Citation
Duration of incarceration and the mental health of currently incarcerated inmates

Material Information

Title:
Duration of incarceration and the mental health of currently incarcerated inmates
Creator:
Menard, Alyssa ( author )
Language:
English
Physical Description:
1 electronic file (49 pages) : ;

Subjects

Subjects / Keywords:
Prisoners -- Mental health ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
This study examines the association between duration of incarceration in state prisons and mental health treatment needs in currently incarcerated men and women. Using nationally representative data from the 2004 Survey of Inmates in State Correctional Facilities, duration of incarceration is measured using number of prior incarcerations and duration of time served in one’s current sentence. A four-category typology of mental health treatment status is used to assess the association between duration of incarceration and mental health status in currently incarcerated inmates. Results from a stratified multivariate logistic regression analysis show that duration of incarceration is significantly associated with mental health treatment for currently incarcerated inmates. This study demonstrates that lengthier incarcerations lead to more mental health treatment needs for inmates who are currently incarcerated. These results demonstrate that incarceration is directly correlated with the mental health of inmates and that there are unmet mental health treatment needs in the prison population.
Thesis:
Thesis (M.A..)--University of Colorado Denver
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Alyssa Menard.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10088 ( NOTIS )
1008887495 ( OCLC )
on1008887495

Downloads

This item has the following downloads:


Full Text
DURATION OF INCARCERATION AND THE MENTAL HEALTH OF CURRENTLY
INCARCERATED INMATES by
ALYSS A MENARD B.A., Norwich University, 2013
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Arts Sociology Program
2017


This thesis for the Master of Arts degree by Alyssa Menard has been approved for the Sociology Program by
Adam M. Lippert, Chair Stacey J. Bosick
Keith W. Guzik


Menard, Alyssa (M.A., Sociology Program)
Duration of Incarceration and the Mental Health of Currently Incarcerated Inmates Thesis Directed by Associate Professor Adam M. Lippert.
ABSTRACT
This study examines the association between duration of incarceration in state prisons and mental health treatment needs in currently incarcerated men and women. Using nationally representative data from the 2004 Survey of Inmates in State Correctional Facilities, duration of incarceration is measured using number of prior incarcerations and duration of time served in ones current sentence. A four-category typology of mental health treatment status is used to assess the association between duration of incarceration and mental health status in currently incarcerated inmates. Results from a stratified multivariate logistic regression analysis show that duration of incarceration is significantly associated with mental health treatment for currently incarcerated inmates. This study demonstrates that lengthier incarcerations lead to more mental health treatment needs for inmates who are currently incarcerated. These results demonstrate that incarceration is directly correlated with the mental health of inmates and that there are unmet mental health treatment needs in the prison population.
The form and content of this abstract are approved. I recommend its publication.
Approved: Adam M. Lippert
111


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION....................................................1
II. BACKGROUND AND LITERATURE REVIEW................................1
A Brief History of Incarceration in the United States............3
Incarceration and Mental Health..................................9
Duration of Incarceration.......................................12
Stress, Coping and Institutional Acclimations.............13
Stress Process Theory.....................................14
Total Institutions........................................16
Current Study.............................................18
III. DA I A AM) METHODS.............................................20
Survey of Inmates in State Correctional Facilities..............20
Dependent Variables.............................................20
Mental Health Treatment...................................20
Independent Variables..........................................21
Duration of Incarceration.................................21
Analysis........................................................21
IV. RESULTS........................................................23
Descriptive Statistics......................................23
Multinomial Regression......................................26
V. DISCUSSION.....................................................31
IV


Strengths and Limitations
35
Conclusion.............................................36
REFERENCES..........................................................38
v


LIST OF TABLES
TABLE
Table 1. Descriptive Statistics of Entire Sample.......................23
Table 2. Multinomial Logistic Models Predicting the Odds of Mental Health
Treatment Compared to Inmates Who Have Never Received Treatment........25
Table 3. Multinomial Logistic Models Predicting the Odds of Mental Health Treatment Compared to Inmates Who Have Met Need..................27
vi


LIST OF FIGURES
FIGURES
Figure 1. Inmates Receiving Mental Health Treatment.....................24
Figure 2. Predicted Probability of Mental Health Treatment Categories By Duration of Incarceration........................................................29
vii


CHAPTER I
INTRODUCTION
Research has increasingly demonstrated that exposure to incarceration exacerbates mental health disorders and complicates mental health treatment needs among imprisoned persons (Massoglia 2008b; Nowatny, Cepeda and James-Hawkins 2015; Schnittker, Massoglia, & Uggen 2012; Turney, Wildeman and Schnittker 2012). Like the non-institutionalized population, common mental disorders experienced by inmates include anxiety disorders, mood disorders, impulse control disorder and substance disorders (Kessler, Berglund, Chiu, Dernier, Heeringa, Hirpi, Jin, Penned, Walters, Zaslavasky and Zheung 2006) However, unlike the general population, the prevalence of such disorders is exponentially higher among incarcerees. Approximately 34% of individuals within the general population have experienced some type of mental health disorder within the last twelve months, whereas over half of all inmatesnearly 56%have been treated for some type of mental health disorder while incarcerated (James and Glaze 2006; Yu, Sung, Mellow and Koenigsmann 2014; Felson, Silver and Remster 2012; Kessler 2001).
Mental health disorders affect a substantial proportion of inmates and have lasting consequences during and after incarceration. More than half of all inmates report a wide range of mental health symptoms including mania, major depression, psychotic disorder, anxiety and mood disorder, impulse disorders, and substance use or dependence (James and Glaze 2006; Kessler 2001; Schnittker, Massoglia and Uggen 2012). Further, many inmates experience co-morbidity or suffer from more than one mental disorder at a time (Peters, Lurigio, and Wexler 2015). Inmates are not only at elevated risk for experiencing multiple
1


mental health co-morbidities, but they also face higher prospects of additional challenges post-release including homelessness, employment discrimination, difficulty maintaining continuity of care, lack of social support, increased stigma, relapse, and higher rates of rearrest compared to former inmates who were not treated for mental health disorders while incarcerated (Schnittker and John 2007; Williams et al. 2010; Binswanger et al., 2011b; Yu eta. 2014).
Two theoretical perspectives offer insights on how different dimensions of incarceration might influence mental health treatment seeking among inmates. Stress process theory (Pearlin 1989) holds that lengthy incarcerations coupled with a history of many such incarcerations will undermine inmate mental health, while Goffmans total institution theory (1961) contends that the institutional familiarity generated by lengthy or repeated prison terms may generate specialized skills and knowledge that inmates can deploy to protect their mental health while incarcerated.
Prior literature has devoted attention to the relationship between incarceration and mental health, but few studies have simultaneously examined how the duration of current time served and ones number of prior detainments are associated with mental well-being and treatment for new or existing mental health problems. Most prior studies on imprisonment and mental health attend to onebut not bothof these dimensions of institutional exposure. However, a more robust specification of such exposure would combine measures of both time served of ones current sentence and ones history of prior incarcerations in order to examine the cumulative exposure to incarceration.
Both prior incarcerations and duration of current time served measure different aspects of institutional exposure. Duration of current time served examines consequences of
2


uninterrupted institutional exposure whereas number of priors measures multiple experiences of institutional exposure. Prior incarcerations as a measure of institutional exposure attends to aspects of incarceration such as job interruption, dislocation from society, separation from friends and family members, as well as the potential for exposure to different institutional environments with repeated incarcerations. Duration of current time served attends to similar yet different aspects of institutional exposure. Inmates who spend more time incarcerated also experience similar consequences as those who serve repeated sentences, however, as duration of time served increases the severity or magnitude of such consequences may be amplified. Moreover, as time served increases, there are additional opportunities for inmates to develop institutional knowledge, coping mechanisms, or to encounter additional risk factors while incarcerated. By examining both repeated incarcerations and length of incarceration this study will be able to determine how various aspects of cumulative exposure are associated with the mental health treatment of currently incarcerated inmates.
Although there has been much research related to the health consequences of incarceration and the mental health of inmates, a majority of studies use general population data to explore differences in the mental health of inmates compared to the general population. This approach fails to capture differences among the mental health statuses of currently incarcerated individuals and may introduce selection bias into prior studies. Using data that compares inmates with the general population ignores much heterogeneity found within inmate populations due to similar risk factors associated with socioeconomics and childhood adversities that may be correlated with both health and crime. Moreover, it is also unclear how selection bias influences prior estimates linking various aspects of incarceration to mental health treatment. Using data that captures the mental health status of inmates while
3


they are currently incarcerated helps to attenuate for this bias and gain insights into the different pathways associated with the effects of incarceration on the mental health of prisoners.
In light of these knowledge gaps, the current study examines the joint and independent associations among current incarceration duration, incarceration history, and mental health treatment using the 2004 National Survey of Inmates, a nationally representative study of currently incarcerated U.S. adults. Specifically, the current study uses multinomial regression modeling using a typology of mental health status to address the following questions:
1. How are incarceration history and time served associated with the odds of having a new mental health treatment experience vs. having never had mental health treatment?
2. How are incarceration history and time served associated with the odds of having an unmet mental health treatment need vs. having continuity in ones mental health treatment?
3. Are the associations between incarceration history, time served, and mental health treatment patterns independent or conditional on one another?
4


CHAPTER II
BACKGROUND AND LITERATURE REVIEW A Brief History of Incarceration in the US
Over the last half-century, excessively punitive sentencing policies have lead to mass incarceration in the United States. The United States incarcerated population has increased from approximately 600,000 in the 1960s to more than 2.2 million in 2014 (Carson 2015). Incarceration in the United States is such a common social phenomenon that nearly one out of every 100 American citizens will end up in jail or prison at some point in their lives (Pew Cent 2008). Of those incarcerated, approximately 95% will be released back into their communities, resulting in roughly 16 million ex-convicts in the United States (Schnittker Massoglia & Uggen 2012).
Much of the dramatic rise in imprisonment can be associated with changes in political policy throughout the latter half of the 20th century resulting from political strategies enacted partly in response to growing racial tensions and claims made by civil rights advocates (Dumont Allen, Brockmann, Alexander and Rich 2013; Alexander 2010). Some examples of policy implementations aimed at reducing crime include the War on Drugs, Tough on Crime, rhetoric and mandatory minimum sentencing laws (Wakefield and Uggen 2010; Dumont et al. 2013; Howell, Feld and Mears 2012; Cid 2009). The federal government incentivized adherence to these policies with cash grants for participating states, which led to an increase in number of arrests as well as harsher and longer sentences, especially in regard to drug crimes (Alexander 2010). Although the intent of these policies was to decrease crime rates, these policies have generated structural oppression of entire groups of people affecting
5


those of socioeconomic disadvantage and minority status in a highly inequitable way (Dumont et al. 2013; Alexander 2010).
Incarceration as a form of criminal punishment has far reaching consequences for inmates while imprisoned, and, for both inmates and communities after inmates are released. Research has shown that inmates, while incarcerated, are at elevated risk for poor physical and mental health outcomes (Sung and Mellow 2011; Binswanger et al. 2009). Some examples include increased risk of exposure to infectious or communicable diseases as well as increased risk for experiencing mental health disorders such as depression and anxiety related to the stressors of the prison environment (Massoglia 2008; Porter 2015; Schnittker and John 2007; Binswanger et al. 2009).
Another consequence of incarceration is the mark or stigmatization associated with serving time. This stigmatization contributes to many continued risk factors associated with incarceration after release from prison including homelessness, recidivism, increased substance abuse, poor health behaviors, employment discrimination, and difficulty maintaining social bonds (Binswanger et al. 2011a; Porter 2015; Schnittker and John 2007; Williams et al. 2010). For instance, Pager (2003) finds that job seekers with a criminal record are one half to one third as likely to gain employment compared to non-offenders resulting in chronic unemployment for previously incarcerated individuals. In a study conducted by Western et al. (2015), it was found that after six months of release, nearly half of ex-inmates were still un/underemployed. Consequently, former detainees report significantly lower personal earnings than those without criminal records. Because social service programs often include eligibility requirements that exclude ex-inmates (Wakeman, McKinney and Rich 2009) chronically unemployed former detainees often have unmet needs for basic social
6


