Abstract
Research has shown negative health outcomes from felony imprisonment. The conditions that create and exacerbate physical and mental health outcomes on the felony side—exposure to disease, lack of health care, and stress—are reflected in other less severe forms of criminal justice contact. Given that the low-level contact has grown along with prison incarceration, the health effects of less severe forms of criminal justice contact should be investigated. Using 10 waves from the National Longitudinal Survey of Youth 1997 ([NLSY97), this project explores the impact on self-reported physical and mental health from the continuum of contact, namely, an arrest, conviction, and jail sentence. The results show that low-level forms of contact negatively affect both physical and mental health throughout the continuum of contact. The role of the type of conviction is investigated, providing a more nuanced understanding of how points of contact operate on essential outcomes such as physical and mental health.
Felony incarceration has been shown to have myriad impacts on health, with the vast majority of the literature finding negative outcomes for both physical and mental well-being (Kirk and Wakefield 2018; Massoglia 2008; Schnittker and John 2007; Travis, Western, and Redburn 2014). The focus in the empirical literature has largely been on the prison population due to length of time spent in the carceral institution and the toll that prolonged sentences take on both physical and mental health. Recent research, however, indicates that less severe forms of criminal justice contact, such as an arrest or jail stay, can pose health risks (Sugie and Turney 2017; Wilper, Woolhandler, Boyd, et al. 2009). While the abbreviated time frame of low-level contact encounters may lessen the possibility of any substantial health outcomes when compared with felony imprisonment, the conditions that create and exacerbate physical and mental health for felony contact are present for low-level contact. This project explores the influence of the continuum of contact, in the form of an arrest, charge, conviction, or jail sentence, on perceptions of physical and mental health outcomes. The existing research on low-level contact shows that even an arrest or jail stay can have a negative impact on self-reported mental health (Sugie and Turney 2017). This project seeks to build on these findings by exploring the influence of extended points on the continuum of contact, including the type of conviction, on both self-reported physical and mental health.
The transmission of disease in the carceral setting, the ability to obtain health care both inside and outside of the criminal justice system, and the stress that contact brings are all hallmarks of the felony incarceration experience (Travis et al. 2014). Such experiences are common in even minor brushes with the criminal justice system. In addition, as a collateral consequence of the meteoric rise of all types of incarceration, jail and prison systems have become de facto health care facilities for state and local jurisdictions (Kirk and Wakefield 2018). Disadvantaged populations are often without adequate health services, as well as the insurance to pay for them, in their neighborhoods; therefore, first contact with diagnosis and treatment may be during a criminal intake screening. While similar effects are seen for felony imprisonment, the increased number of individuals funneled through the entry points of the criminal justice system and jail facilities heightens the potential for a more widespread impact. It is imperative to consider all forms of criminal justice contact and the jail institution as key to ascertaining the full extent of the collateral consequences of the expanded reach of the criminal justice system.
Research has shown that both physical and mental health outcomes for imprisoned populations are affected by residence in a carceral institution (Sugie and Turney 2017; Travis et al. 2014). The exposure to individuals with communicable diseases in a crowded and contained environment facilitates the transmission of disease, while the stress of adjudication and incarceration and lack of adequate medical facilities or mental health services assist in exacerbating existing conditions regardless of the type of contact. Using the National Longitudinal Survey of Youth 1997 (NLSY97) (U.S. Department of Labor 2015), I explore the effect of the continuum of criminal justice contact, from arrest and charge to conviction and sentencing, on the perception of physical and mental health outcomes to assess the full impact of criminal justice contact on the individual, their families, and communities.
Background
The established scholarship on the consequences of imprisonment provides a theoretical framework for understanding the mechanisms that link criminal justice contact and health outcomes. The deleterious effect of felony imprisonment on health has been well documented, suggesting increased transmission of diseases and the aggravation of stress-related physical and mental illnesses (Hoyert et al. 2006; Schnittker and John 2007; Schnittker, Massoglia, and Uggen 2011). While other forms of contact are less protracted, recent research suggests that similar health consequences can emerge from low-level criminal justice contact and a jail stay (Sugie and Turney 2017; Travis et al. 2014). The effect on physical and mental health has been linked to three main processes that affect health, namely, exposure to communicable diseases, access to health care services, and the stress and strain produced by contact. This project seeks to further explore the linkages between low-level contact and both physical and mental health, using the mechanisms that drive these outcomes for felony imprisonment as a theoretical framework. While these mechanisms will not be explicitly tested, they provide an important foundation for investigating the similar conditions between low-level contact and felony contact that can have substantial impacts on health and well-being. This project will also examine how the nature of contact—between violent and non-violent convictions—differentially affects health outcomes.
