There have been many assertions about how the effects of criminal justice involvement may reach beyond the imprisoned individual and affect his or her family, extended family, and the community at large in many different ways. People who are incarcerated endure physical and emotional struggles during the period of imprisonment, and their reentry into the outside world is also full of challenges. It has been argued that the spouses, partners, and children of incarcerated people are likely to face a unique set of financial and emotional challenges that can adversely affect familial relationships. Community dynamics may also play a role: the quality of the environment and the degree of resource accessibility ultimately determine the level of support available to families of incarcerated people and to the people themselves when they reenter society. Although there is some evidence to support these assertions, the research is limited. The workshop session reported on in this chapter focused on identifying key points about the relationship between criminal justice involvement and public health that need to be better understood in order to begin proposing viable solutions.
John Hagan (Northwestern University) and Holly Foster (Texas A&M University) presented findings from their work on the National Longitudinal Study of Adolescent to Adult Health (Add Health),1 and on other
1 Add Health is a longitudinal study of a nationally representative sample of adolescents in the United States who were in grades 7-12 during the 1994-1995 school year. Subjects underwent four waves of in-home interviews; the most recent was in 2008. Data have been collected from adolescents, as well as from their fellow students in the first wave, school administrators in the first wave, parents in the first wave, and siblings, friends, and romantic partners through multiple data collection components.
existing literature about the impacts of paternal and maternal imprisonment on children and families. Hagan began the workshop discussion by noting the significant escalation of the U.S. federal and state prison population over the last century, growing steadily from about 100,000 in 1925 to just over 300,000 in 1980, and then rising rapidly to 1.6 million over the next 30 years. In recent years, however, the incarceration rate has begun to decline. Several states with histories of high incarceration rates, including California, New York, and New Jersey, have seen marked decline (22% or more in 2009-2014), but other states, including Nebraska and Arkansas, are still noting increases.2 Foster noted that recent studies show that men made up 93 percent of the prison population in 2015. Research shows the influence of male incarceration on people’s lack of employment prospects, risk of divorce, diminished political participation, housing insecurity, and homelessness. From a health standpoint, men’s incarceration has been linked to depressive symptoms and psychiatric disorders. While women only constituted 7 percent of the incarcerated population, they are a growing minority and an important one because almost two-thirds of them are mothers of minor children. While there is still much to be learned about the female incarceration population, recent data from the Fragile Families Study (see Chapter 7) on maternal incarceration did link the phenomenon to poorer self-rated health.
Foster noted that because one in four women has a family member in prison (Lee et al., 2015), researchers have begun investigating the effects of male incarceration on inmates’ female partners. In Doing Time Together, Comfort (2008) coined the term “secondary prisonization” to describe the experiences of inmates’ female spouses and partners. Turney and colleagues (2012) also investigated how paternal incarceration is linked to diminished mental health and reduced in-kind support for mothers of inmates’ children. They found positive association between incarceration and maternal neglect as well as harsh parenting.
Incarceration of either or both parents can have debilitating consequences for children. Foster pointed to research showing that more than one-half of adult prisoners are parents of minor children (Mumola, 2000; Glaze and Maruschak, 2008). Table 3-1 shows a summary of recent research reviewed by Foster and Hagan on paternal and maternal incarceration effects on children, as indicated by children’s internalizing and externalizing problems and their social exclusion and marginalization. Foster noted
2 The data are from the publication series on prisoners from the Bureau of Justice Statistics.
TABLE 3-1 Examples of Findings on Maternal and Paternal Influences on Children’s Outcomes
|Outcome||Children’s Internalizing and Externalizing Symptoms||Children’s Social Exclusion and Marginalization|
Education: Hagan and Foster (2012a)
Foster Care: Swann and Sylvester (2006)
Null and Heterogeneous Effects:
Homelessness: Wildeman (2014)
Education: Foster and Hagan (2007)
Homelessness: Wildeman (2014)
Powerlessness: Hagan and Foster (2015)
SOURCE: Foster and Hagan (2016).
evidence in the literature showing that maternal incarceration has adverse or “risk” effects on the manner in which children externalize and internalize symptoms, as well as on their educational success. She said it was also important to note that maternal incarceration was shown to have no distinct effect (null) or a heterogeneous effect (where the incarceration alone could not be identified as the sole factor) in some social and emotional realms for children—namely, behavioral problems and homelessness. A broad range of studies have shown more consistent risk effects of paternal incarceration on children’s internalizing and externalizing behavior problems and in the realm of social exclusion, including educational deficits, homelessness, and powerlessness.
