suicide attempts. Physically, prior incarceration correlates to worse self-reported health and higher rates of inflammation biomarkers, cardiovascular disease, geriatric and chronic health conditions, and mortality.

The wide variety of negative health outcomes associated with prior incarceration in older adults demonstrates that prior incarceration is a major social determinant of health (SDOH). The strong impacts after release suggest that reintegration struggles are as important as the conditions of confinement themselves.

Evelyn Patterson, Georgetown University

Incarceration of an individual occurs in the context of larger social factors, such as police surveillance and local social support. Incarceration, therefore, does not act as a single explanatory variable to determine health outcomes for individuals, but plays one role in inseparable systems.

Heterogeneity and Social Processes

Survey data often lacks variation in the duration and impact of incarceration and other forms of criminal involvement. Even interactions, such as police car stops that are usually minor, may lead to disruption of roles (e.g., preventing timely pickup of a child) or even death. People may be impacted simultaneously by multiple aspects of the criminal justice system. For example, an individual may live in a highly surveilled area and experience the incarceration of a family member. Forms of criminal justice involvement, including incarceration, may be repeated, making direct causal relationships more difficult to establish.

Direct and Indirect Pathways

Much of the research on health disparities after incarceration uses a model in which incarceration is the cause of a series of health outcomes, such as morbidity, mortality, and worsened mental health and ADL measures. The relationship between incarceration and health outcomes is, in fact, more complicated. Many types of criminal justice involvement may contribute to negative health outcomes, and negative health outcomes may also make criminal justice involvement more likely.

One way to understand the relationship between incarceration and health outcomes is to split causes and effects into direct and indirect pathways. A direct pathway to negative health outcomes might be the chronic stress experienced by an incarcerated individual, and an indirect pathway to negative health outcomes might be difficulty accessing housing.

Structural Inequality

Pathways to negative health outcomes for incarcerated individuals are also linked to social and structural factors. Inequalities in SDOH, such as minoritization, may contribute to incarceration and to negative health outcomes, which may thereby explain some of the disparities in health outcomes for individuals who are or have been incarcerated. Beyond single characteristics, cumulative social disadvantage manifests as early physical deterioration or accelerated biological aging, which also leads to negative health outcomes.

Methodological Challenges, Data Needs, and Emerging Issues

Christopher Wildeman, Duke University

Many large-scale health studies and datasets (e.g., HRS) were not designed to capture interactions with the criminal justice system. Even when these datasets include some data about criminal justice involvement, they often lack key details about duration and type of incarceration and may only cover subsets of affected populations, such as those who are imprisoned rather than jailed. In the absence of a longitudinal study that is uniquely suited to testing the effects of criminal justice contact on health, and since such a study would require a significant influx of funding, researchers can seek to improve existing surveys and their methods of use.

Survey Enhancements

Existing surveys may not address negative health behaviors, criminal activity, or other stress-driven behaviors to avoid stigmatizing respondents. However, these behaviors may be important indicators for health outcomes, and surveys should include questions on these topics.

Survey enhancements can also be used to differentiate between carceral settings (e.g., by asking about the number and duration of incarcerations or by asking about the health care individuals received in the carceral system). The large sample size of NLSY will allow data from this survey to be disaggregated by carceral setting. Smaller surveys could include more granular questions about types and repetition of criminal justice contact.

Methodological Concerns

Incarceration and other forms of criminal justice contact affect health outcomes in an environment of confounding variables, such as social marginalization, negative health behaviors, crime, and delinquency, some of which precede contact with the criminal justice system. The interplay between these variables and multiple interactions with the criminal justice system make identification of contact itself as a causal variable impossible. Rather, criminal justice contact can be shown to be one risk factor for worsened health outcomes among many. Because criminal justice involvement is a risk factor but cannot be shown to be a cause of negative health outcomes, common econometric and modeling techniques that examine the effects of decision-makers (e.g., judges) on outcomes for individuals can be difficult to conduct.

Bias Reduction

Surveys that collect data from medical providers can underrepresent individuals who have had contact with the criminal justice system due to patients’ mistrust of medical care; existing surveys have studied this situation and can predict health outcomes for these individuals. Some existing surveys capture self-reported income rather than taxable income, which can be important for criminal justice involved populations, who are often barred from many types of official employment.