Felony Imprisonment and Health
Detrimental health effects of felony imprisonment on health have been well documented due to the increased exposure to communicable diseases and the aggravation of stress-related physical and mental illnesses (Binswanger, Krueger, and Steiner 2009; Schnittker et al. 2011; Wilper, Woolhandler, Boyd, et al. 2009). The existing literature documents the relative ease of chronic disease transmission given the existing conditions of the population, the lack of health care in the communities that inmates generally originate from, and the close quarters, which facilitate both the transference of new diseases and worsening of existing conditions (Hoyert et al. 2006; Massoglia 2008). Schnittker et al. (2011) reported that the levels of communicable diseases such as tuberculosis and Hepatitis C are elevated for imprisoned populations. The conditions in prison create or exacerbate medical ailments due to overcrowding, unsanitary conditions, and the stress and strain of the institution.
Due to federal and state-mandated health care provided in prison, a vein of research finds short-lived ameliorative effects on physical and mental health (Hammett, Harmon, and Rhodes 2002; Patterson 2010; Schnittker, Massoglia, and Uggen 2012). Schnittker et al. (2011) found improved outcomes for African American inmates given the availability of medical services in the carceral environment. Such findings, however, highlight the lack of health services and access to health insurance outside of prison walls, especially in predominately low-income, African American communities (Smedley, Smith, and Nelson 2003; D. R. Williams and Mohammed 2009; Wilper, Woolhandler, Lasser, et al. 2009; Williams 2007). The populations disproportionately represented in jails and prisons are also those who have limited access to medical health services and treatments in their communities (Smedley et al. 2003). Public health research has shown a wide disparity in health outcomes, especially among low-income African American males. Such outcomes result from lower rates of health insurance in these communities and decreased access to low-cost medical clinics. The increase in incarceration among these populations can potentially improve preexisting conditions due to the mandated access to basic medical screenings and treatment. Any access to health care may assist in ameliorating the medical disadvantage that exists in their outside communities. However, this effect is largely conditional on the time spent incarcerated and the availability and quality of the medical services available in the institution.
Research on imprisonment has found consistent and often sustained physical and mental health effects from the stress of the institution as well as the stigma that is experienced after release (Dumont et al. 2013). The stress of isolation, absence from family and support networks, and the threat of violence can play a role in exacerbating existing physical conditions and creating or worsening mental health (Massoglia 2008; Schnittker and John 2007; Wilper, Woolhandler, Boyd, et al. 2009). The stigma that accompanies felony incarceration and other forms of criminal justice contact can heighten stress effects, allowing them to persist after contact or imprisonment has ended. In their investigation of stigma-related effects of prison incarceration on health, Schnittker and John (2007) found that it is not the length of contact with the prison institution but any contact at all that results in pervasive stigma effects. A more recent wave of research suggests that these health impacts extend beyond the incarcerated individual and transfer to their family members and intimate partners through the financial, social, and psychological stress of incarceration (Lee et al. 2014). While an arrest or jail sentence is not as severe, low-level forms of contact have been shown to have similar effects, resulting in long-ranging mental health implications for even fleeting encounters with the criminal justice system.