services like food assistance (Harding 2014), housing support (Western, Braga, Davis, and Sirois 2015), and clinical re-entry needs (Gill and Wilson 2016).
It has also been shown that incarceration has a criminogenic effect and increases the likelihood of recidivism (Nagin, Cullen and Johnson 2009; Cid 2009; Bales and Piquero 2011). Of all inmates who have been previously incarcerated more than half will recidivate within three years of release (Wakefield and Uggen 2010). For some former detainees, repeat offenses sometimes constitute adaptive means to secure basic resources (e.g., food, cash, clinical needs) that are unavailable to them due to employment discrimination and a general lack of public assistance programs for ex-inmates (Gill and Wilson 2016; Williams et al. 2010). Irrespective of the causes provoking criminal offenses following release, ex-inmates who have their basic needs met through re-entry services or by other means such as family and social support, are less likely to recidivate compared to ex-inmates who experience a high level of unmet need post release (Gill and Wilson 2016).
Although there are empirically demonstrated disadvantages to using incarceration as a primary form of punishment in American society (Thomas, Lavandowski, Isler and Wilson 2007; Maruschak 2008; Kinner and Milloy 2011; Donahue 2014), the prison industry has grown exponentially in the last half-century. The rapid expansion of punitive sentencing policies during the latter part of the 20th century coupled with the privatization of prisons generated a major economic opportunity for big business to capitalize on the free market establishing the prison industrial complex (PIC) (Palaez 2008; Dumont et al. 2013). In the years following the implementation of such policies the increase in punitive action including mandatory minimum sentences and three strikes policies by the government created new market space for privatized prisons. Currently, 10% of all prisoners in the US are sent to
7


privately owned prison (Dumont et al. 2013). With the number of people being incarcerated, an opportunity for the privatization of prisons was created leading to a mutually beneficial arrangement between government and big business. In addition to the need for private prisons for housing inmates, many state and federal prisons contract out services such as health care, construction, and food services to private companies generating market rewards for major corporations and Wall Street investors (Palaez 2008). These incentives for profit generate backing for politicians and help to support political campaigns that offer continued support for the implementation of tough on crime policies in the United States that continue to generate profit for the investors of privatized correctional services (Palaez 2008). Efforts to reduce crime and protect society from crime have resulted in major changes to the way the justice system punishes criminal engagement and has a wide variety of consequences to our society, with an especially consequential effect on the well being and mental health of inmates (Massoglia 2008b; Nowotny et al. 2015; James and Glaze 2006).
In the U.S., many prisons house more mentally ill individuals than inpatient psychiatric facilities (Kim, Becker-Cohen and Serakos 2015), over half of inmates reporting some type of mental disorder (James and Glaze 2006). Prior literature has demonstrated a strong association between incarceration and mental health disorders in inmate populations (Sered 2008; Massoglia 2008b). Prisoners suffering from mental illness are more prone to commit violent offenses, more likely to recidivate, and more likely to experience increased risk factors associated with a cyclical process of incarceration and release (Kim, Becker-Cohen and Serakos 2015). Currently incarcerated inmates are also highly likely to be undertreated and under diagnosed for mental health disorders while incarcerated (Schnittker, Massoglia and Uggen 2012; Baillargeon, Hoge and Penn 2010). With an estimated 7.5% of
8


the population, about 16 million people, being felons or ex-felons, the impact of mental health disorders associated with exposure to incarceration has great consequences not only to those incarcerated but to the general population as 95% of inmates are eventually released (Nowotny 2016; Schnittker, Massoglia and Uggen 2012). Understanding mental health treatment while incarcerated, therefore, is essential to the successful reentry of inmates to their communities as well as understanding the relationship between incarceration and mental health.
Incarceration and Mental Health
The high prevalence of mental health disorders present in incarcerated populations highlights the role prisons and correctional facilities play in the mental health treatment needs of inmates. Many factors connected with the prison environment have been associated with the production or reproduction of poor mental health outcomes among inmates (Massoglia 2008b), while prisons simultaneously provide needed care for many individuals with unmet or undiagnosed mental health treatment need outside of the prison environment (Turney, Wildeman and Schnittker 2012). As prisons disproportionately house those of limited financial, social, and human capital, these surroundings generate and maintain environments of disadvantage (Wakefeild and Uggen 2010). Despite the access to health care services in prisons, health benefits associated with institutionalized treatment do not persist post-release (Schnittker and John 2007) leading to a high proportion of inmates and exinmates with untreated mental health disorders.
Currently incarcerated inmates have significantly higher rates of mental health disorders compared to non-institutionalized persons. Over half of all inmates will receive treatment for mental health disorders while incarcerated (James and Glaze 2006). Although
9


there is a high proportion of individuals who have been treated for a mental health disorder prior to incarceration, many individuals with no history of mental health problems who have been exposed to the prison environment will seek treatment for new mental health disorders once incarcerated (Massoglia and Pridemore 2015; Schnittker and John 2007). Research has shown that individuals who recidivate are more likely to use the health services that are provided by correctional facilities, yet despite receiving treatment while incarcerated, individuals with a history of incarceration still report worse health overall than the general population (Schnittker and John 2007). These findings suggest that incarceration relates to both shaping ones risk factors for mental health pathologies, as well as influencing access to mental health treatment services.
Scholarship falling at the intersection of criminology and medical sociology has recently begun to focus on deconstructing the mechanisms between incarceration and mental health. Incarceration is known to have direct consequences on the mental health of inmates and increased mental health treatment needs in incarcerated populations. While research has shown that incarceration may lead to some temporary improvements in physical health, especially among minority populations (Houle 2014; Baily et al. 2015b), the opposite has been found for the mental health of prisoners (Turney, Wildeman and Schnittker 2012). Being incarcerated is associated with sudden changes in many areas of ones life and is coupled with a major shock to ones lifestyle, which can lead to a variety of chronic stressors (Massoglia and Pridemore 2015; Turney, Wildeman, and Schnittker 2012). Prior literature (Schnittker and John 2007; Massoglia 2008b; Schnittker, Massoglia and Uggen 2012) has shown that being exposed to the prison environment results in primary and secondary stressors both in daily life and in the long term. Such stressors inhibit positive coping
10


strategies and are associated with a decline of the mental health of inmates (Kim 2014; Massoglia 2008b). The prison environment often exacerbates mental health disorders in individuals with a history of mental illness as well as engendering new mental health disorders in previously asymptomatic individuals (Massoglia and Pridemore 2015; Nowatny et al. 2015; Kim 2014; Williams et al. 2010) leading lead to an increased need for mental health treatment in prisons.
Mental health disorders among inmates have been shown to have many long lasting consequences for the overall health of inmates both during and after incarceration. Some difficulties experienced by inmates with mental health disorders post incarceration include increased risk of homelessness and housing discrimination (Baillargeon, Hoge and Penn 2010; Schnittker, Massoglia and Uggen 2012), increased likelihood of recidivism (Lamb, Wienberger and Gross 2004), limited access to health care, (Lamberti and Wiseman 2004) difficulty gaining employment or benefits (Baillargeon Hoge and Penn 2010), as well as stigmatization across multiple domains. Increased risk factors such as these not only complicate mental well-being while incarcerated, but also contribute to chronic stress throughout the life course (Massoglia & Pridemore 2015; Massoglia 2008b) through durable pathways such as endured stigma, loss of economic or social roles post release or impaired health status (Turney, Wildeman, and Schnittker 2012). Chronic stress can lead to relapse in mental health disorders as well as inhibit the bodys essential functions including cell-mediated immunity (Glaser and Kiecolt-Glaser 2005; Massoglia and Pridemore 2015) and cardiometabolic health (Thoits 2010), which may lead to additional health complications following release from prison. Moreover, the more risk factors ex-inmates experience after being released from prison, the higher their likelihood of returning to prison perpetuating a
11


cyclical process of recidivism and release (Binswanger, Nowels. Corsi, Long, Booth, Kutner and Stiener 2011b; Baillargeon, Binswanger, Penn, Williams and Murray 2009).
Duration of Incarceration: Current And Repeated Incarcerations An implicit supposition in research on incarceration and the mental well-being of prisoners is that the accumulation of exposure to prison-based stressors will confer poorer mental health outcomes for detainees with lengthy incarceration spells. Although some studies have used inclusive measures to assess the cumulative duration of incarceration by examining the number of priors and incarceration status, results have been mixed (Nagin, Cullen and Johnson 2009). Previous research examining cumulative effects of incarceration have failed to examine the joint effect of duration of time served and number of prior incarcerations on mental health disorders in currently incarcerated populations, and the treatment needs of inmates with diminished mental health.
In a study conducted by Bailey et al. (2015), duration of incarceration was operationalized using a self-report of number of prior incarcerations and total lifetime duration of incarceration. This study found no significant association between total duration of incarceration and health behaviors while incarcerated, but did find a significant association between the number of prior incarcerations and current health behaviors post-release. In another study conducted by Schnittker and John (2007), cumulative duration of incarceration was analyzed as a primary predictor for severe health limitations. Incarceration was captured using separate measures of number of prior incarcerations and current incarceration status. Results from this study were mixed; current incarceration was negatively associated with functional health limitations, but history of incarceration was positively linked to such limitations. These patterns were not moderated by race or education levels. This study did not
12


evaluate whether current incarceration status moderated the relationship between prior incarceration and functional limitations.
These approaches may be problematic due to the association of processes that incarceration and health may share, suggesting that these processes may not be independent, but //7/tv'dependent. Therefore, more research is needed to assess the joint contribution of both duration of time served and number of priors on the mental health of prisoners and mental health treatment in prison as well as prior to ones incarceration. Two complementary theoretical frameworks offer insight that may be useful in understanding the association between cumulative duration of incarceration and mental health treatment outcomes in incarcerated populations.
Stress, Coping, and Institutional Acclimation
Two theoretical scenarios, Stress Process Theory and the Theory of Total Institutions, offer support linking the total cumulative duration of incarceration and mental health disorder in incarcerated populations. Both these theories support the views of cumulative advantage/disadvantage (CAD) theory. CAD theory posits that both advantage and disadvantage accumulate through the life course based on systematic exposure to available resources both within individual societies and within individuals (Dannefer 2003).
Cumulative dis/advantage is a complex structural process of socialization and resource allocation that results in measurable differences in access to resources, health outcomes, and inequality for different social groups. Advantages and disadvantages accumulate throughout the life course over multiple pathways including education, economic resources, and access to health care. Accumulated disadvantage in the form of repeated and/or lengthy
13


incarceration spells has several consequences in relation to exposure to the prison environment.
Stress Process Theory
Stress process theory (SPT) is a foundational theory in medical sociology, and holds that human stress occurs as a result between a mismatch between the demands placed upon a person and the resources they have to meet those demands. Specifically, SPT suggests that cumulative exposures to stressful environments (i.e. prison) disproportionally disadvantage individuals due to structural systems of stratification that unevenly expose them to stress while simultaneously straining their resources to cope with taxing demands, a confluence of challenges that may result in health-related inequalities (Pearlin 1989; Thoits 2010).
Pearlin (1989) describes how life events requiring major adjustments may be potentially harmful, especially in undesirable situations. Stressful events generally occur through multifaceted and interrelated environmental and structural factors such as social stratification, interpersonal relationships, and social institutions (Turney, Wildeman and Schnittker 2012). People who share similar life experiences commonly experience similar stressors, and patterns of stress distribution are often rooted in social and economic situations. SPT discusses the many components of stress and relates stressors back to chronic strain in ones life, often resulting from an initial event or situation. Moreover, stressors occur in clusters, are rarely independent of other parts of ones life, and accumulate throughout the life course (Massoglia and Pridemore 2015; Thoits 2010).
SPT proposes that there are both primary and secondary stressors associated with chronic strain. Primary stressors are the result of major or non-normative change in ones life. Once a change has taken place, additional secondary stressors generally occur. These
14


stressors are often associated with social, economic, and environmental circumstances. For instance, in a non-institutionalized population, a primary stressor may constitute job loss while a secondary stressor may materialize in the forms of relationship instability, food insecurity, or homelessness. Incarceration is an example of an undesirable event and is usually a major change for any individual getting incarcerated; this would be considered the primary stressor (Massoglia and Pridemore 2015). However, one can easily see that there are several secondary stressors associated with the realities of living in the prison environment such as fear of violence, loss of autonomy, social isolation and economic hardship (Massoglia 2008b; Turney, Wildeman and Schnittker 2012).
Pearlin goes on to explain that institutional roles can increase stressors due to persistent and repeated features of an institution. Challenges associated with institutional stressors include interpersonal conflict, repetitive activities, and role captivity (Pearlin 1989). Role captivity exists when one is unwillingly stuck in an undesirable role, as is most often the case with inmates. In addition to stressors specifically associated with an institutionalized setting, many experience ambient strains associated with the burden of becoming incarcerated. Ambient strains include physical and environmental strains such as a serious or chronic illness or fear of vehemence or discrimination by staff or other inmates. Moreover, all strains and stressors contain multiple sub-types of stressors, creating persistent chronic strain, especially within the prison environment.
In order to manage such stressful situations, individuals must learn to enact coping mechanisms. Coping is the act of attempting to avoid or lessen the consequences associated with the impact of life problems and stressors (Pearlin 1989). Some ways to implement coping mechanisms include trying to change the situation or escape, trying to manage the
15