Foster and Hagan discussed findings from Add Health regarding the effects of parental incarceration on economic and food security, receipt of medical services, and college access and completion rates for young adults in waves III and IV of the study (Hagan and Foster, 2015). Parental imprisonment was measured on the basis of survey questions that asked
respondents whether their biological father or mother has ever served time in jail or prison. Economic or income insecurity was measured by a person’s ability to pay for rent or mortgage, telephone, and utility bills; food insecurity was defined by a person’s ability to pay for food.
Young adults who were experiencing or who had experienced either paternal or maternal imprisonment were more likely to report income insecurity relative to those adolescents whose parents had never been to prison. The figures were higher for those with incarcerated mothers. Similar results were found in the case of food insecurity. The findings still held after controlling for such covariates as ethnicity, education, marital status, and parental characteristics. Comparing any of the four forms of income insecurity across two time periods—Wave III (2001-2002) and Wave IV (2007-2008)—the authors found that up to 40 percent of the children of mothers who have been incarcerated have problems with one or more of these income insecurities. The rates were higher in Wave IV than they were in Wave III,3 and the results were consistent across waves in terms of paternal incarceration having a slightly less adverse effect than maternal incarceration.
Another aspect investigated by Hagan and Foster was the linkage of criminal justice system involvement to the receipt of appropriate health care services when needed. There was a negative association between parental incarceration and the receipt of health care services by young adults: fewer young adults received medical services when needed if they had an incarcerated parent, and the effect was stronger for young adults with incarcerated mothers. Multivariate analysis showed that maternal incarceration significantly reduced the odds of receiving medical care when needed, even after controlling for other covariates. However, the analysis also showed that adding health insurance coverage and mitigating economic insecurity reduced the impact of maternal incarceration to non-significance. While both of those factors mediated the effects, the pathway through economic insecurity and marginality was especially pronounced for maternal incarceration. The analysis also showed race disparities in receipt of health services for young adults with incarcerated parents. African Americans were less likely to receive health care when needed than were non-Hispanic whites: 17 percent of the racial disparity was explained by access to health insurance, and 28 percent was explained by economic deprivation.
One of Add Health’s unique features is that it contains information on schools attended by children of survey respondents, of which there were upwards of 90 schools. Hagan and Foster said that this information allowed them to study the effects of being in a school in which a large number of the students have had an incarcerated parent—both on children in
3 As the fourth wave of the survey was conducted in 2007-2008, effects of the Great Recession on young adults were also taken into consideration.
that category and on children who did not have an incarcerated parent. In a representative sample of U.S. schools, the rate of maternal imprisonment ranged from 2 percent to 13 percent (with a mean of 3 percent) and the rate of paternal imprisonment ranged from 3 percent to 26 percent (with a mean of 10 percent). Hagan and Foster said they believe that these rates may influence youths’ expectations of educational attainment. According to their research, even children with incarcerated parents fare better at schools in which the level of parental imprisonment is near zero. The probability of college graduation for these children is 25 percent; in schools with 20 percent paternal imprisonment, the graduation probability for that same group is cut in half. The graduation probability for children whose fathers were never imprisoned drops from 44 to 25 percent when the paternal imprisonment increases from near zero to 20 percent.
Hagan and Foster also reported on a similar analysis to see the differential effects of attending schools with low and high maternal incarceration rates. Students with incarcerated mothers had a 2 percent chance of graduating college if the school they attended had a maternal incarceration rate near zero. When the school’s maternal incarceration rate was close to 10 percent, those students had just a 1 percent chance of graduating college. Students whose mothers were not incarcerated but attended schools with a 10 percent rate of maternal incarceration had a 25 percent likelihood of college graduation.
Hagan and Foster concluded their presentation by emphasizing the societal benefits of government investment in measuring the effects of paternal incarceration on health outcomes. Studies like Add Health allow researchers to investigate health-specific and broader scale outcomes of young adults. Add Health in particular is also able to provide information on racial and ethnic disadvantages, such as how minority youth are more affected by paternal incarceration because they are exposed to the phenomenon more prevalently. The survey has also yielded estimates of effects of paternal and maternal incarceration at community and school levels.
Christopher Uggen (University of Minnesota) discussed the cumulative effects of criminal punishments. Crime exists, and as such so must punishment in an effort to enforce and enhance public safety. The focus in his work, however, has been to identify instances and areas of excess punishment. Uggen calls the cumulative effects of incarceration and criminal punishment—things like employment barriers, welfare restrictions, or perpetual disenfranchisement—“collateral sanctions.” Uggen said that these
sanctions are often studied in isolation, which does not give an accurate picture of the way their effects can “pile up” on an individual.