Continuum of Contact and Health
According to the Bureau of Justice Statistics (BJS), the jail population rose steadily throughout the last 20 years, in line with the felony prison population. In the 12-month period ending in June 2010, almost 13 million people were admitted to local jails, with 9 million unique admissions (Minton 2011). The rising imprisonment rate captured a significant proportion of economically and socially disadvantaged individuals who often lack access to basic health services. Over one-third of male jail inmates and more than half of female inmates report an existing medical condition at admission (Minton 2011). The rate of misdemeanor adjudication has similarly increased, with an estimated 10 million cases processed every year (Natapoff 2015). The rapid increase in the populations accessing the criminal justice system either from the courts or from jails has resulted in conditions that provide prime opportunities for the transmission of disease, the worsening of existing medical illnesses, and the increased susceptibility to negative mental health outcomes as a result of heightened stress levels (Lindquist and Lindquist 1999; Sugie and Turney 2017). Sugie and Turney (2017) investigated the role of an arrest, conviction, and incarceration on mental health outcomes, finding that various points of contact, specifically arrest and incarceration, can have detrimental mental health outcomes. Furthermore, they extend their analyses to discuss how such outcomes can be patterned by race and ecological context. Such results highlight the need to investigate other forms of contact and their attendant collateral consequences. This study builds upon the work of Sugie and Turney (2017), as well as others (Pogrebin, Dodge, and Katsampes 2001; Weisheit and Klofas 1989), by exploring criminal justice contact across the continuum of contact, from arrest, charge, and conviction to jail and prison, on both physical and mental health outcomes. 1
Physical health
A number of public health studies have investigated the link between short-term imprisonment and a range of health effects, but with a concentrated focus on chronic diseases, sexually transmitted infections, and injuries. The bulk of the research finds that the exposure to communicable and infectious diseases in jail can create or exacerbate physical health due to lack of quality health care or access to medication while incarcerated (Bell et al. 2004; Binswanger et al. 2009; Conklin, Lincoln, and Tuthill 2000; Hammett et al. 2002; Javanbakht et al. 2009; Potter et al. 2011; Solomon et al. 2004; Weisheit and Klofas 1989; Wilper, Woolhandler, Boyd, et al. 2009). Wilper, Woolhandler, Boyd, et al. (2009) explored the prevalence of chronic illnesses in both the jail and prison environment, finding that a sizable proportion of inmates report conditions such as diabetes, asthma, hypertension, and HIV/AIDS upon admission. Bell et al. (2004) and Binswanger et al. (2009) focused on the public health effects of exposure to both jail and prison institutions, both finding detrimental consequences due to increased stress in the carceral environment and the uneven distribution of medical care and treatment.
In their investigation of the collateral costs of jail stays, Weisheit and Klofas (1989) found that respondents believed that their physical well-being would be affected by being in jail. Whether through the mechanisms of exposure, lack of medical care, or stress of incarceration, the individuals in the study self-reported concerns about health impacts. These perceptions reflect not only the reality of the impacts of low-level contact but also the added stress that can result from anticipating the negative aspects or ultimate outcomes of contact. While Weisheit and Klofas only focus on the jail experience, it is possible that other forms of contact can also influence self-reports and perceptions of health consequences.
Similar to felony imprisonment, low-level criminal justice contact reflects the existing racial health disparities, with the institution becoming a de facto health clinic for basic diagnosis and treatment (Dumont et al. 2013). Empirical work has suggested that time in the carceral institution can provide access to medical screenings and treatment for a segment of the population that is not generally privy to these services (Hammett et al. 2002; Schnittker et al. 2011). However, access within jails is not evenly distributed and is often predicated on the size, location, and available resources of the facility (Lindquist and Lindquist 1999; Wilper, Woolhandler, Boyd, et al. 2009). Due to the rapid cycling of jail inmates, medical assessments are often limited, while diagnosis and treatment are difficult to obtain. Exposure to communicable and infectious diseases in the jail facilities can lead to the establishment or exacerbation of medical complications due to lack of quality health care and/or access to medication while incarcerated (Bell et al. 2004; Binswanger et al. 2009; Wilper, Woolhandler, Boyd, et al. 2009). Wilper, Woolhandler, Boyd, et al. (2009) found divergent levels of access to medical care, with the vast majority of jail inmates (68.4 percent) reporting that they did not receive a medical screening, compared with 20.1 percent in state prisons and 13.9 percent in federal prisons. For those progressing through the system, no such services are available due to the existing health care disparities that exist in underserved and disadvantaged communities.
The vast majority of studies on physical health and low-level contact have focused on jail and the mechanism of exposure as the main driver of negative health impacts. However, even brief detention by law enforcement can have substantial consequences (Sugie and Turney 2017; Weisheit and Klofas 1989). While not as protracted as felony contact, an arrest represents a separation from society, removing the individual from their daily life and exposing them to conditions that may heighten risk of injury or disease. The brief incapacitation of an arrest can make it difficult to attend to health-related concerns, such as attending doctor’s appointments or picking up prescriptions. Recent work by Kohler-Hausmann (2013) and Gonzalez Van Cleve (2016) suggested that dealing with a charge or conviction can represent a substantial time burden, taking an individual away from daily responsibilities. In particular, Kohler-Hausmann (2013) introduced the concept of the “procedural hassle” of misdemeanor cases, describing the process that requires defendants to traverse back and forth to attend to the charge, with these court dates often characterized by the “endless waiting” for adjudication (Gonzalez Van Cleve 2016). Therefore, incapacitation in the case of low-level criminal justice contact extends not only to a jail stay but can be attributed to an arrest, charge, or conviction. An arrest can keep a person from their daily responsibilities, if only temporarily, while the ongoing process that comes with a criminal charge can compel the individual to engage in the “procedural hassle” of criminal justice processing, from attorney meetings to court hearings to community service or incarceration. In addition, the adjudication process and the weight of a potential conviction may cause attendant health consequences, exacerbated by the stress of the process (Schnittker and John 2007).