meaning of the situation, or to find ways to manage the symptoms of the stress through behavioral and emotional coping (Leban, Cardwell, Copes and Brezina 2016). In incarcerated populations, this may be challenging as often times, there is little an inmate can do to change the situation and resources to lessen the impact of stressful situations may be in short supply. Therefore, the methods of coping one can engage in tend to be more limited than during other facets of life. Coping mechanisms can be positive or negative, and they can be healthy or unhealthy. Some common coping mechanisms used in prisons include seeking out social support or help, cognitive coping such as minimizing the importance or ignoring the stressor, accepting responsibility for ones actions, attempting to escape the situation, emotional coping such as seeking time alone to draw or read, or behavioral coping such as retaliation (Leban, Cardwell, Copes and Brezina 2016).
Despite attempts to cope with chronic strain and stressors there can be a substantially detrimental effect on health functioning. In the prison environment, multidimensional stress can lead to increased stigma, poor physical health, and poor coping strategies all contributing to diminished mental health in incarcerated populations (Massoglia 2008a; Nowotny et al. 2015; Massoglia and Pridemore 2015; Porter 2015; Sered and Norton-Hawk 2008). By implementing coping mechanisms through utilizing available resources, inmates may be able to better adapt to the adversities of the prison environment.
Total Institutions Theory
Another theory that resonates with the perspectives of CAD theory is Goffmans theory of Total Institutions (TI) (Goffman 1961). In Goffmans discussion of total institutions, he describes total institutions as authoritarian organizations of corrective sanction in which inmate behavior is regularly judged and restricted (Scott 2010). The
16


authoritarian environment of total institutions results in the simultaneous depersonalization (i.e. elimination of all personal items upon entry) and standardization (i.e. the use of uniforms and provision of necessary material) of individuals. Personal identity is limited and provisions are provided for all equally and lack uniqueness. Total institutions are built on a system of house rules and there are severe consequences for disobedience. These rules provide punishments and small rewards for good or bad behaviors while in the institution (Goffman 1961). Due to this challenging and incapacitating environment an individuals autonomy is highly restricted and one is stripped of their ability to regulate the self.
As total institutions are severely restrictive, Goffman offers an explanation of coping mechanisms deployed in order to manage the stress of such environments. TI suggests that cumulative exposure to an institutionalized environment creates uniquely specialized skills that enable stronger coping methods allowing individuals to better handle the demands of the prison environment (Schnittker, Massoglia and Uggen 2012). TI theory tells us that repeated and lengthy exposure to incarceration aids inmates in learning to utilize valued resources within the prison environment through adaptations and adjustments. This adaptation generally involves skills such as developing an institutional language that is used to describe the inmate world, an accumulation of specific institutional knowledge such as knowledge about the institution, various officials, and comparative information about other similar institutions (through prior incarcerations), the implementation of secondary adjustments, building social controls at the inmate level, bond formation, and creating barriers to the psychological stressors associated in the prison environment (Goffman 1961).
A primary tool used to develop unique coping methods and reduce psychological stressors relates to the use and implementation of secondary adjustments, which are
17


practices that may allow an inmate to obtain forbidden pleasures without directly challenging the prison staff. Some examples include procuring prohibited resources such as homemade alcohol or cigarettes or engaging in discouraged activities such as gambling or sexual acts (Goffman 1961). These activities offer an inmate a level of personal autonomy that is often absent in prison environment (Scott 2010). Moreover, TI theory posits that inmates who have had lengthy or repeated incarcerations are more likely to be able to use these adaptive techniques to perfect their coping mechanisms while incarcerated, especially with repeated exposures (Munn 2011). Implementing such coping methods while incarcerated may help to reduce consequences on mental health during ones current incarceration.
Current Study
While prior research has extensively examined mental health consequences of incarceration, not all extant research in this area simultaneously attends to prior incarceration history and the length of ones current incarceration. Indeed, much of this research is based on general population samples, constrained to small subsamples with some prior incarceration history. Further, no research to the authors knowledge has investigated both the independent and joint effects of prior incarcerations, length of current incarceration, and mental health treatment outcomes among currently incarcerated persons. Given these research gaps, the current study uses Stress Process and Total Institutions theories, multinomial logistic regression techniques and data from the Survey of Inmates in State Correctional Facilities to address the following research aims: (1) assess the relationship between duration of time served and the mental health treatment needs of currently incarcerated inmates, (2) evaluate how repeated incarcerations contributes to the mental health treatment needs of currently incarcerated inmates, and (3) examine the joint effect of
18


both number of prior incarcerations and duration of time served in ones current sentence on the mental health treatment of currently incarnated inmates to evaluate interaction effects of these two measures. By exploring the treatment seeking behaviors of inmates this study will be able to evaluate how sentence length and number of prior incarcerations may contribute to managing the stressors associated with the harsh prison environment, allowing us to gain a stronger understanding of the consequences of incarceration.
19


CHAPTER III
DATA AND METHODS
2004 Survey of Inmates in State Correctional Facilities
This study uses nationally representative cross sectional data from the 2004 Survey of Inmates in State Correctional Facilities (SISCF). The SISCF was designed by the Bureau of Justice Statistics and measures extensive demographic and socioeconomic characteristics as well as criminal history backgrounds, drug and weapon use, health of prisoners, and military service. The Bureau of the Census collected data for this survey between October 2003 and May 2004. A two stage sampling procedure was used to collect the data. First, individual prisons were selected to participate and then from within each prison, individual inmates were selected from within selected prisons. Inmates were interviewed by computer-assisted personal interviews and telephone interviews and had an 89.1% response rate. This study uses public use data, which were downloaded from the ICPSR website on September 2,
2016. These data have 14,499 observations in the public use data set. List-wise deletion was used to eliminate data with missing responses on variables of interest resulting in a total sample of N=10,907.
Dependent Variables
Mental Health Treatment
This study focuses on pre-incarceration and current mental health treatment of prisoners. Two variables measuring history of mental health treatment in the 12 months prior to the current incarceration and treatment during current incarceration were used to create a four-category dependent variable indicating each of the following: has not received mental health treatment while incarcerated and has no history of mental health treatment (=0); has
20


not received treatment but has history of mental health treatment (=1); has received treatment but has no history of mental health treatment (=2); and has received treatment and has history of mental health treatment (=3) (see Figure 1 for a descriptive picture of the prevalence of each of these categories in the analytic sample). For parsimony, I refer to these categories as having never received treatment, has unmet treatment needs, has new treatment needs, and has consistently met received treatment.
Independent Variables
Duration of Incarceration
The main independent variables are the main effects and interaction terms of duration of time served in ones current sentence and number of prior incarcerations. Using a data driven method, duration of time served was broken into non-linear categorical measures based on tertile categories of short, medium and long duration of time served. Short sentence duration (=0) is defined as 0-19 months of time served. Medium duration (=1) of time served is 20-54 months served, and long duration (=2) of time served is 55-523 months served. Number of prior incarceration is a dichotomous variable divided between those with no prior incarcerations and those with one or more prior incarcerations. Interaction terms were generated for all interactions between time served and prior incarcerations.
Covariates
Other covariates in this analysis included age (18 96), gender (female=0, male=l), race (White =0, Black =1, Hispanic =2, Other =3), income the month prior to incarceration measured in dollars ($0 $8,750), high school graduation status (did not graduate =0, graduated high school =1), and sentence length in months (0- 523).
21


Analyses
This study first estimates both descriptive statistics and then more sophisticated multivariate multinomial logistic regression models using Stata 13 to analyze the data. List-wise deletion was used to omit respondents who had missing data on variables of interest. Table 1 presents descriptive characteristics of the entire sample (N=10,907). Using multinomial logistic regression analyses, Model 1 examines the odds of receiving mental health treatment using the four-category typology constructed from two measures of current and prior mental health treatment in a currently incarcerated population. Model 1 uses both duration of time served in ones current sentence and number or prior incarcerations as primary predictors for mental health treatment during current incarceration. Multinomial logistic regression models allows for between group comparisons based on the four-category typology of mental health treatment status in currently incarcerated inmate populations.
Model 1 uses two reference groups, never treated and met need, to examine group differences in the odds of receiving mental health treatment while currently incarcerated. Model 2 includes interactions between number of prior incarcerations and duration of time served in ones current sentence. All models include demographic and socioeconomic control variables. All tables denote statistically significant differences (p < .05, p < .01 ,p <001).
22


CHAPTER IV
RESULTS
Descriptive Statistics
Table 1 presents descriptive statistics of the entire sample. Among all sample participants the majority of individuals incarcerated are male. The mean age of the sample is 36. The average sentence length across all inmates is 55 months, or almost four and a half years. Nearly three quarters of inmates currently incarcerated in state prisons have had at least one prior incarceration (73%). Those who have had a short duration (0-19 months) of time served make up 34% of the sample, those with a medium duration (20-54 months) of time served represent 32% of the sample, and those with a long duration (55-523 months) of time served are 33% of the sample.
Table 1. Sample Description, Survey of Inmates In State Correctional Facilities (N= =10,907)
Variable Mean/Proportion SE
Mental Health Treatment
Never Treated .74 .004
Met Need .10 .003
Unmet Need .04 .002
Newly Treated .11 .003
Prior Incarcerations .73 .004
Time Served
Short (reference) .34 .005
Medium .32 .005
Long .33 .005
Age 36 .100
Male .80 .004
Race
White (reference) .39 .005
Black .38 .005
Hispanic .17 .004
Other .06 .002
Education 10.91 .022
Graduated .84 .003
Monthly Income 2,038 21.5
Sentence Length in Months 902 25.4
23


Of the entire sample, 74% of inmates have never received mental health treatment prior to incarceration nor are receiving treatment for mental health disorders during their current incarceration. This group represents the never treated group. Those who were previously receiving mental health treatment prior to their current incarceration and are still receiving mental health treatment while incarcerated represent 10% of the sample and represent met need. Inmates who were previously receiving mental health treatment prior to their current incarceration but are no longer receiving mental health treatment represent 4% of the sample. Such inmates represent unmet need. Inmates who are receiving new treatment with no history of mental health treatment equal 11% of the sample. These inmates are the new treatment group. Figure 1 illustrates the percentage of inmates in each typology of mental health treatment history and current mental health treatment in the entire sample.
Figure 1. Typology and Prevelance of Mental Health Treatment Among Currently Incarcerated Inmates in State Prisons (N=10,907)
No History of Treatment History of Treatment
24


Table 2. Multinomial Logistic Models Predicting the Odds of Receiving Mental Health Treatment (N= 1 0.907)e

Model 1 Model 2
Met Need Unmet Need New Treatment Met Need Unmet Need New Treatmen
OR (95% Cl) OR (95% Cl) OR (95% Cl) OR (95% Cl) OR (95% Cl) OR (95% Cl)
Age 1.01** 0.99* q 99*** 1.00** 0.99* 0 99***
(1.00,1.01) (.98,1.00) (.98, .99) (1.00,1.01) (.98,1.00) (.98, .99)
Male 0.26*** q 44*** 0.33*** 0.26*** q 43*** 0.33***
(.23, .30) (.35, .55) (.29, .39) (.29, .30) (.35, .55) (.29, .39)
Race (reference White)
Black q 27*** Q 4]*** 0.50*** q 37*** q 4Q*** 0.5***
(.32, .43) (.32, .51) (.43, .57) (.32, .43) (.32, .52) (.43, .57)
Hispanic 0.36*** 0.36*** 49*** 0.36*** 0 40*** 0 49***
(.29, .44) (.26, .50) (.40, .59) (.29, .44) (.32, .51) (.40, .59)
Other 0.89 0.54** 0.80 0.88 0.54** 0.80
(.69,1.13) (.34, .85) (.61,1.03) (.69,1.13) (.34, .86) (.61,1.03)
Graduated 74*** 0.68** q 74*** 0.68** 0.78**
(.62, .88) (.53, .88) (.66, .93) (.62, .87) (.59, .88) (.66, .93)
Monthly Income 95*** 0.96* .97* 0.95*** 0.96* 0 97***
(.93, .97) (.93, .99) (.95, .99) (.93, .97) (.93, .99) (.95, .98)
Sentence Length in Months 1.00 1.00 1.00* 1.00 1.00 1.00*
(1.00,1.00) (1.00,1.00) (1.00,1.00) (1.00,1.00) (1.0, 1.0) (1.00,1.00)
Prior Incarcerations 1 31*** 1.19 1.30*** 1.52** 1.23 1.61**
(1.12,1.52) (.95,1.50) (1.12-1.51) (1.17,1.97) (.88,1.71) (1.17,2.2)
Time Served (reference short)
Medium j 4j*** 0.77* 2 Q9*** j 71*** 1.01 2 4***
(1.21,1.65) (.61, .97) (1.76,2.48) (1.26,2.34) (.65,1.55) (1.66,3.45)
Long 1.20* 0.67** 3.38*** 1.38* 0.48** 4 29***
(1.01,1.44) (.50, .87) (2.83,4.03) (1.00,1.92) (.28, .84) (3.05,6.02)
PriorXMedium 0.77 0.69 0.84
(.54,1.10) (.41,1.14) (.56,1.26)
PriorXLong 0.83 1.51 0.73
(.57,1.19) (.83,1.76) (.50,1.06)
Constant .55*** 0.45** 4g*** q 49*** 0.44** Q 4]***
(.38, .78) (.26,0.77) (.33, .69) (.33, .72) (.25, .77) (.27, .67)
Confidence intervals in parenthesis. p<05,** p<01, *** p<001 a Omitted reference group is never received treatment
25