Self-reported data about short-term jail stays and prior incarceration may be subject to differential, recall, and self-report bias as well as survey attrition and uncertainty in instrument sensitivity. Reconciling administrative data with self-reported data on short-term jails stays and prior incarceration could help determine the levels of bias in survey data and thereby give more accurate insights into levels of incarceration and the impact of short-term jail stays. The level of bias in different populations should also be disaggregated, by race, ethnicity, number of arrests, and type of incarceration.

Mike Mueller-Smith, University of Michigan

Two major factors contribute to the growing population of older adults with criminal records: (1) mass incarceration in the United States rose steadily since 1974 and peaked two decades later, and (2) most people commit criminal activity when they are young—the median age of first contact with the criminal justice system is 18. Thus, a generation of people who turned 18 during the peak period of mass incarceration will reach retirement age within the next few decades. For example, people born in 1982 hit the prime crime age of 18 in 2000 and will reach retirement age in two decades.

This combination of factors has led to a large and growing population of individuals with criminal records who are older than retirement age, which will likely result in a crisis of elderly individuals excluded from the essential features of the current social safety net. People with a criminal record face labor market discrimination and difficulty finding and keeping legal work, so they are less likely than the general population to have access to Social Security insurance, pensions, or private savings. Thus, new safety net approaches may be necessary to support these individuals as they age.

Furthermore, economic data on individuals with histories of criminal justice contact can be difficult to assemble. CJARS began in 2016 as a novel data platform to modernize research and successful reporting on the U.S. criminal justice system to account for economic disparities in the population of criminal justice involved populations throughout their life courses.

Data Access

Accurate information about the portion of people with criminal records is essential to understand and plan for the aging of criminal justice involved individuals, particularly because they are more likely to have worse health outcomes and less access to care and savings resources. Administrative records of individuals with early life contact with the criminal justice system could bolster the accuracy of survey data by protecting against social desirability and recall biases as well as the high rate of survey attrition in this population.

CJARS was built to incorporate individual-level survey data with event-level criminal justice administrative data on a nationwide level. However, CJARS data remain incomplete due to some jurisdictions reporting no data and others reporting only prison or court data without other types of criminal justice contact. Decentralizing data creates barriers to harmonizing and integrating data, which, in turn, limits researchers’ ability to measure the impact of criminal justice policies or recommend investments in social safety net programs or education.

New sources of information should also be leveraged to understand the well-being of populations with histories of criminal justice involvement. Currently, CJARS does not have data about measures of well-being later in life. Individuals age into BJS’s definition of older adults an average of 40 years after first contact with criminal justice systems. CJARS’ administrators also seek new information to clarify changes in incarceration and its effects across birth cohorts, racial and ethnic disparities, spillover effects of criminal justice involvement onto other household members, and differences across local jurisdictions and geographies. Although information from the National Corrections Reporting Program disaggregates data by year and procedural stage, additional information is needed to track individuals as they move through, and sometimes repeat, phases of the criminal justice contact.

Modeling Demographics of Criminal Justice Involved Populations

Whether or not an individual will be involved with criminal justice systems is likely to be determined by age 20, long before retirement age. Integrating CJARS, Census, and Social Security data allows researchers to use what is known about current levels of young adults with criminal records to model future demographics of criminal justice involved populations and to estimate the share of the retirement age population who will have survived criminal justice contact.

The portion of adults with criminal histories who will reach retirement age is expected to reach 24 percent in 2040, up from 13 percent in 2018. By 2040, about a third of men surviving to age 62 will have some form of criminal record and 7 percent will have gone to prison. It is estimated that close to half of Black men will have a criminal record and over a fifth will have been to prison. About a quarter of White non-Hispanic people will have a criminal record of some kind. Fewer women than men will have a criminal record in 2040, but the current rate at which women are involved with criminal justice is growing at a faster pace than men.

Impacts of Criminal Justice Involvement

The 2006 to 2019 results of the American Community Survey (ACS) gave insight into life after 50 by comparing people with and without a history of criminal justice involvement. On average, involvement in criminal justice systems ended at age 42, but the direct and indirect effects of involvement continued throughout the lifespan. After weighting data by race, education, age, and demographic profiles, researchers found that criminal justice involvement impacted a range of lifecycle outcomes that influenced well-being in old age, including employment and earnings, health, and family structure and household attachment. These direct effects also impacted indirect effects, such as differences in health insurance coverage, eligibility for Medicaid or private insurance, rates of disabilities, and use of facilities (e.g., nursing homes, drug treatment facilities, and homeless shelters).