Mental health
The stress of processing and the resulting collateral consequences of low-level contact can be detrimental to the perception of mental health outcomes. Existing research has shown that felony incarceration and jail stays can result in the creation or exacerbation of stress-related illnesses (Binswanger et al. 2009, 2011; Lindquist and Lindquist 1999; Massoglia 2008). However, contact without incarceration can increase stress responses due to the anticipatory stress and strain of impending court proceedings and the separation from family and everyday routines (Freudenberg 2001; Sugie and Turney 2017; Turney, Wildeman, and Schnittker 2012; Weisheit and Klofas 1989). The recent work of Sugie and Turney (2017) investigates the potential impacts of an arrest, conviction, and incarceration stay on mental health, using the stress process paradigm and data from the NLSY97. Their findings show the negative mental health outcomes for an arrest and incarceration, with an emphasis on how such consequences are increasingly detrimental for highly disadvantaged communities, especially those characterized by high rates of poverty, unemployment, and female-headed households. Specifically, Sugie and Turney (2017) found that arrests are incredibly influential, explaining almost half of the association between incarceration and mental health. Their work suggests that a variety of primary stressors are involved in shaping mental health outcomes from an arrest, including the “procedural hassle” (Kohler-Hausmann 2013) and the stress and stigma that accompanies an arrest. While these primary stressors can be influential, Sugie and Turney suggest that the secondary stressors of anticipated or realized collateral consequences of the arrest, including impacts on employment, housing, and family relationships, may also play substantial role for mental health. This work, as well as others, suggests that even the most minor form of contact—an arrest—can result in negative self-reported mental health due to not only the primary stressors of contact but also the secondary stressors that result from an arrest. Such findings underscore the utility of looking beyond incarceration when considering the collateral impact of criminal justice processing and adjudication. This study offers an extension of Sugie and Turney’s (2017) work, looking at both physical and mental health outcomes and an extended range of contact points, including type of conviction, to provide a nuanced context for these outcomes.
The impact on health outcomes may be predicated on the type of offense, either based on severity or type. More serious convictions require extended time engaging with the criminal justice system, which may affect incapacitation and consequently health outcomes. Furthermore, the type of offense may increase the stress of impending adjudications as well as the consequences of those proceedings. Heightened stress responses are influential not only for mental health outcomes but also for physical health (Binswanger et al. 2009; Massoglia 2008; Wilper, Woolhandler, Lasser, et al. 2009). Investigating how the nature of a conviction affects both physical and mental health may provide a more nuanced understanding of how contact operates on all levels in terms of health outcomes.
Existing research on the collateral consequences of criminal justice contact suggests that conviction type can be a factor in the experiences post-contact (Vuolo et al. 2017). While research on the role of conviction type and health outcomes is limited, the existing literature does suggest that there are differentiating factors for certain forms of conviction compared with others (Steffensmeier et al. 1998; Weisheit and Klofas 1992). More serious convictions for violent offenses, for instance, may have a greater impact on mental and physical health as compared with non-violent offenses because they require more time engaging with the system. However, it is also possible that non-violent offenses, which are most common, can be just as influential because despite the decreased severity of the offense, the time required to resolve such a conviction can be substantial. Gonzalez Van Cleve (2016) and Kohler-Hausmann (2013) described the time constraints and demands that misdemeanor justice requires, especially for marginalized populations, thereby increasing the potential stressors of even a minor conviction. For first-time offenses, the stress and strain and potential mental health impact may be even greater due to the steep learning curve required to navigate the system and the uncertainty of the process. Whether these impacts are influenced by increased time in the system, heightened stress of separation from support systems and everyday responsibilities as well as the anticipatory stress of the impending criminal justice process remain largely understudied. This project offers an overview of how the continuum of contact affects perceptions of physical and mental health as well as findings on the role that conviction type plays on these outcomes.