Multinomial Regression
Table 2 presents results from a multinomial model comparing the odds of having (1) unmet mental health treatment needs, (2) newly-treated mental health needs, and (3) consistently treated mental health needs, relative to (0) having never received mental health treatment, as a function of time served, prior incarcerations, and controls. Model 1 demonstrates the main effects of duration of incarceration and prior incarcerations on the mental health treatment status of currently incarcerated inmates. Results show that duration of time served is positively and significantly associated with mental health treatment. For inmates who have served a medium vs. short sentence, the odds of having new treatment is 2.09 times the odds of never being treated (95% confidence interval [Cl] = 1.76, 2.48). Those with a long vs. short duration of time served have 3.38 the odds of having new treatment vs. having never been treated (95% Cl = 2.83, 4.03). Inmates who have prior incarcerations have 1.30 times higher odds of having new treatment compared to inmates who have never received treatment (95 % Cl = 1.12, 1.51).
Model 1 shows that inmates who served a medium vs. a short period of incarceration are more likely to have met need compared to being never treated. For inmates who have served a medium vs. short sentence, the odds of having met need is 1.41 times the odds of never being treated for any mental health disorder (95% Cl = 1.21, 1.65). Those who have served a long vs. a short sentence have 1.20 times higher odds of having met need compared to never receiving mental health treatment 95% Cl = 1.01, 1.44). Inmates who have a prior incarceration have 1.31 times higher odds of having met need compared to never being treated (95% Cl = 1.12, 1.52). Model 1 shows that for those with a medium duration of time served vs. a short duration of time served the odds of having unmet need compared to never
26


Table 3. Multinomial Logistic Regression Models Predicting Mental Health Treatment Outcomes vs. Having Met Treatment Needs (N=10,907)a_______
Model 1 Model 2
Unmet Need New Treatment Unmet Need New Treatment
OR OR OR OR
(95% Cl) (95% Cl) (95% Cl) (95% Cl)
Age Q 9g*** Q 9g*** Q C)g*** Q C)g***
(.97, .99) (.98, .99) (.97, .99) (.97, .99)
Male ^ gy*** 1.26** 1.66*** 1.27*
Race (reference White) (1.30,2.14) (1.05, 1.51) (1.30,2.14) (1.06, 1.52)
Black 1.10 1.34** 1.09 1.35**
(.84, 1.43) (1.11, 1.63) (.84, 1.42) (1.11,1.52)
Hispanic 1.01 1.37* 1.01 1.37*
(.70, 1.47) (1.05, 1.52) (.69, 1.47) (1.05, 1.63)
Other 0.61 0.91 0.61* 0.90
(.37, 1.00) (.65, 1.25) (.37, 1.01) (1.05, 1.80)
Graduated 0.92 1.06 0.92 1.06
(.69, 1.23) (.85, 1.32) (.69, 1.23) (.85, 1.32)
Monthly Income 1.02 1.02 1.01 1.00
(.98, 1.05) (1.00, 1.05) (.98, 1.05) (.99, 1.05)
Sentence Length in Months 1.00 1.00** 1.00 1.00**
(1.00, 1.00) (1.00, 1.00) (1.00, 1.00) (1.00, 1.00)
Prior Incarcerations 0.91 1.00 0.81 1.06
Time Served (reference short) (.70, 1.18) (.82, 1.21) (.55, 1.20) (.72, 1.56)
Medium 0.54*** \ 4g*** 0.59* 1.40
(.42, .71) (1.12, 1.82) (.35, .97) (.90, 1.1)7
Long 0.55*** 2.81*** 0.35*** 3 Q9***
PriorXMedium PriorXLong (.40, .76) (2.23, 3.54) (.19, .64) 0.89 (.49, 1.60) 1.83 (.93, 3.60) (2.00, 4.77) 1.08 (.66, 1.79) 0.88 (.54, 1.43)
Constant 0.82 0.88 0.89 0.84
(.45, 1.51) (.56, 1.39) (.47, 1.70) (.50, 1.42)
Confidence intervals in parenthesis p <.05,** p<01, *** p< 001 a Omitted reference group is met need
27


receiving treatment are .77 times lower (95% Cl = .61, .97). For inmates who have served a long sentence vs. a short sentence there is .67 times the odds of having unmet need compared to never receiving treatment (95% Cl = .50, .87).
Model 2 shows no significant effect of the interaction between number of prior incarcerations and duration of time served, suggesting that duration of time served and number of priors have only significant independent effects.
Table 3 presents results from a multinomial model comparing the odds of having (1) unmet mental health treatment needs, (2) newly-treated mental health needs, relative to (3) consistently treated mental health needs, as a function of time served, prior incarcerations, and controls. Model 1 demonstrates the main effects of duration of incarceration and prior incarcerations. These results show that inmates are more likely to have met need compared to unmet need. The odds of having met need compared to having unmet need are .54 times lower for inmates serving a medium sentence vs. a short sentence (95% Cl = .42, .71). For inmates with a long vs. short duration of time served, the odds of having met need compare to unmet need are .55 times lower (95% Cl = .40, .76).
Results in Table 3 show that duration of time served is significantly associated with mental health treatment inmate populations in inmates with new treatment compared to those with met need. Inmates with who have served a medium sentence vs. a short sentence have 1.48 times the odds of being newly treated compared to having met need (95% Cl = 1.12, 1.82). For those inmates who have served long sentences vs. short sentences, the odds are 2.81 times higher for receiving new treatment group compared to having met need (95% Cl = 2.23, 3.54).
28


Model 2 in Table 3 includes the interaction between time served and number of prior incarcerations. There are no significant effects of the interactions between duration of time served and prior incarcerations in this model.
Figure 2 summarizes the findings from Tables 2 and 3 in the form of fitted predicted probabilities of membership in each of the mental health treatment categories by duration of incarceration. The fitted probabilities are adjusted for the mean effect of all controls in the fully-adjusted models. Results indicate that the probability of being in the never treated category was highest for those with short durations of incarceration (82%), followed by those with medium durations of incarceration (75%), and lowest among those with the longest intervals of incarceration (72%).
Figure 2. Predicted Probability of Mental Health Treatment Catagories by Duration of Incarceration
Notes: adjusted for age, sex, race, income, education, and sentence length
For those inmates in the met need category, results demonstrate the probabliltity for being in this category were highest for those with medium duration of incarceration (10%). For those who had served a short sentence duariton or a long sentence duration the probablility of being in the met need category decreased to 8%. The proabablity of being in
29


the unmet need category decreased over time. Results show that for inmates who have a short duration of incarceration the probablitly of being in the unmet need category is highest (5%), but as time served increases the proability of being in the unmet need category decreses. For those who have served a meduim or long sentence the probablity of being in this category decreased by almost half (3%).
Results specify that the proablility of being in the new treatment category increased as duration of time served increases. The probability of being in this category is lowest for those who have served a short sentence (6%), followed by those with medium duratinon of incarcertion (11%) and is highest among inmates who have served a long sentence duration (16%).
30


CHAPTER V
DISCUSSION
Recent studies have shown that time spent incarcerated is correlated with declines in the mental health of inmates (Kim, Beck-Cohen and Remster 2012; Lamb, Weinberger and Gross 2014). Research has shown that inmates have a much higher prevalence of mental health disorder compared to the general population. While incarcerated, more than half of inmates will seek some type of treatment for a mental health disorder (James and Glaze 2006), regardless of prior history of mental disorder. Several consequences related to the prison environment have been associated with poor mental health outcomes of inmates such as increased stress, environmental risk factors, and post release consequences (Massoglia 2008b, Schnittker and John 2007). With over 2 million inmates currently incarcerated, understanding the association between mental health and duration of time served is important for both scholars and policymakers.
Although much research has examined the strains associated with being incarcerated (Binswanger, Kruger and Steiner 2009; Turney, Wildeman and Schnittker 2012; Kim, Beck-Cohen and Remster 2012), few studies have examined the effects of both number of prior incarcerations and duration of time served in ones current sentence to assess the consequences of duration of incarceration on the mental health of currently incarcerated inmates, limiting our understanding of how duration of incarceration is associated with receiving treatment for new and previously diagnosed mental health disorders in currently incarcerated inmates. Moreover, studies on the mental health of inmates often use a subsample from general population data to assess the hazards of incarceration making it difficult to tease out differences within the inmate population.
31


The current study examined these gaps in the literature by using multinomial logistic regression to analyze the odds of receiving new or continued treatment, and discontinuing treatment or never having received treatment, for mental health disorders while incarcerated as a function of the independent and joint effects of duration of current incarceration and prior incarcerations. Data from the 2004 National Survey of Inmates in State Correctional Facilities were used to assess mental health treatment patterns in currently incarcerated populations. The four-category typology considered here includes inmates who have never been treated, inmates with met need, inmates with unmet need, and inmates who are currently receiving new treatment. Using this typology I was able to address the flowing research aims: (1) assess the relationship between duration of time served and mental health outcomes, (2) examine the association between prior incarcerations and mental health treatment, and (3) evaluate the joint effect of number of priors and duration of time served.
Duration of time served was significant across all treatment typologies. An interesting finding showed that prior incarcerations were only significant when comparing inmates with met need and new treatment to inmates who never received treatment. A history of prior incarcerations therefore, was positively associated with receiving mental health treatment while incarcerated in subsequent incarcerations. An unexpected result demonstrated that in no case was the interaction between number of prior incarcerations and duration of time served significant. However, this finding shows that although there are significant main effects for both variables, these measures are independent and not inter-dependent.
The four-category treatment typology included two origin patterns: those with a history of treatment prior to incarceration and those with no history of incarceration. The difference in origin correlates to distinct patterns in expected treatment of inmates based on
32


duration of time served. Inmates with met and unmet need both share a common history of mental health treatment prior to their current incarceration. Inmates with met need demonstrate continuity of treatment whereas unmet need shows interrupted treatment suggesting that these inmates have different selection pathways once incarcerated. This study finds that inmates who have a history of mental health treatment are more likely to have met need compared to having unmet need the longer incarcerated.
Consistent with Total Institutions theory, this finding suggests that inmates who have a history of mental health treatment and who spend more time incarcerated are more likely to have their mental health treatment needs met. This may be due to the accumulation of institutional knowledge gained through prolonged exposure to the prison environment. Total Institutions theory posits that repeated or extended lengths of time spent in a total institution reduces ones autonomy leading to utilizing coping mechanisms through the use of adaptation and institutional knowledge (Goffman 1961). Inmates with who are incarcerated for longer durations are more likely to develop specialized skills such as learning how to access and take advantage of services provided within the prison, including access to mental health care treatments. Additionally, the results in this analysis demonstrate that individuals with a history of multiple incarcerations are more likely to access mental health treatment compared to individuals who are serving their first sentence, which shows that those who have repeated institutional exposures may be more capable or willing to use prison resources. Moreover, the results show that longer duration of time served is associated with met need compared to unmet need suggesting that inmates who are incarcerated for greater lengths develop coping mechanisms not limited to seeking treatment, but perhaps also relating to personal autonomy helping to reduce stressors associated with the prison environment.
33


The second origin pattern illuminated in the current study examines inmates who have no history of mental health treatment prior to their current incarceration. Results from the analysis show that the longer inmates are incarcerated the more likely they are to receive new treatment and the less likely they are to have been never treated. This pattern suggests that factors associated with the prison environment contribute to increased mental health anguish the longer one spends incarcerated, as well as with repeated exposures through multiple incarcerations. In accordance with stress process theory, major changes within ones life are associated with increased stress and difficulty coping (Pearlin 1989). Being incarcerated is a major shock to ones life, and the longer duration of incarceration or the more times one is incarcerated, the more challenging one may find coping with such a change. These findings support stress process theory and suggest that incarceration is highly correlated with increased stressors and chronic strain, as well as the manifestation of new or under-diagnosed mental health disorders that are associated with time spent incarcerated.
Inmates who have never received treatment for a mental health disorder prior to or during their current incarceration are most likely to have served a short sentence compared to inmates who have received new treatment. It is possible that inmates who serve short sentences are less likely to seek treatment simply due to the limited time spent incarcerated. Moreover, it is possible that inmates with shorter sentence duration lack the extended exposure to the stressors of the prison environment and therefore may not suffer the same mental health consequences as those who are incarcerated for longer durations. It is also plausible that those who serve longer sentences experience different or cumulative stressors compared to inmates who served short sentences resulting in a greater need for mental health treatment while incarcerated.
34