Uggen described the example of Michael, who had several convictions for driving while intoxicated. Michael served his punishments, got sober, and was admitted to college for social work, which in itself was a challenge because of his record. Once he got his degree, however, he had to petition the board to get a social work license because of his record. He still had difficulty finding a job, but did so, moving his family to Minnesota. But when someone on his county’s commission took issue with his history, he lost his job—and his family lost their health benefits. Michael did subsequently find employment as a social worker, but it came after having to jump through and over several hoops and hurdles.4 Uggen also shared the story of Alex, who arrived in the United States illegally as a child but obtained his green card. At age 16 Alex was convicted of statutory rape. He served his sentence, but he did not then register as a sex offender. Alex spent 7 years without criminal incident, but then the U.S. Customs and Immigration Enforcement agency ruled that his failure to register warranted punitive action, and Alex is now facing deportation. Uggen emphasized again that, while some sanctions are merited and essential for public safety, others are capturing a large number of people who may not pose any threat. When one studies the health effects of incarceration, one should also consider the stigmas faced by ex-prisoners and individuals with felony convictions.
Uggen said there are an estimated 2.5 million people currently in prison (Carson, 2015). That total includes 2 percent of the nation’s adult males, and 5 percent of its African American males. According to state-level estimates from 2010, 5.1 million U.S. adults have been previously incarcerated—including another 10 percent of the African American male population. He also reminded the workshop participants that while the rates of incarceration have seen recent declines, the number of ex-prisoners continues to grow, so the count now in 2016 is undoubtedly much larger.
Uggen outlined basic statistics on racial composition of ex-prisoners and how they are geographically distributed across the country. Southern and southeastern states have a high concentration of ex-prisoners, as of 2010. African American ex-felons form about 18 percent of the voting age population or about 25 percent of the African American male voting-age population. Some states have high racial disproportionality in criminal justice sysem involvement—notably including California, Illinois, and Indiana—as well as high probation rates and long probation sentences.
4 As detailed in a recent report on the growth of incarceration in the United States, a growing proportion of U.S. citizens have been denied a range of rights and access to many public benefits, becoming increasingly marginalized in civic and political life (National Research Council, 2014).
Uggen referred to recent research on the effects of incarceration on individuals’ mental health. Schnittker and colleagues (2012) found strong evidence for a positive correlation between depressed mood and prison experience. The correlative evidence was less strong for anxiety, panic, and social phobia. From a community standpoint, Uggen and some colleagues tested the hypothesis that states with higher incarceration rates would have strong global effects on disease rates. They found that a percentage point increase in male ex-prisoners led to a 7-17 percent increase in death from HIV/AIDS among non-incarcerated men. Among women, the association ranged from 21 to 27 percent. Their research also showed that the rates were highly contingent on the availability of testing and treatment in a particular area: where the testing and treatment were widely available, having a large number of ex-prisoners in circulation was associated with lower rates of disease than in other areas. Furthermore, the researchers found that the effect of incarceration seemed to change over time: this relationship has declined among men since 2003, which may be due to better detection of diseases in prison and treatment of incarcerated populations.
Uggen laid out the theoretical framework for understanding the spillover effects of ex-prisoners on community:
- Ex-prisoners reentering the community could adversely affect health care utilization in an area through an increase in uncompensated care (care given to those lacking insurance), which could lead to reduction in services and diminished quality. The burden of financing care for the uninsured affects the availability of health care to others, therefore leading to worse community health outcomes in an area. Uggen noted that Schnittker and colleagues (2015) found that as the rate of ex-prisoners rises in a state there are clear effects on health care:
- — Accessibility: residents have less access to care and specialists, reduced physician trust, and less satisfaction with care.
- — Capacity: care received by the uninsured people reduces the number of open beds and appointments.
- — Quality: higher quantity diminishes the quality of ambulatory visits, mammograms, and annual tests.
- — Breadth: insured people, women, adults over age 50, and non-Hispanic whites are most affected.
Uggen ended his presentation by listing some recent policy changes that he believes will continue to reduce collateral consequences for incarcerated individuals, their families, and the community as a whole. He noted that more than 20 states have banned employers from asking questions on criminal history on job applications, which might improve the chances of
ex-felons finding employment. Twelve states have opted out of the ban that denies welfare benefits to people with drug felonies. Although only two states allow felons to vote from prison, 19 states restore voting rights after a person’s incarceration, parole, and probation are completed. These are significant policy changes that reduce the number of sanctions on ex-prisoners.
John Laub (University of Maryland) asked Hagan and Foster whether wave V of Add Health will include questions on the Affordable Care Act, which would enable researchers to see the benefits of the expansion of health care coverage to previously uninsured populations. Josiah Rich (Medical School, Brown University) suggested that the researchers would want to disaggregate the results by expansion and non-expansion states. Hagan said that the fifth wave of the survey is being conducted and will continue until 2018, which will provide the opportunity to investigate the effects of the Affordable Care Act. On the basis of an analysis done on wave IV of Add Health, he conjectured that the Affordable Care Act would further mediate the effect of maternal incarceration.