Data and Method
Using the NLSY97, I evaluate the effect of the continuum of contact, namely, arrests, convictions, and jail sentences on perceived physical and mental health outcomes. The data set offers the ability to track respondents through 14 waves of data from 1997 to 2010, and assess their level of contact with the criminal justice system. The initial 1997 cohort consists of 12- to 16-year-old young adults, with an oversample of African Americans and Latinos. The sample for all waves ranges in age between 12 and 31 years; however, for these analyses, the sample was restricted to those over 18 years of age, resulting in an overall sample of 8,984 and over 74,000 person-waves, with a retention rate of 83.2 percent across all waves, to accurately measure adult health outcomes and adult criminal justice contact. The NLSY97 also provides a variety of health-related measures, ranging from parent’s assessment to the child’s health as well as behaviors that may have detrimental or ameliorative effects on overall health. There are only a limited number of data sets that capture both the incidence and effect of other forms of criminal justice contact. The NLSY97 provides a substantial amount of data on all forms of criminal justice contact most often left out of traditional large-scale surveys. Those surveys that do inquire about contact usually only ask about prison incarceration or aggregate both jail and prison stays into one incarceration measure. The NLSY97 is unique in that it records criminal justice contact and incarceration for a variety of contact points (arrest, charge, conviction, jail sentence, and prison sentence) and also provides a wide array of health variables to measure the potential impact of contact.
Health Measures
For physical health, the measure is a self-reported assessment of general health for all 10 waves of data: “In general, how is your health?” coded 1 for poor, 2 for fair, 3 for good, 4 for very good, and 5 for excellent. 2 For mental health, a scale variable is used that is comprised of five mental health variables. The respondents were asked how often in the past month they felt (1) nervous, (2) calm and peaceful, (3) downhearted and blue, (4) happy, and (5) down in the dumps. The response categories are a 4-point Likert-type scale, with responses ranging from all of the time, most of the time, some of the time, and none of the time. Scale reliability was .93 in each year that the variables were assessed. While the physical health measure was asked in every wave, mental health measures were only captured in 2002, 2004, 2006, 2008, and 2010, limiting the number of person-waves available for analysis to 42,214. The mental health scale was coded with 1 representing negative self-reported mental health, while 5 represents more positive self-reports of mental health. 3
Criminal Justice Contact
Five measures are used to explore the trajectory of criminal justice contact and its potential impacts on health outcomes: arrest, charge, conviction, jail sentence, and prison sentence. 4 These are dichotomous measures, asking respondents if they have experienced each point of contact during the interview year. For analysis, a factor variable was created for all five points of contact, which allows for cross-contact comparison. In analysis, the arrest measure represents only the effect of an arrest. However, the charge measure represents the effect of an arrest but no other points of contact beyond a charge. The conviction measure represents the effect of an arrest, charge, and conviction but not the sentencing measures. All contact measures were provided by the NLSY97, asking respondents if they have experienced that form of contact in the last calendar year. Including all of the contact variables within the model allows for an evaluation of the differential impacts of these points of contact as well as changes to outcomes based on the severity of the contact. In addition, the NLSY97 provides dichotomous measures for the type of conviction for all major offense types: assault, robbery, burglary, theft, drug possession, drug sales, property crimes, public order, and traffic. The measures were isolated to contain the most severe charge (and not co-occurring charges and convictions) and then separated into a violent conviction category 5 and a non-violent conviction category. Table 1 provides counts for these variables, both individually and within their combined categories.
Count of Contact Variables and Count of Conviction Type and Offense.
There were missing values for approximately 338 conviction-type variables, with the conviction type not classified by the respondent. This missing data explain the discrepancy between the total count of convictions and the conviction-type totals.
Covariates
All models control for basic demographic variables such as race, gender, marital status, presence of children in the household, immigration status, education level, and cohort. 6 Year dummy variables were also added to capture the effect of time on the models. The NLSY97 also includes measures of cigarette and alcohol consumption behavior, asking respondents how many cigarettes they smoked 7 and how many drinks they consumed per day in the last 30 days. In addition, there is a measure asking if the respondent had health insurance in the last year to control for access to health care. To capture previous health concerns, there is also a measure of the parental report of the respondent’s health in 1997, asking parents to assess the nature of their child’s physical health, from poor to excellent. In addition, a measure was included that gives the parent’s self-reported health in 1997, rating it from poor to excellent. These two were included in the models for physical health. For mental health, a measure was included of the parental report of the respondent’s mental health in 1997 as well as the respondent’s assessment of their own mental health in 1997.
Method
To analyze the impact of the continuum of contact on perceived physical health, I use multi-level mixed effects ordinal logistic regression models in Stata 8 with the individual respondents nested in waves of the survey. The outcome variable is ordinal; therefore, ordered logistic regression is needed to properly fit the models and produce unbiased results (Armstrong and Sloan 1989; Hartzel, Agresti, and Caffo 2001). For the mental health scale, I use multi-level mixed effects generalized linear models (GLMs) 9 due to the continuous nature of the outcome measure, with individuals nested in waves of the NLSY97.