Strengths and Limitations
This study has several strengths. The first strength of this study is the large nationally representative data (N= 10,907), which provides enough power to detect subtle patterns and differences among mental health outcomes in incarcerated populations. In order to provide an objective measure of model fit and parsimony, this study used AIC scores to select the bestfitting model for the data. Second, this study uses a stratified sample of only currently incarcerated inmates instead of using a control sample from the general public, which reduces selection bias in estimating the consequences of incarceration on mental health treatment. Comparing only inmates significantly reduces selection bias within the sample, and allows us to infer pathways between incarceration and mental health treatment in inmate populations. Third, this is one of few studies to examine actual duration of time served using data driven percentile ranks (i.e. tertiles) as a predictor for mental health treatment in incarcerated populations. Finally, the current study considers an outcome that includes categories for pre-incarceration treatment experiences as well as experiences during ones current incarceration. This innovation provides more detail on the history and current status of ones interaction with mental health professionals, and better identifies those positively selected into prison based on prior mental health problems from those who develop new treatment needs while incarcerated.
This study has several noteworthy limitations. Currently, there is no perfect measure for total cumulative duration of incarceration over the life-course. While this research generated a new approach to assessing cumulative duration of incarceration by examining both number of prior incarcerations and duration of time served, future studies should produce more precise measures of total cumulative duration of incarceration over the life
35


course. Improving this measure would more accurately capture the consequences of the cumulative exposure of incarceration allowing for a greater understanding of the specific mechanisms contributing to poor mental health in incarcerated populations. A second limitation of this study is that it used cross sectional data, which limits the observations to a single observation for each participant and makes determining causality difficult. However, by using a stratified model to help adjust for bias based on prior mental health treatment the study was able to partially attenuate for this limitation. Using longitudinal data would increase the ability to examine changes in the mental health within individuals over time allowing for a better understanding of precise pathways to mental health outcomes among inmates. A third limitation to this study was that based on the results, it is difficult to make claims regarding the primary motivation for mental health treatment seeking among inmates. Both increased stressors and acclimation of the institutional environment may contribute to treatment seeking in currently incarcerated populations.
Conclusion
This research provides clear evidence of the consequences to mental health based on duration of incarceration using a typology of mental health status in currently imprisoned populations. However, more research is needed to explore how cumulative duration of incarceration over the life course is associated with mental health outcomes in inmate populations and what factors contribute to seeking mental health treatment while imprisoned. Future studies should work to improve the measure of cumulative lifetime duration of incarceration allowing for greater causal inference of the specific pathways leading to poor mental health among inmates while simultaneously allowing for the observation of changes in the mental health of currently incarcerated inmates over time. Overall, the results
36


demonstrate a complex association between declines in the mental health of inmates and duration of time served, and can aid both correctional facilities and public health officials with understanding exposure risks, integration of treatment plans, and continuity of care post release.
37


REFERENCES
Alexander Michelle. 2010. The New Jim Crow: Mass Incarceration in the Age of Colorblindness. New York, NY: The New Press.
Bailey, Zinzi D., Cassandra Okechukwu, Ichiro Kawachi and David. R. Williams. 2015. Incarceration and Current Tobacco Smoking Among Black and Caribbean Black Americans in the National Survey of American Life. American Journal of Public Health. 11: 2275-2282.
Bales, William D., and Alex R. Piquero. 2011. Assessing the impact of imprisonment on recidivism Journal of Experimental Criminology, 5(1), 71-101. doi:10.1007/sll292-011-9139-3.
Binswanger, Ingrid A., Patrick M. Kruger, and IF Steiner. 2009. Prevalence of Chronic Medical Conditions Among Jail and Prison Inmates in the USA Compared With the General Population. Journal of Epidemiology and Community Health (1979-). (63)11: 912-919.
Binswanger, Ingrid A., Patrick J. Blatchford, Rebecca G. Lindsay and Marc F. Stern. 2011a. Risk Factors For All-Cause, Overdose and Early Deaths After Release From Prison in Washington Stat q Drug and Alcohol Dependence. 117: 1-6.
Binswanger, Ingrid A., Carolyn Nowels, Karen F. Corsi, Jeremy Long, Robert E. Booth, Jean Kutner and John F. Stiener. 2011b. From the Prison Door Right to the Sidewalk, Everything Went Downhill, A Qualitative study of Health Experiences or Recently Released Inmates. International Journal of Law and Psychiatry (34) 249-255.
Carson, E. Ann. 2015. Prisoners in 2014. Washington, DC: Bureau of Justice Statistics. http://www.bi s.gov/index.cfm?ty=pbdetail&iid=5387
Dannefer, Dale. 2003. Cumulative Advantage/Disadvantage in the Life Course: Cross-Fertilizing Age and Social Science Theory. Journal of Gerentology: SOCIAL SCIENCES. (58B)6 327-337.
Donahue, John. 2014. Coronary Artery Disease in Offender Populations: Incarceration as a Risk Factor and Point of Intervention. Journal of Correctional Health. (20)4: 302-312.
Felson, Richard B, Eric Silver and Brianna Remster. 2012. Mental Disorder and Offending in Prison 7' Criminal Justice and Behavior. (39)2: 125-143.
Glaser, Ronald and Janice K. Kiecolt-Glaser. 2005. Science and Society Stress-induced Immune Dysfunction: Implications for Health. Nature Reviews Immunology. (5)3243-251.
Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates, Garden City, NY.: Doubleday.
38


Harding, David J. 2014. Making Ends Meet After Prison. Journal of Policy Analysis and Management. (33)2: 440-470.
Houle, Brian. 2014. The Effect of Incarceration on Adult Male BMI Trajectories, USA, 1981-2006. Journal of Racial and Ethnic Health Disparities. 1: 21-28.
Howell, James C., Feld, Barry C. and Daniel P. Mears. 2002. Pp. 200 244 in From
juvenile delinquency to adult crime: criminal careers, justice policy and prevention. Edited by Loeber, Rolf and David P. Farrington. New York: Oxford University Press.
James, Doris J. and Lauren E. Glaze. 2006. Mental Health Problems of Prison and Jail Inmates. Bureau of Justice Statistics.
http://bjs.ojp.usdoj.gov/index.cfm?ty=pbdetail&iid=789
Kessler, Ronald C., Patricia Berglund, Wai Tat Chiu, Olga Dernier, Steven Heeringa,
Eva Hiripi, Robert Jin, Beth-Ellen Penned, Ellen E. Walters, Alan Zaslavsky, and Hui Zheung. 2006. "The US National Comorbid- ity Survey Replication (NCS-R): Design and Field Procedures." International Journal of Methods in Psychiatric Research (13):69-92.
Kim, KiDek, Miriam Beck-Cohen and Maria Serakos. 2015. The Processing and Treatment of Mental 111 Persons in the Criminal Justice System: A Scan of Practice and Background Analysis. Crime and Justice: Urban Institute
Kim, Yujin. 2014. The Effect of Incarceration on Midlife Health: A Life Course Approach. Population Research and Policy Review. 34: 827-849.
Kinner, Stuart A. and MJ Milloy. 2011. Collateral Consequences of an Ever Expanding Prison System. Canadian Medical Association. (183)5 632.
Lamb, Richard H., Linda E. Weinberger and Bruce H. Gross. 2004. Mentally 111 Persons in the Criminal Justice System: Some Perspectives. Psychiatric Quarterly. (75)2: 107-126.
Lamberti, Steven J. and Robert L. Wiseman. 2004. Persons with Severe Mental Disorders in the Criminal Justice System: Challenges and Opportunities. Psychiatric Quarterly.(75)2: 151-164.
Maruschak, Laura M. 2008. Medical Problems of Prisoners. Washington, DC: Bureau of Justice Statistics. http://bjs. gov/content/pub/pdf/mpp.pdf.
Massoglia, Michael. 2008a. Incarceration, Health, and Racial Disparities in Health. Law and Society Review. (42)2: 275-306.
39


Massoglia, Michael. 2008b. Incarceration as Exposure: The Prison, Infectious Disease, and Other Stress Related Illnesses. Journal of Health and Social Behavior. (49) March 56-71.
Massoglia, Michael and William Alex Pridemore. 2015. Incarceration and Health. Annual Review of Sociology. (41) 291-310.
Mears, Daniel, Joshua C. Cochran and Francis T. Cullen. 2015. Incarceration Heterogeneity and Its Implication for Assessing the Effectiveness on Recidivism. Criminal Justice Policy Review. (26)7: 691-712.
Mumola, Christopher J. 2007. Medical Causes of Death in State Prisons, 2001-2004 Washington DC: Bureau of Justice Statistics.
http://www.bjs.gov/content/pub/pdf/mcdsp04.pdf.
Munn, Melissa. 2011. Living in the Aftermath: The Impact of Lengthy Incarceration on Post-Carceral Success. The Howard Journal. (50)3: 233-246.
Nagin, Daniel S., Francis T. Cullen and Cheryl Lero Jonson. 2009. Imprisonment and Reoffending. Crime and Justice. (38)1: 115-200.
Nowotny, Kathryn M., 2016. Social Factors Related to the Utilization of Health Care Among Prison Inmates. Journal of Correctional Health Care. 22(2): 129-138.
Nowotny, Kathryn M., Alice Cepeda, Laurie James-Hawkins and Jason D. Boardman. 2015. Growing Old Behind Bars: Health Profiles of the Older Inmate Population in the United States. Journal of Aging and Health. 1-22.
Pelaez, Vicky. 2008. The Prison Industry in the United States: Big Business or a New Form of Slavery? Global Research, http://www.globalresearch.ca/the-prison-industry-in-the-united-states-big-business-or-a-new-form-of-slavery/8289
Pager, Devah. 2003. The Mark of a Criminal Record. American Journal of Sociology. (108)5: 937-995.
Pearlin, Leonard I. 1989. The Sociological Study of Stress. Journal of Health and Social Behavior. (30) 3: 241-256.
Peters, Roger H., Aurthor J. Lurigio and Harry K. Wexler. 2015. Co-Occurring Substance Use and Mental Disorder in the Criminal Justice System: A New Frontier of Clinical Practice and Research. Psychiatric Rehabilitation .Journal. (38)1 1-6
Pew Cent. States. 2008. One in 100: Behind Bars in America 2008.Washington, DC: Pew Cent. States.
40


Porter, Lauren C. 2015. Incarceration and Post-release Health Behaviors. Journal of Health and Social Behavior. (55)2 234-249.
Schnittker, Jason and Andrea John. 2007. Enduring Stigma: The Long-Term Effects of Incarceration. Journal of Health and Social Behavior. (48) 2:115-130.
Schnittker, Jason, Michael Massoglia, and Christopher Uggen. 2012. Out and Down: Incarceration and Psychiatric Disorders. Journal of Health and Social Behavior. (53)4: 448-464.
Schnittker, Jason and Michael Massoglia. 2015. A Sociological Approach to Studying the Effects of Incarceration. Wisconsin Law Review. 350-374.
Scott, Susie. 2010. Revisiting the Total Institution: Performative Regulation in the Reinventive Institution. Sociology. (44)2: 213-231.
Sered, Susan and Maureen Norton-Hawk. 2008. Disrupted Lives, Fragmented Care: Illness Experiences of Criminalized Women. Women and Health. (48)1: 43-61.
Sung, Hung-En and Jeff Mellow. 2011. Criminalization of Disease, medicalization of Justice and Beyond. Corrections Compendium. 11-16.
Thoits, Peggy A. 2010. Stress and Health: Major Findings and Policy Implications. Journal of HelathHealth and Social Behavior. (51): S41-S53.
Thomas, James C., Brooke A. Lavandowski, Malika Roman Isler, Elizabeth Torrone and George Wilson. 2007. Incarceration and Sexually Transmitted Infections: A Neighborhood Perspective. Journal of Urban Health: Bulletin of the New York Academy of Medicine. (85)1 90-99
Turney, Kristen, Christopher Wildeman and Jason Schnittker. 2012. As Fathers and Felons: Explaining the Effects of Current and Recent Incarceration on Major Depression. Journal of Health and Social Behavior. 53(4) 465-481.
Wakefield, Sara and Christopher Uggen. 2010. Incarceration and Stratification. Annual Review of Sociology. (36): 387-406.
Wakeman, Sarah E., Margaret E. McKinney and Josiah D. Rich. 2009. Filling the Gap: The Importance of Medicaid Continuity for Former Inmates. Journal of General Internal Medicine. (24)7: 860-862.
Western, Bruce, Anthony A. Braga, Jaclyn Davis and Catherine Sirois. 2015. Stress and Hardship After Prison. American Journal of Sociology. (120)5: 1512-1547.
Williams, Brie A., James McGuire, Rebecca G Lindsay, Jacques Baillargeon, Ireana Stijacic Cenzar, Sel J. Lee and Margot Kushel. 2010. Coming Home: Health Status of
41


Homelessness Risk of Older Pre-release Prisoners. Journal of General Internal Medicine (25)10 1038-1044.
Yu, Sung-suk Violet, Hung-En Sung, Jeff Mellow and Carl J. Koenigsmann. 2015. Self-Perceived Health Improvements Among Prison Inmates. Journal of Correctional Health Care. (21)1: 59-69.
42