Workshop chair Wendy Manning (Bowling Green University) asked the three presenters to comment on remaining data gaps and benefits that can be derived from expansion of available data from other national data sources. Uggen described the difficulties in teasing out state-level and institution-level variation from the data, which would be better suited to produce aggregate statistics. Foster said she would like to broaden the scope of research to incorporate variation in state policies. On this point, Hagan mentioned a study by Beckett and Western (2001) that looked at the effects of the stringency of regimes on welfare support. He believes that such variations over time and across states can be captured by a nationally representative survey such as Add Health.
Ruth Peterson (Ohio State University, emerita) asked Hagan and Foster if the Add Health Study intended to look at children of the original respondents who are now parents themselves, potentially affecting the behavior of their own children in similar ways. Foster said she hopes that the study continues and is able to interview these adult children, enabling them to look at the intergenerational effects of incarceration.
Alexis Bakos (Office of Minority Health) commented on the phenomenon of maternal imprisonment leading to greater economic and food insecurity. She hypothesized that mothers tend to have larger networks (grandmothers, sisters, aunts) and therefore greater social capital than fathers. Bakos also wondered about the interaction effects of paternal and maternal incarceration—whether there were studies on households in which both parents had histories of incarceration. Foster noted the work
of Lee and colleagues (2013) at the U.S. Centers for Disease Control and Prevention, whose findings were stronger for fathers than mothers. Hagan referred to the work done by Wakefield and Wildeman (2011) on the use of foster care with maternal incarceration. He added that the research is moving toward looking at aftereffects of foster care and how Medicaid is linked to these arrangements, which is also pushing the cause for having longitudinal data. Candace Kruttschnitt (University of Toronto) pointed out that a mother’s addiction habit can be a confounding factor in the relationship between maternal incarceration and food insecurity. Hagan and Foster agreed with Kruttschnitt and said that the next wave of Add Health will allow them to use a fixed-effects analysis to account for this confounding factor.
Daniel Nagin (Carnegie Mellon University) mentioned alternative data sources, such as the Pathways to Desistance Study, which have collected information on health care services received by 1,500 juveniles in Philadelphia, Pennsylvania, and Maricopa County, Arizona (Phoenix). He also highlighted the utility of administrative data by describing his work using court data on convictions and linking it to the Department of Corrections’ records to understand the effects of incarceration on recidivism. Nagin added that administrative data are a valuable source of information for studying dose–response relationships (if incarceration is perceived as a dose) and how length of incarceration is related to different outcomes.
In response, Ingrid Binswanger (Kaiser Permanente Colorado and University of Colorado School of Medicine) said that medical records in state and county criminal justice agencies are stored centrally and mostly are not digitized, and she emphasized the need for electronic health records in the correctional health system. She acknowledged that most of these health records are not harmonized, and some of the records are proprietary, which creates data access challenges. The information in those health records can provide prevalence estimates for certain diseases and quality of care received by inmates. Nagin added that developments in the field of text analysis can increase the efficiency of extracting useful information from administrative health records.
Amanda Geller (New York University) added that for her Fragile Families study the researchers were able to get New York State’s Division of Criminal Justice Services to match their respondent identifiers to records in the system that had been reported from police departments. That method had a lot of potential but also a lot of pitfalls, such as funding and confidentiality issues, which she will address in her presentation (see Chapter 7). Kathryn Porter (National Center for Health Statistics) recommended the National Survey of Children in Non-Parental Care, conducted by National Center for Health Statistics, as an alternative data source. The survey was a follow-up to the National Survey of Children’s Health and looked at
children in formal and informal foster care situations (such as kin care). The survey has a module on adverse childhood experiences, which includes parental incarceration.
Michael Massoglia (University of Wisconsin) mentioned a book, Give Us the Ballot: The Modern Struggle for Voting Rights in America (Berman, 2015), which argues that 30 states have actually recently enacted more stringent voting restrictions for ex-offenders. He asked Uggen if he is really optimistic about the direction the country is moving or if the gains are negated by the losses. Uggen stated that in terms of re-enfranchisement, he believes the country is moving in the right direction. Voting restriction, however, is a much broader issue, but it definitely includes and is affected by collateral sanctions. There has been work toward change, but with technology the records “stick” with individuals and are harder to overcome.
Chris Wildeman (Cornell University) and Emily Wang (Yale School of Medicine) pointed out that health surveys provide subjective measures on health while administrative records provide information on health indicators of prisoners while they are in prison—not before and not after incarceration. Such data gaps hinder the ability to gather information on the odds and prevalence of diseases for criminal justice involved populations. Wang stressed the need for indicators or variables that simultaneously capture criminal justice involvement and objective measures of health.