Multi-level mixed effects ordered logistic regression and GLMs were chosen because they offered the ability to test fixed effects with a random intercept at the individual level for both outcome variables (Peng and Lu 2012). Mixed effects models offer a flexible alternative to generalized linear regression or ordinal logistic regression models, with fixed effects estimated directly, while the random effects in this particular analysis are used on the respondent level as a random intercept. Mixed effects models are beneficial for longitudinal panel data because they explicitly model individual change across time, especially for data with repeated observations for individuals within waves. Models using fixed effects assume independence of variables, but given the longitudinal nature of the data used for the study, it is reasonable to assume that variables across waves may be correlated. Modeling both the fixed and random effects allows for a more nuanced analytic approach for repeated measures (Gelman and Hill 2007). In addition, mixed effects models offer the ability to model multiple sources of random variation by incorporating both a fixed effects component and a random effects component, which allows for variation for individuals over time.
A model was run for each outcome, physical and mental health, with all five levels of contact included as a factor variable: arrest, charge, conviction, jail sentence, and prison sentence. In terms of timing, the health outcome was selected after the initial and most severe point of contact. The reference group for both models are those respondents who have not experienced any points of contact. For the models investigating the impact of the type of conviction, the contact variables were inserted into the model, with the conviction measure separated into violent and non-violent categories within a factor variable.
Results
Table 2 provides results for the multi-level ordered logistic regression models for physical health, showing a complex and nuanced impact of criminal justice contact on health outcomes . 10 The least severe forms of contact—an arrest or charge—do not seem to have an effect on physical health outcomes as compared with other forms of contact, possibly due to the quick resolution of these points of contact, limiting the possible exposure to infectious diseases or time away from medical treatments for existing conditions. However, conviction has a negative and significant impact on self-reported health, suggesting that the process of advancing to this stage in criminal justice adjudication may be detrimental to perceptions of physical health. The potential mechanisms behind this result could be that the stress of an arrest, the adjudication process, and the weight of a conviction may cause attendant health consequences, exacerbated by stress. Being convicted of a charge may also spur concerns about the consequences of the judgment and the sentencing implications.
Mixed Effects Ordinal Logistic Regression: Contact and Physical Health.
Note. Standard errors in parentheses; year dummies were included in models but omitted due to space considerations.
p < .05. **p < .01. ***p < .001.
The effect of a jail sentence does not show a significant result; however, a prison sentence does show an improvement of health outcomes. The neutral result for a jail sentence could be due to the abbreviated nature of jail stays, especially for those awaiting trial, and the limited amount of medical health services provided. Prison sentence has a positive effect on physical health, confirming previous empirical findings that show an ameliorative effect of incarceration on health (Schnittker et al. 2011). The access to even a modicum of health care within the institution can provide a temporary positive benefit on the perception of health.
The controls operated similarly to the established literature with access to health insurance and higher levels of educational attainment resulting in better perceived physical health. The parent’s reporting of both the youth’s health and their own health in 1997 shows a significant and positive effect on self-reported physical health. These two measures serve as a baseline for self-reported health to control for existing health conditions that may affect the overall result. As expected, those who reported increased smoking and drinking behaviors had lower perceived physical health.
Table 3 provides results for the models for mental health, showing that criminal justice contact in all forms results in higher rates of poor mental health. 11 All forms of contact show a negative and significant effect on the mental health scale. Interestingly, most of the coefficients for contact, with the exception of a conviction, are all higher than those for a prison sentence. This result suggests that low-level contact is a salient influence on mental health outcomes. The impact of an arrest is substantial, signaling the importance of even relatively minor forms of contact. These findings could be the result of the collateral consequences on employment, housing, and family life that accompany all forms of contact. This may cause individual to despair given that a criminal record of any type can be detrimental to employment or educational prospects. In addition, an arrest can destabilize an individual’s sense of self, putting them into an emotional tailspin, resulting in higher rates of depression and anxiety. However, the result for a prison sentence could also be a factor of the small number of respondents who had been sentenced to prison in the sample. Further investigation of this relationship should be investigated with the future waves of the NLSY97.
Mixed Effects Regression: Contact and Mental Health Scale.
Note. Standard errors in parentheses; year dummies were included in models but omitted due to space considerations.
p < .05. **p < .01. ***p < .001.
The control variables operate in line with existing research. For instance, health insurance, being married, and higher educational status decrease depressive symptoms. The negative and significant effect for the number of drinks consumed in a 30-day period suggests that substance use may be part of the relationship between contact and outcomes affecting mental health. Much of the work on imprisonment—whether short term or long term—focuses on co-occurring substance use that aids in exacerbating existing mental health concerns (Travis et al. 2014).