Full Text

PAGE 1

DURATION OF INCARCERATION AND THE MENTAL HEALTH OF CURRENTLY INCARCERATED INMATES by ALYSSA ME N ARD B.A. Norwich University, 2013 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Arts Sociology Program 2017

PAGE 2

! ! "" This thesis for the Master of Arts degree by Alyssa Menard has been approved for the Sociology Program b y Adam M. Lippert, Chair Stacey J. Bosick Keith W. Guzik Date: May 1 3, 2017

PAGE 3

! ! """ Menard, Alyssa (M.A., Sociology Program) Duration of Incarceration and the Mental Health of Currently Incarcerated Inmates Thesis Directed by Associate Professor Adam M. Lippert. ABSTRACT This study examines the association between duration of incarceration in state prisons and mental health treatment needs in currently incarcerated men and women. Using nationally representative data from the 2004 Survey of Inmates in State Correctional Facilities, duration of incarceration is measured using number of prior incarcerations and duration of time served in one's current sentence. A four category typology of mental health treatment status is used to assess the association between duration of incarceration and mental health status in currently inca rcerated inmates. Results from a stratified multivariate logistic regression analysis show that duration of incarceration is significantly associated with mental health treatment for currently incarcerated inmates. This study demonstrates that lengthier in carcerations lead to more mental health treatment needs for inmates who are currently incarcerated These results demonstrate that incarceration is directly correlated with the mental health of inmates and that there are unme t mental health treatment needs in the prison population. The form and content of this abstract are approved. I recommend its publication. Approved: Adam M. Lippert

PAGE 4

! ! "# TABLE OF CONTENTS CHAPTER I. INTRODUCTION ... .1 II. BACKGROUND AND LITERATURE REVIEW ..1 A Br ief History of Incarceration in the United States .. 3 Incarceration and Mental Health .9 Duration of Incarceration .12 Stress, Coping and Instit utional Acclimations. 13 Stress Process Theory ..14 Total Institutions ...16 Current Study. .......18 III. DATA AND METHODS 20 Survey of Inmates in State Correctional Facilities .. 20 Dependent Variables 20 Mental Hea lth Treatment ..20 Independent Variables ...21 Duration of Incarceration ..21 Analysis 21 IV. RESULTS 23 Descriptive Statistics 23 Multinomial Reg ression ... 26 V. DISCUSSI ON .31

PAGE 5

! ! # Strengths and Limita tions 35 Conclusion .......36 REFERENCES 3 8

PAGE 6

! ! #" LIST OF TABLES TABLE Table 1. Descriptive Statistics of Entire Sample .23 Tabl e 2. Multinomial Logistic Models Predicting the Odds of Mental Health Treatment Compared to Inmates Who Have Never Received Treatment .... ...25 Table 3. Multinomial Logistic Models Predicting the Odds of Mental Health Treatment Compared to Inmates Who Ha ve Met Need .27

PAGE 7

! ! #"" LIST OF FIGURES FIGURE S Figure 1. Inmates Receiving Mental Health Treatment ....... 24 Figure 2. Predicted Probability of Mental Health Treatment Categories B y D uration of Incarceration.. .. 29

PAGE 8

! ! $ CHAPTER I INTRODUCTION Research has increasingly demonstrated that exposure to incarceration exacerbates mental health disorders and complicates mental health treatment needs among imprisoned persons (Massoglia 2008b; Nowatny, Cepeda and James Hawkins 2015; Schnittker, Massoglia, & Uggen 2012; Turney, Wildeman and Schnittker 2012). Like the non institutionalized population, common mental disorders experienced by inmates include anxiety disorders, mood disorders, impulse control disorder and substance d isorders ( Kessler, Berglund, Chiœ, Demier, Heeringa, Hirpi, Jin, Penneil, Walters, Zaslavasky and Zheung 2006) However, unlike the general population, the prevalence of such disorders is exponentially higher among incarcerees. Approximately 34% of individ uals within the general population have experienced some type of mental health disorder within the last twelve months, whereas over half of all inmates nearly 56% have been treated for some type of mental health disorder while incarcerated (James and Glaze 2006; Yu, Sung, Mellow and Koenigsmann 2014; Felson, Silver and Remster 2012; Kessler 2001). Mental health disorders affect a substantial proportion of inmates and have lasting consequences during and after incarceration. More than half of all inmates rep ort a wide range of mental health symptoms including mania, major depression, psychotic disorder, anxiety and mood disorder, impulse disorders, and substance use or dependence (James and Glaze 2006; Kessler 2001; Schnittker, Massoglia and Uggen 2012). Fur ther, many inmates experience co morbidity or suffer from more than one mental disorder at a time (Peters, Lurigio, and Wexler 2015). Inmates are not only at elevated risk for experiencing multiple

PAGE 9

! ! % mental health co morbidities, but they also face higher pr ospects of additional challenges post release including homelessness, employment discrimination, difficulty maintaining continuity of care, lack of social support, increased stigma, relapse, and higher rates of re arrest compared to former inmates who were not treated for mental health disorders while incarcerated (Schnittker and John 2007; Williams et al. 2010; Binswanger et al., 2011b; Yu et a. 2014). Two theoretical perspectives offer insights on how different dimensions of incarceration might influence mental health treatment seeking among inmates. Stress process theory (Pearlin 1989) holds that lengthy incarcerations coupled with a history of many such incarcerations will undermine inmate mental health, while Goffman's total institution theory (1961) contends that the institutional familiarity generated by lengthy or repeated prison terms may generate specialized skills and knowledge that inmates can deploy to protect their mental health while incarcerated. Prior literature has devoted attention to t he relationship between incarceration and mental health, but few studies have simultaneously examined how the duration of current time served and one's number of prior detainments are associated with mental well being and treatment for new or existing ment al health problems. Most prior studies on imprisonment and mental health attend to one but not both of these dimensions of institutional exposure. However, a more robust specification of such exposure would combine measures of both time served of one's c urrent sentence and one's history of prior incarcerations in order to examine the cumulative exposure to incarceration. Both prior incarcerations and duration of current time served measure different aspects of institutional exposure. Duration of current time served examines consequences of

PAGE 10

! ! & uninterrupted institutional exposure whereas number of priors measures multiple experiences of institutional exposure. Prior incarcerations as a measure of institutional exposure attends to aspects of incarceration suc h as job interruption, dislocation from society, separation from friends and family members, as well as the potential for exposure to different institutional environments with repeated incarcerations. Duration of current time served attends to similar yet different aspects of institutional exposure. Inmates who spend more time incarcerated also experience similar consequences as those who serve repeated sentences, however, as duration of time served increases the severity or magnitude of such consequences m ay be amplified. Moreover as time served increases there are additional opportunities for inmates to develop institutional knowledge, coping mechanisms, or to encounter additional risk factors while incarcerated. By examining both repeated incarcerations and length of incarceration this study will be able to determine how various aspects of cumulative exposure are associated with the mental health treatment of currently incarcerated inmates. Although there has been much research related to the health co nsequences of incarceration and the mental health of inmates, a majority of studies use general population data to explore differences in the mental health of inmates compared to the general population. This approach fails to capture differences among the mental health statuses of currently incarcerated individuals and may introduce selection bias into prior studies. Using data that compares inmates with the general population ignores much heterogeneity found within inmate populations due to similar risk fa ctors associated with socioeconomics and childhood adversities that may be correlated with both health and crime. Moreover, it is also unclear how selection bias influences prior estimates linking various aspects of incarceration to mental health treatment Using data that captures the mental health status of inmates while

PAGE 11

! ! they are currently incarcerated helps to attenuate for this bias and gain insights into the different pathways associated with the effects of incarceration on the mental health of prisone rs. In light of these knowledge gaps the current study examines the joint and independent associations among current incarceration duration, incarceration history, and mental health treatment using the 2004 National Survey of Inmates, a nationally repres entative study of currently incarcerated U S adults Specifically, the current study uses multinomial regression modeling using a typology of mental health status to address the following questions: 1. How are incarceration history and time served associate d with the odds of having a new mental health treatment experience vs. having never had mental health treatment? 2. How are incarceration history and time served associated with the odds of having an unmet mental health treatment need vs. having continuity in one's mental health treatment? 3. Are the associations between incarceration history, time served, and mental health treatment patterns independent or conditional on one another?

PAGE 12

! ! ( CHAPTER II BACKGROUND AND LITERATURE REVIEW A Brief History of Incarcerat ion in the US Over the last half century, excessively punitive sentencing policies have lead to mass incarceration in the United States. The United States' incarcerated population has increased from approximately 600,000 in the 1960s to more than 2.2 milli on in 2014 (Carson 2015). Incarceration in the United States is such a common social phenomenon that nearly one out of every 100 American citizens will end up in jail or prison at some point in their lives (Pew Cent 2008). Of those incarcerated, approximat ely 95% will be released back into their communities resulting in roughly 16 million ex convicts in the United States (Schnittker Massoglia & Uggen 2012). Much of the dramatic rise in imprisonment can be associated with changes in political policy throug hout the latter half of the 20 th century resulting from political strategies enacted partly in response to growing racial tensions and claims made by civil rights advocates (Dumont Allen, Brockmann, Alexander and Rich 2013; Alexander 2010). Some examples o f policy implementations aimed at reducing crime include the War on Drugs,' Tough on Crime,' rhetoric and mandatory minimum sentencing laws (Wakefield and Uggen 2010; Dumont et al. 2013; Howell, Feld and Mears 2012; Cid 2009). The federal government ince ntivized adherence to these policies with cash grants for participating states, which led to an increase in number of arrests as well as harsher and longer sentences especially in regard to drug crimes (Alexander 2010). Although the intent of these polici es was to decrease crime rates, these policies have generated structural oppression of entire groups of people affecting

PAGE 13

! ! ) those of socioeconomic disadvantage and minority status in a highly inequitable way (Dumont et al. 2013; Alexander 2010). Incarceratio n as a form of criminal punishment has far reaching consequences for inmates while imprisoned, and for both inmates and communities after inmates are released. Research has shown that inmates, while incarcerated, are at elevated risk for poor physical and mental health outcomes (Sung and Mellow 2011; Binswanger et al. 2009). Some examples include increased risk of exposure to infectious or communicable diseases as well as increased risk for experiencing mental health disorders such as depression and anxiet y related to the stressors of the prison environment (Massoglia 2008; Porter 2015; Schnittker and John 2007; Binswanger et al. 2009). Another consequence of incarceration is the mark' or stigmatization associated with serving time. This stigmatization co ntributes to many continued risk factors associated with incarceration after release from prison including homelessness, recidivism, increased substance abuse, poor health behaviors, employment discrimination, and difficulty maintaining social bonds (Binsw anger et al. 2011a; Porter 2015; Schnittker and John 2007; Williams et al. 2010). For instance, Pager (2003) finds that job seekers with a criminal record are one half to one third as likely to gain employment compared to non offenders resulting in chronic unemployment for previously incarcerated individuals. In a study conducted by Western et al. (2015), it was found that after six months of release, nearly half of ex inmates were still un/underemployed. Consequently, former detainees report significantly lower personal earnings than those without criminal records. Because social service programs often include eligibility requirements that exclude ex inmates (Wakeman, McKinney and Rich 2009) chronically unemployed former detainees often have unmet needs for basic social

PAGE 14

! ! services like food assistance (Harding 2014), housing support (Western, Braga, Davis, and Sirois 2015), and clinical re entry needs (Gill and Wilson 2016). It has also been shown that incarceration has a criminogenic effect and increases th e likelihood of recidivism (Nagin, Cullen and Johnson 2009; Cid 2009; Bales and Piquero 2011). Of all inmates who have been previously incarcerated more than half will recidivate within three years of release (Wakefield and Uggen 2010). For some former det ainees, repeat offenses sometimes constitute adaptive means to secure basic resources (e.g., food, cash, clinical needs) that are unavailable to them due to employment discrimination and a general lack of public assistance programs for ex inmates (Gill and Wilson 2016; Williams et al. 2010). Irrespective of the causes provoking criminal offenses following release, ex inmates who have their basic needs met through re entry services or by other means such as family and social support, are less likely to recid ivate compared to ex inmates who experience a high level of unmet need post release (Gill and Wilson 2016). Although there are empirically demonstrated disadvantages to using incarceration as a primary form of punishment in American society (Thomas, Lav andowski, Isler and Wilson 2007; Maruschak 2008; Kinner and Milloy 2011; Donahue 2014), the prison industry has grown exponentially in the last half century. The rapid expansion of punitive sentencing policies during the latter part of the 20 th century cou pled with the privatization of prisons generated a major economic opportunity for big business to capitalize on the free market establishing the prison industrial complex (PIC ) (Palaez 2008; Dumont et al. 2013). In the years following the implementation of such policies the increase in punitive action including mandatory minimum sentences and three strikes policies by the government created new market space for privatized prisons. Currently, 10% of all prisoners in the US are sent to