Caitlin Patler, University of California, Berkeley

ICE is responsible for enforcement of federal immigration laws in the interior of the United States. ICE detention is a civil procedure that runs parallel to the criminal legal system. Because ICE detention is considered civil rather than criminal detention, some constitutional protections (e.g., right to a public defender, systematic access to bond, and limitations on length of detention) do not apply to immigrants detained by ICE. This detention is also not considered a sentence and is, thus, not considered legally punitive.

Immigration law (and therefore, detention) is federal law, but states and local jurisdictions have significant leeway to determine whether to alert ICE about the presence of noncitizens. Any noncitizen can be detained and deported, including lawful permanent residents who have contact with the criminal justice system, which can trigger deportation proceedings.

Between 2008 and 2018, ICE made more than 2 million apprehensions; the average daily detained population is currently around 40,000 individuals, from a historic high of approximately 55,000 in 2019. Historically, detained immigrants are overwhelmingly male and mostly from Mexico and Central America. Like the criminal legal system, ICE detention facilitates the exclusion and expulsion of racialized minority populations from full participation in daily life under the guise of race-neutral policies. Therefore, immigration policies are racialized structural determinants of health.

Conditions of Immigration Detention in the United States

ICE subcontracts detention to privately operated facilities and local jails. In practice, ICE detention centers have similar conditions to other carceral settings, such as restrictive and controlled movement, punitive experiences (e.g., solitary confinement and shackling are common), and limited communication with the outside world, including with legal representation.

ICE detention and incarceration also have some key distinctions, which may have implications for health and healthcare. Private, for-profit corporations currently account for around 90 percent of ICE detention beds. In addition, ICE detention is mandatory for many people seeking asylum and for most people with a criminal record, including those convicted of nonviolent misdemeanor offenses, and there are few constitutional limits on the length of detention. Together, these policies can create indefinite imprisonment without trial. The uncertainty of immigration detention may influence health.

Health Harms of the Immigration Detention System

Medical care in ICE detention centers is substandard and has little oversight. ICE regularly fails to meet its own standards for care; violations of national detention standards were found in 78 percent of deaths in ICE custody from 2011 to 2018. The health care provided to detainees often focuses on acute care rather than chronic or preventative care.

Confinement in ICE detention centers has been linked to mental health deterioration among people who have and have not been previously diagnosed with a mental illness. Detained immigrants may experience stress and trauma before detention (e.g., reasons they left their country, the migration journey, and settlement) and during detention (e.g., family separation, indefinite uncertainty, and regular use of solitary confinement, even for people with serious mental illness). Outcomes of this stress can include an increase in suicidal ideation, self-harm, and suicide; exacerbation of past trauma and stressors as detention length increases; and high burden of hospital admission for psychiatric illnesses.

Confinement in ICE detention centers has also been linked to poor physical health, through such mechanisms as neglect of basic needs, poor sanitation, overcrowding, lack of recreation, and increased risk of physical and sexual violence, especially for sexual and gender minorities. The punitive conditions of confinement are individually associated with worse health and, together, they accumulate to further worsen health. Since the immigrant population tends to be underserved by the medical community even outside of detention centers, there is a strong possibility that conditions are underdiagnosed in the immigrant population.

Health Research on Detained Individuals

ICE does not regularly release detailed information about the populations it detains, including demographic data, making health research on this population difficult. Relatively few studies have examined the health of immigration detainees due to the difficulty of accessing data. Administrative data may be accessed through FOIA requests, but this information is often heavily redacted. Tracking individuals through the immigration detention system is not currently possible. Many researchers use observational convenience samples or referral samples of former detainees. Strengthening data access while protecting the privacy and confidentiality of detained immigrants would greatly aid research.

Criminal justice involvement often precedes ICE detention, and many other factors, such as lower access to insurance and health care, may also influence health. More research is needed to understand the potentially overlapping health impacts of the criminal and immigration legal systems for individuals who have been incarcerated in both systems.

Finally, little is known about the impact of ICE detention on aging populations. A growing body of research on aging in immigrant communities points to the central role that immigration laws—as structural determinants of health—play in the process. For example, the undocumented Latino immigrant population is aging, with 40 percent predicted to be older than age 55 by 2038. Data collected during ICE detention shows that detainees have high rates of chronic conditions, such as hypertension and cardiovascular disease. More research is needed to understand age-related health disparities among detained immigrants. Additional data and research on immigrants in the immigration and criminal legal systems will be crucial moving forward.