In Table 4, the conviction results for physical health (Model 1) suggest an interesting pattern, with violent convictions showing no significant results, but non-violent convictions with a negative and significant result. 12 This finding points to the need for further exploration of the nature and type of convictions and the potential for differential impacts. The positive and significant result for prison sentence stays consistent from the main model, while charge registers a marginally significant effect. The differential impacts between violent and non-violent convictions may be the result of a few potential factors. The age of the sample is relatively young; therefore, they are more likely to have committed a non-violent offense rather than a violent offense. As seen in Table 1, non-violent convictions, including traffic, drug, and property offenses, represent the majority of contact for the sample. These results represent a divergence from previous research that suggests that the severity of the offense may be influential in terms of outcomes from contact. Therefore, we should see violent convictions being more salient in terms of the impacts because non-violent convictions, arguably, take less time to adjudicate and resolve. These results suggest further investigations into types of conviction and their potential differential impacts on both physical and mental health. Part of the explanation for these results may be an artifact of the age of the sample, with respondents more likely to have a non-violent conviction on their record. In fact, one of the most common types of conviction for this age group is traffic violations, a relatively minor form of conviction. 13 These results could also speak to differences in the timing of the offense. If this is a first offense, the respondents may be ill-equipped, both in terms of information and resources, for the “procedural hassle” that Kohler-Hausmann (2013) described and the toll that it can take on one’s physical and mental health.
Mixed Effects Model for Physical and Mental Health and Conviction Category.
Note. Standard errors in parentheses; year dummies were included in models but omitted due to space considerations.
For Model 1, the parent’s report of youth’s health and parent’s self-reported health are measures of physical health. For Model 2, both variables are measures of mental health.
p < .05. **p < .01. ***p < .001.
In Model 2, in Table 4, the results for mental health diverge from the physical health results. In particular, both violent and non-violent convictions seem to have an impact on mental health, even when controlling for other contact variables. The other contact variables are consistent with the Model 1 in Table 3, showing negative and significant effects from an arrest to a prison sentence. The effect of violent and non-violent convictions may signal the impact of not only the “procedural hassle” of dealing with a conviction but also the stress and strain that come from criminal adjudication. The concern about the potential implications of any conviction, regardless of type, may create or exacerbate depressed or anxious feelings, prompting the individual to perceive a decline in their overall mental health. The control variables show a similar patter to the other results, with health insurance and higher educational attainment resulting in positive physical and mental health outcomes, while increased drinking and cigarette use having a negative impact on both outcomes. Overall, these findings suggest that conviction type may be a salient point of investigation for determining how contact renders impacts on health, but that such impacts may differ across physical and mental health outcomes.
Discussion
The empirical work on both felony and low-level contact presents a complex landscape of physical and mental health–related outcomes (Bell et al. 2004; Patterson 2010; Schnittker et al. 2012; Sugie and Turney 2017). This project builds on the foundations of such work, exploring the outcomes of criminal justice contact on self-reported physical and mental health. The results suggest that there are negative relationships between the least severe forms of contact and health outcomes, and therefore, the continuum of contact should continue to be taken into account when assessing the costs and consequences of criminal justice contact. Additional analysis shows differential impacts depending on conviction type, suggesting that more nuanced analysis can be useful in understanding how these points of contact operate on health outcomes.
These results highlight the importance of looking beyond prison to a broader continuum of criminal justice contact and their potential impacts on physical and mental health outcomes. As these results suggest, even a conviction can negatively affect an individual’s perception of health, potentially erecting another barrier to gainful and consistent employment and other markers of stability. The health impacts of felony imprisonment were thought to hinge on the length of time in the carceral institution and the severity of the conviction. Due to prolonged sentences and harsh surroundings, the prevailing empirical conclusion settled on time out of society and length of exposure as the key mechanisms that drove negative health outcomes. These results suggest that criminal justice contact can be detrimental even at the earliest stages of the adjudication process and not necessarily for the most severe offenses. The results for conviction show that perceptions of physical and mental health are affected even for non-violent convictions. Therefore, it is not necessarily about the length of sentence or the severity of contact but rather any level of exposure to the criminal justice system that can either create stress-related illnesses or exacerbate existing physical and mental health conditions.