PAGE 15

! ! + privately owned prison (Dumont et al. 2013). With the number of people being incarcerated, an opportunity for the privatization of prisons was created leading to a mutually beneficial arrangement between government and big business. In addition to the need for private prisons fo r housing inmates, many state and federal prisons contract out services such as health care, con struction, and food services to private companies generating market rewards for major corporations and Wall Street investors (Palaez 2008). These incentives for profit generate backing for politicians and help to support political campaigns that offer continued support for the implementation of tough on crime policies in the United States that continue to generate profit for the investors of privatized correction al services (Palaez 2008). Efforts to reduce crime and protect society from crime have resulted in major changes to the way the justice system punishes criminal engagement and has a wide variety of consequences to our society, with an especially consequent ial effect on the well being and mental health of inmates (Massoglia 2008b; Nowotny et al. 2015; James and Glaze 2006). In the U S many prisons house more mentally ill individuals than inpatient psychiatric facilities (Kim, Becker Cohen and Serakos 2015 ), over half of inmates reporting some type of mental disorder (James and Glaze 2006). Prior literature has demonstrated a strong association between incarceration and mental health disorders in inmate populations (Sered 2008; Massoglia 2008b) Prisoners s uffering from mental illness are more prone to commit violent offenses, more likely to recidivate, and more likely to experience increased risk factors associated with a cyclical process of incarceration and release (Kim, Becker Cohen and Serakos 2015). Cu rrently incarcerated inmates are also highly likely to be undertreated and under diagnosed for mental health disorders while incarcerated (Schnittker, Massoglia and Uggen 2012 ; Baillargeon Hoge and Penn 2010 ). With an estimated 7.5% of

PAGE 16

! ! the population, abo ut 16 million people, being felons or ex felons, the impact of mental health disorders associated with exposure to incarceration has great consequences not only to those incarcerated but to the general population as 95% of inmates are eventually released ( Nowotny 2016; Schnittker, Massoglia and Uggen 2012). Understanding m ental health treatment while incarcerated therefore, is essential to the successful reentry of inmates to their communities as well as understanding the relationship between incarceration and mental health. Incarceration and Mental Health The high prevalence of mental health disorders present in incarcerated populations highlights the role prisons and correctional facilities play in the mental health treatment needs of inmates. Many fact ors connected with the prison environment have been associated with the production or reproduction of poor mental health outcomes among inmates (Massoglia 2008b), while prisons simultaneously provide needed care for many individuals with unmet or undiagnos ed mental health treatment need outside of the prison environment (Turney, Wildeman and Schnittker 2012). As prisons disproportionately house those of limited financial, social, and human capital, these surroundings generate and maintain environments of di sadvantage (Wakefeild and Uggen 2010). Despite the access to health care services in prisons, health benefits associated with institutionalized treatment do not persist post release (Schnittker and John 2007) leading to a high proportion of inmates and ex inmates with untreated mental health disorders. Currently incarcerated inmates have significantly higher rates of mental health disorders compared to non institutionalized persons. Over half of all inmates will receive treatment for mental health disorder s while incarcerated (James and Glaze 2006). Although

PAGE 17

! ! $! there is a high proportion of individuals who have been treated for a mental health disorder prior to incarceration, many individuals with no history of mental health problems who have been exposed to t he prison environment will seek treatment for new mental health disorders once incarcerated (Massoglia and Pridemore 2015; Schnittker and John 2007). Research has shown that individuals who recidivate are more likely to use the health services that are pro vided by correctional facilities, yet despite receiving treatment while incarcerated, individuals with a history of incarceration still report worse health overall than the general population (Schnittker and John 2007). These findings suggest that incarcer ation relates to both shaping one's risk factors for mental health pathologies, as well as influencing access to mental health treatment services. Scholarship falling at the intersection of criminology and m edical sociology has recently begun to focus on d econstructing the mechanisms between incarceration and mental health. Incarceration is known to have direct consequences on the mental health of inmates and increased mental health treatment needs in incarcerated populations. While research has shown that incarceration may lead to some temporary improvements in physical health, especially among minority populations (Houle 2014; Baily et al. 2015b), the opposite has been found for the mental health of prisoners (Turney, Wildeman and Schnittker 2012). Being i ncarcerated is associated with sudden changes in many areas of one's life and is coupled with a major shock to one's lifestyle, which can lead to a variety of chronic stressors (Massoglia and Pridemore 2015; Turney, Wildeman, and Schnittker 2012). Prior li terature (Schnittker and John 2007; Massoglia 2008b; Schnittker, Massoglia and Uggen 2012) has shown that being exposed to the prison environment results in primary and secondary stressors both in daily life and in the long term. Such stressors inhibit pos itive coping

PAGE 18

! ! $$ strategies and are associated with a decline of the mental health of inmates (Kim 2014; Massoglia 2008b). The prison environment often exacerbate s mental health disorders in individuals with a history of mental illness as well as engendering n ew mental health disorder s in previously asymptomatic individuals (Massoglia and Pridemore 2015; Nowatny et al. 2015; K im 2014; Williams et al. 2010) leading lead to an increased need for mental health treatment in prisons. Mental health disorders among in mates have been shown to have many long lasting consequences for the overall health of inmates both during and after incarceration. Some difficulties experienced by inmates with mental health disorders post incarceration include increased risk of homelessn ess and housing discrimination (Baillargeon, Hoge and Penn 2010; Schnittker, Massoglia and Uggen 2012), increased likelihood of recidivism (Lamb, Wienberger and Gross 2004), limited access to health care, (Lamberti and Wiseman 2004) difficulty gaining empl oyment or benefits (Baillargeon Hoge and Penn 2010), as well as stigmatization across multiple domains. Increased risk factors such as these not only complicate mental well being while incarcerated, but also contribute to chronic stress throughout the li fe course (Massoglia & Pridemore 2015; Massoglia 2008b) through durable pathways such as endured stigma, loss of economic or social roles post release or impaired health status (Turney, Wildeman, and Schnittker 2012). Chronic stress can lead to relapse in mental health disorders as well as inhibit the body's essential functions including cell mediated immunity (Glaser and Kiecolt Glaser 2005; Massoglia and Pridemore 2015) and cardiometabolic health (Thoits 2010), whic h may lead to additional health complica tions following release from prison. Moreover, the more risk factors ex inmates experience after being released from prison, the higher their likelihood of returning to prison perpetuating a

PAGE 19

! ! $% cyclical process of recidivism and release (Bi nswa n ger, Nowels C orsi, Long, Booth, Kutner and Stiener 2011b ; Baillargeon, Binswanger, Penn, Williams and Murray 2009). Duration of Incarceration: Current And Repeated Incarcerations An implicit supposition in research on incarceration and the mental well being of prisoner s is that the accumulation of exposure to prison based stressors will confer poorer mental health outcomes for detainees with lengthy incarceration spells. Although some studies have used inclusive measures to assess the cumulative duration of incarcerati on by examining the number of priors and incarceration status, results have been mixed ( Nagin, Cullen and Johnson 2009) Previous research examining cumulative effects of incarceration have failed to examine the joint effect of duration of time served and number of prior incarcerations on mental health disorders in currently incarcerated populations, and the treatment needs of inmates with diminished mental health. In a study conducted by Bail e y et al. (2015), duration of incarceration was operationalize d using a self report of number of prior incarcerations and total lifetime duration of incarceration. This study found no significant association between total duration of incarceration and health behaviors while incarcerated, but did find a significant as sociation between the number of prior incarcerations and current health behaviors post release. In another study conducted by Schnittker and John (2007) cumulative duration of incarceration was analyzed as a primary predictor for severe health limitations Incarceration was captured using separate measures of number of prior incarcerations and current incarceration status. Results from this study were mixed; current incarceration was negatively associated with functional health limitations, but history of incarceration was positively linked to such limitations. These patterns were not moderated by race or education levels. This study did not

PAGE 20

! ! $& evaluate whether current incarceration status moderated the relationship between prior incarceration and functional l imitations. These approaches may be problematic due to the association of processes that incarceration and health may share, suggesting that these processes may not be independent, but inter dependent. Therefore, more research is needed to assess the joint contribution of both duration of time served and number of priors on the mental health of prisoners and mental health treatment in prison as well as prior to one's incarceration. Two complementary theoretical frameworks offer insight that may be useful in understanding the association between cumulative duration of incarceration and mental health treatment outcomes in incarcerated populations. Stress, C oping, and I nstitutional A cclimation Two theoretical scenarios, Stress Process Theory and the Theory of Total Institutions, offer support linking the total cumulative duration of incarceration and mental health disorder in incarcerated populations. Both these theories support the views of cumulative advantage/disadvantage (CAD) theory. CAD theory posits that both advantage and disadvantage accumulate through the life course based on systematic exposure to available resources both within individual societies and within individuals (Dannefer 2003). Cumulative dis/advantage is a complex structural process of soc ialization and resource allocation that results in measurable differences in access to resources, health outcomes, and inequality for different social groups. Advantage s and disadvantage s ac cumulate throughout the life course over multiple pathways includi ng education, economic resources, and access to health care. Accumulated disadvantage in the form of repeated and/ or lengthy

PAGE 21

! ! $' incarceration spells has several consequences in relation to exposure to the prison environment. Stress Process Theory Stress pr ocess theory (SPT) is a foundational theory in medical sociology, and holds that human stress occurs as a result between a mismatch between the demands placed upon a person and the resources they have to meet those demands. Specifically, SPT suggests that cumulative exposure s to stressful environments (i.e prison) disproportionally disadvantage individuals due to structural systems of stratification that unevenly expose them to stress while simultaneously straining their resources to cope with taxing dema nds a confluence of challenges that may result in health related inequalities (Pearlin 1989 ; Thoits 2010 ). Pearlin (1989) describes how life events requiring major adjustments may be potentially harmful, especially in undesirable situations. Stressful ev ents generally occur through multifaceted and interrelated environmental and structural factors such as social stratification, interpersonal relationships, and social institutions (Turney, Wildeman and Schnittker 2012) People who share similar life experi ences commonly experience similar stressors, and patterns of stress distribution are often rooted in social and economic situations. SPT discusses the many components of stress and relates stressors back to chronic strain in ones life, often resulting from an initial event or situation. Moreover, stressors occur in cluster s, are rarely independent of other parts of one s life and accumulate throughout the life course (Massoglia and Pridemore 2015; Thoits 2010) SPT proposes that there are both primary an d secondary stressors associated with chronic strain. Primary stressors are the result of major or non normative change in one s life. Once a change has taken place, additional secondary stressors generally occur. These

PAGE 22

! ! $( stressors are often associated with social, economic, and environmental circumstances. For instance, in a non institutionalized population, a primary stressor may constitute job loss while a secondary stressor may materialize in the forms of relationship instability, food insecurity, or hom elessness. Incarceration is an example of an undesirable event and is usually a major change for any individual getting incarcerated; this would be considered the primary stressor (Massoglia and Pridemore 2015) However, one can easily see that there are several secondary stressors associated with the realities of living in the prison environment such as fear of violence, loss of autonomy, social isolation and economic hardship (Massoglia 2008b; Turney, Wildeman and Schnittker 2012) Pearlin goes on to ex plain that institutional roles can increase stressors due to persistent and repeated features of an institution. C hallenges associated with institutional stressors include interpersonal conflict, repetitive activities, and role captivity (Pearlin 1989) Ro le captivity exists when one is unwillingly stuck in an undesirable role, as is most often the case with inmates. In addition to stressors specifically associated with an institutionalized setting, many experience ambient strains associated with the burden of becoming incarcerated. Ambient strains include physical and environmental strains such as a serious or chronic illness or fear of vehemence or discrimination by staff or other inmates. Moreover, all strains and stressors contain multiple sub types of s tressors, creating persistent chronic strain, especially within the prison environment. In order to manage such stressful situations, individuals must learn to enact coping mechanisms. Coping is the act of attempting to avoid or lessen the consequences as sociated with the impact of life problems and stressors (Pearlin 1989) Some ways to implement coping mechanisms include trying to change the situation or escape trying to manage the

PAGE 23

! ! $) meaning of the situation, or to find ways to manage the symptoms of the stress through behavioral and emotional coping (Leban, Cardwell, Copes and Brezina 2016). In incarcerated populations, this may be challenging as often times, there is little an inmate can do to change the situation and resources to lessen the impact of st ressful situations may be in short supply Therefore, the methods of coping one can engage in tend to be more limited than during other facets of life. Coping mechanisms can be positive or negative, and they can be healthy or unhealthy. Some common coping mechanisms used in prisons include seeking out social support or help, cognitive coping such as minimizing the importance or ignoring the stressor, accepting responsibility for one's actions, attempting to escape the situation, emotional coping such as see king time alone to draw or read, or behavioral coping such as retaliation ( Leban, Cardwell, Copes and Brezina 2016) Despite attempts to cope with chronic strain and stressors there can be a substantially detrimental effect on health functioning. In the prison environment, multidimensional stress can lead to increased stigma, poor physical health, and poor coping strategies all contributing to diminished mental health in incarcerated populations (Massoglia 2008a; Nowotny et al. 2015; Massoglia and Pridemo re 2015; Porter 2015; Sered and Norton Hawk 2008). By implementing coping mechanisms through utilizing available resources, inmates may be able to better adapt to the adversities of the prison environment. Total Institutions Theory Another theory that res onates with the perspectives of CAD theory is Goffman's theory of Total Institutions (TI) (Goffman 1961). In Goffman's discussion of total institutions, he describes total institutions as authoritarian organizations of corrective sanction in which inmate b ehavior is regularly judged and restricted (Scott 2010) The