The findings for mental health suggest an even more pervasive and pernicious impact. From an arrest to a conviction to a jail sentence, results show an impact on self-reported mental health across all points of contact. These findings suggest that there may be multiple mechanisms at play, with time engaged with the criminal justice system operating in conjunction with the stress and strain of the proceedings. The potential role of anticipatory stress is particularly relevant, with negative effects being seen in the initial stages of contact—arrest and charge, for instance—suggesting that the concern about next steps in the contact continuum may play a role in affecting perceptions of current mental health. Similar to Sugie and Turney’s (2017) findings, the role of an arrest remains a salient influence on mental health. This finding alone suggests an extension of the reach of both the criminal justice system and the attendant collateral consequences that result from contact. As Sugie and Turney (2017) noted in their analysis, the primary stressors of an arrest can affect respondent’s perceptions of their mental health due to the hassles and time constraints of traversing the system. Furthermore, they note the contribution of secondary stressors, borne out of the concern about the impacts of an arrest on employment, housing, and familial responsibilities . The findings from Sugie and Turney’s research as well as this current study show that the collateral consequences of criminal justice contact can and do extend far beyond felony incarceration.
When a decrease in mental health can be seen across the continuum of contact, from an arrest to a conviction to a jail sentence, this suggests that the reach of the criminal justice system has become deeper and wider than previously documented. This reach has the potential to extend far beyond the holding cell or pre-trial adjudication to consequences that may have far-reaching implications beyond the immediate point of contact. The negative effect on mental health can have a similar impact to those of physical health–related concerns; individuals are hindered in their ability to attend to everyday responsibilities, which can affect their successful reentry into society, even after a short-term stay for an arrest or jail sentence. Without access to reentry and mental health treatment services, such impacts can spiral into wide-reaching consequences that can aid in losing a job, educational opportunity, or possibly reverting to criminal offending.
Reentry is often discussed in terms of a return from felony imprisonment; however, these findings suggest that the discussion should include the reentry processes and consequences that result from contact. While not as severe, these results show the potential for worsening reports of physical and mental health as a result of a conviction or even an arrest, respectively. Negative perceptions of physical and mental health can impede individual’s efforts toward a successful reentry by increasing their absences from work, making them unable to financially attend to core responsibilities such as paying rent or a mortgage or providing for family members or paying court fees. For those employed in the secondary labor market, where flexibility and time off are at a premium, any absence could place a job in jeopardy (Crutchfield 1989). Add to that increased absences or time off to attend to subsequent medical or mental health concerns post-contact, and the ties to gainful and steady employment are increasingly frayed. The effect of contact at any level may further isolate these individuals and communities by restricting potential job and educational opportunities due to their criminal record and poor physical and mental health. The nature of these relationships can provide insight into how even less severe forms of contact can influence health outcomes and can heap more disadvantage onto those individuals who face economic, occupational, educational, and social barriers.
Several important limitations exist. The NLSY97 only records self-reported contact and is not verified by official records. It is possible that respondents may omit incidents of criminal justice contact. More importantly, the NLSY97 does not provide a clear and systematic way to separate out misdemeanor points of contact from those of felony contact, except in terms of incarceration. Therefore, the arrests, charges, and convictions in this article could be for either a misdemeanor or felony. Differentiating between these two is an important step for future research. Also, the number of people in the original population that have had contact is relatively small, which is a challenge given the nature of the research questions being posed for this current project. However, there are a limited number of data sets that ask questions about contact beyond prison stays and none that are currently available that tap into questions about outcomes. Therefore, the NLSY97 provides a unique opportunity to gauge the effect of incarceration and less severe forms of criminal justice contact and evaluate the mechanisms that undergird these outcomes. The collateral consequences of short-term incarceration may be cumulative over time; however, the NLSY97 is not necessarily able to show such effects due to the time frame in which the data were collected. For project of this kind, however, the NLSY97 can provide an understanding of the potential outcomes of an arrest as compared with both short-term and extended incarceration.
While the implications of a negative health impact loom large for felony incarceration, those arrested or sentenced to a jail stay constitute a larger segment of the population, with jails reporting around 9 million individual admissions per year (Minton 2011). The majority of these individuals are employed and are supporting families at the time of contact. Therefore, widening the net to look at lower level forms of criminal justice contact allows a greater understanding of the broad implications of criminal justice expansion that occurred alongside the prison boom. The effects of even low forms of contact can extend beyond court dates and jail stays to raise levels of inequality and disadvantage by affecting access to employment, education, housing, and other essential markers of stability. Investigating these low-level forms of contact may be key to truly understanding the interplay between the criminal justice system and disadvantage and inequality.
Footnotes
Acknowledgements
The author thanks anonymous reviewers, Robert Crutchfield, Becky Pettit, Hedwig (Hedy) Lee, Jerald Herting, and Katrina Leupp for helpful comments on earlier drafts.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