PAGE 24

! ! $* authoritarian environment of total institutions results in the simultaneous depersonalization (i.e. elimination of all personal items upon entry) and standardization ( i.e. the use of uniforms and provision of necessary material) of individuals. Personal identity is limited and provisions are provided for all equally and lack uniqueness Total institutions are built on a system of house rules and there are severe consequences for disobedience. Thes e rules provide punishments and small rewards for good or bad behaviors while in the institution (Goffman 1961) Due to this challenging and incapacitating environment an individual's autonomy is highly restricted and one is strip p ed of their ability to re gulate the self. As total institutions are severely restrictive, Goffman offers an explanation of coping mechanisms deployed in order to manage the stress of such environments. TI suggests that cumulative exposure to an institutionalized environment creat es uniquely specialized skills that enable stronger coping methods allowing individuals to better handle the demands of the prison environment (Schnittker, Massoglia and Uggen 2012) TI theory tells us that repeated and lengthy exposure to incarceration ai ds inmates in learning to utilize value d resources within the prison environment through adaptations and adjustments. This adaptation generally involves skills such as developing an institutional language that is used to describe the inmate world, an accum ulation of specific institutional knowledge such as knowledge about the institution, various officials, and comparative information about other similar institutions (through prior incarcerations), the implementation of secondary adjustments, building socia l controls at the inmate level, bond formation, and creating barriers to the psychological stressors associated in the prison environment (Goffman 1961). A primary tool used to develop unique coping methods and reduce psychological stressors relates to t h e use and implementation of secondary adjustments, which are

PAGE 25

! ! $+ practices that may allow an inmate to obtain forbidden pleasures without directly challenging the prison staff. Some examples include procuring prohibited resources such as homemade alcohol or cigarettes or engaging in discouraged activities such as gambling or sexual acts (Goffman 1961). These activities offer an inmate a level of personal autonomy that is often absent in prison environment (Scott 2010) Moreover, TI theory posits that inmates who have had lengthy or repeated incarcerations are more likely to be able to use these adaptive techniques to perfect their coping mechanisms while incarcerated, especially with repeated exposures (Munn 2011) Implementing such coping methods while incar cerated may help to reduce consequences on mental health during one's current incarceration. Current Study While prior research has extensively examined mental health consequences of incarceration, not all extant research in this area simultaneously atten ds to prior incarceration history and the length of one's current incarceration. Indeed, much of this research is based on general population samples, constrained to small subsamples with som e prior incarceration history. Further, no research to the autho r's knowledge has investigated both the independent and joint effects of prior incarcerations, length of current incarceration, and mental health treatment outcomes among c urrently incarcerated persons. Given these research gaps, the current study uses Str ess Process and Total Institutions theories, multinomial logistic regression techniques and data from the Survey of Inmates in State Correctional Facilities to address the following research aims: (1) assess the relationship between duration of time served and the mental health treatment needs of currently incarcerated inmates, (2) evaluate how repeated incarcerations contributes to the mental health treatment needs of currently incarcerated inmates, and (3) examine the joint effect of

PAGE 26

! ! $, both number of prior incarcerations and duration of time served in one's current sentence on the mental health treatment of currently incarnated inmates to evaluate interaction effects of these two measures. By ex ploring the treatmen t seeking behaviors of inmates this study w ill be able to evaluate how sentence length and number of prior incarcerations may contribute to managing the stressors associated with the harsh prison environment allowing us to gain a stronger understanding of the consequences of incarceration.

PAGE 27

! ! %! CHA PTE R III DATA AND METHODS 2004 Survey of Inmates in State Correctional Facilities This study uses nationally representative cross sectional data from the 2004 Survey of Inmates in State Correctional Facilities (SISCF). The SISCF was designed by the Bureau of Justice Statistics and measures extensive demographic and socioeconomic characteristics as well as criminal history backgrounds, drug and weapon use, health of prisoners, and military service. The Bureau of the Census collected data for this survey bet ween October 2003 and May 2004. A two stage sampling procedure was used to collect the data. First, individual prisons were selected to participate and then from within each prison, individual inmates were selected from within selected prisons. Inmates we re interviewed by computer assisted personal interviews and telephone interviews and had an 89.1% response rate. This study uses public use data, which were downloaded from the ICPSR website on September 2, 2016. These data have 14,499 observations in the public use data set. List wise deletion was used to eliminate data with missing responses on variables of interest resulting in a total sample of N=10,907. Dependent Variables Mental Health Treatment This study focuses on pre incarceration and current m ental health treatment of prisoner s Two variables measuring history of mental health treatment in the 12 months prior to the current incarceration and treatment during current incarceration were used to create a four category dependent variable indicating each of the following: has not received mental health treatment while incarcerated and has no history of mental health treatment (=0); has

PAGE 28

! ! %$ not received treatment but has history of mental health treatment (=1); has received treatment but has no history of mental health treatment (=2); and has received treatment and has history of mental health treatment (=3) (see Figure 1 for a descriptive picture of the prevalence of each of these categories in the analytic sample). For parsimony, I refer to these categor ies as having never received treatment, has unmet treatment needs, has new treatment needs, and has consistently met received treatment. Independent Variables Duration of Incarceration The main independent variables are the main effects and interaction terms of duration of time served in one's current sentence and number of prior incarcerations. Using a data driven method, duration of time served was broken into non linear categorical measures based on tertile categories of short, medium and long duratio n of time served. Short sentence duration (=0) is defined as 0 19 months of time served. Medium duration (=1) of time served is 20 54 months served, and long duration (=2) of time served is 55 523 months served. Number of prior incarceration is a dichotomo us variable divided between those with no prior incarcerations and those with one or more prior incarcerations. Interaction terms were generated for all interactions between time served and prior incarcerations. Covariates Other covariates in this analys is included age (18 96), gender (female=0, male=1), rac e (White =0, Black =1, Hispanic = 2, Other =3), income the month prior to incarceration measured in dollars ($0 $8,750), high school graduation status (did not graduate =0, graduated high school =1 ) and sentence length in month s (0 523).

PAGE 29

! ! %% Analyses This study first estimates both descriptive statistics and then more sophisticated multivariate multinomial logistic regression models using Stata 13 to analyze the data. List wise deletion was used to omit respondents who had missing data on variables of interest. Table 1 presents descriptive characteristics of the entire sample (N=10,907). Using multinomial logistic regression analyses, Model 1 examines the odds of receiving mental health treatment us ing the four category typology constructed from two measures of current and prior mental health treatment in a currently incarcerated population. Model 1 uses both duration of time served in one's current sentence and number or prior incarcerations as prim ary predictors for mental health treatment during current incarceration. Multinomial logistic regression models allows for between group comparisons based on the four category typology of mental health treatment status in currently incarcerated inmate popu lations. Model 1 uses two reference groups, never treated and met need, to examine group differences in the odds of receiving mental health treatment while currently incarcerated. Model 2 includes interactions between number of prior incarcerations and du ration of time served in one s current sentence. All models include demographic and socioeconomic control variables. All tables denote statistically significant differences ( p < .05 p < .01, p <.001 ).

PAGE 30

! ! %& CHAPTER IV RESULTS Descriptive Statistics Table 1 presents descriptive statistics of the entire sample. Among all sample participants the majority of individuals incarcerated are male. The mean age of the sample is 36. The average sentence length across all inmates is 55 months or almost four and a half years. Nearly three quarters of inmates currently incarcerated in state prisons have had at least one prior incarceration (73%) Those who have had a short duration (0 19 months) of time served make up 34% of the sample those with a medium duration (20 5 4 months) of time served represent 32% of the sample and those with a long duration (55 523 months) of ti m e served are 33% of the sample. Table 1. Sample Description, Survey of Inmates In State Correctional Facilities (N=10,907) Variable Mean/Proportion SE Mental Health Treatment Never Treated .74 .004 Met Need .10 .003 Unmet Need .04 .002 Newly Treated .11 .003 Prior Incarcerations .73 .004 Time Served Short (reference) .34 .005 Med ium .32 .005 Long .33 .005 Age 36 .100 Male .80 .004 Race White (reference) .39 .005 Black .38 .005 Hispanic .17 .004 Other .06 .002 Education 10.91 .022 Graduated .84 .003 Monthly Income 2,038 21.5 Sentence Length in Months 902 25.4

PAGE 31

! ! %' Of the entire sample, 74% of inmates have never received mental health treatment prior to incarceration nor are receiving treatment for mental health disorders during their current incarceration. This group represen ts the never treated' group. Those who were previously receiving mental health treatment prior to their current incarceration and are still receiving mental health treatment while incarcerated represent 10% of the sample and represent met need'. Inmates who were previously receiving mental health treatment prior to their current incarceration but are no longer receiving mental health treatment represent 4% of the sample. Such inmates represent unmet need'. Inmates who are receiving new treatment with no history of mental health treatment equal 11% of the sample. These inmates are the new treatment' group. Figure 1 illustrates the percentage of inmates in each typology of mental health treatment history and current mental health treatment in the entire sa mple. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Not Receiving Treatment Currently Receiving Treatment Figure 1. Typology and Prevelance of Mental Health Treatment Among Currently Incarcerated Inmates in State Prisons (N=10,907) No History of Treatment History of Treatment

PAGE 32

! ! %( Table 2. Multinomial Logistic Models Predicting the Odds of Receiving Mental Health Treatment (N=10,907)a Model 1 Model 2 Met Need Unmet Need New Treatment Met Need Unmet Need New Treatment OR OR OR OR OR OR (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) (95% CI) Age 1.01** 0.99* 0.99*** 1.00** 0.99* 0.99*** (1.00, 1.01) (.98, 1.00) (.98, .99) (1.00, 1.01) (.98, 1.00) (.98, .99) Male 0.26*** 0.44*** 0.33*** 0.26*** 0.43*** 0.33*** (.23, .30) (.35, .55) (.29, .39) (.29, .30) (.35, .55) (.29, .39) Race (reference White) Black 0.37*** 0.41*** 0.50*** 0.37*** 0.40*** 0.5*** (.32, .43) (.32, .51) (.43, .57) (.32, .43) (.32, .52) (.43, .57) Hispanic 0.36*** 0.36*** .49*** 0.36*** 0.40*** 0.49*** (.29, .44) (.26, .50) (.40, .59) (.29, .44) (.32, .51) (.40, .59) Other 0.89 0.54** 0.80 0.88 0.54** 0.80 (.69, 1.13) (.34, .85) (.61, 1.03) (.69, 1.13) (.34, .86) (.61, 1.03) Graduated .74*** 0.68** .78** 0.74*** 0.68** 0.78** (.62, .88) (.53, .88) (.66, .93) (.62, .87) (.59, .88) (.66, .93) Monthly Income .95*** 0.96* .97* 0.95*** 0.96* 0.97*** (.93, .97) (.93, .99) (.95, .99) (.93, .97) (.93, .99) (.95, .98) Sentence Length in Months 1.00 1.00 1.00* 1.00 1.00 1.00* (1.00, 1.00) (1.00, 1.00) (1.00, 1.00) (1.00, 1.00) (1.0, 1.0) (1.00, 1.00) Prior Incarcerations 1.31*** 1.19 1.30*** 1.52** 1.23 1.61** (1.12, 1.52) (.95, 1.50) (1.12-1.51) (1.17, 1.97) (.88, 1.71) (1.17, 2.2) Time Served (reference short) Medium 1.41*** 0.77* 2.09*** 1.71*** 1.01 2.4*** (1.21, 1.65) (.61, .97) (1.76, 2.48) (1.26, 2.34) (.65, 1.55) (1.66, 3.45) Long 1.20* 0.67** 3.38*** 1.38* 0.48** 4.29*** (1.01, 1.44) (.50, .87) (2.83, 4.03) (1.00, 1.92) (.28, .84) (3.05, 6.02) PriorXMedium 0.77 0.69 0.84 (.54, 1.10) (.41, 1.14) (.56, 1.26) PriorXLong 0.83 1.51 0.73 (.57, 1.19) (.83, 1.76) (.50, 1.06) Constant .55*** 0.45** .48*** 0.49*** 0.44** 0.41*** (.38, .78) (.26, 0.77) (.33, .69) (.33, .72) (.25, .77) (.27, .67) Confidence intervals in parenthesis p <.05,** p<.01, *** p<.001 a Omitted reference group is never received treatment

PAGE 33

! ! %) Multinomial Regression Table 2 presents results from a multinomial model comparing the odds of having (1) unmet mental health treatment needs, (2) newly treated mental health needs, and (3) consistently treated mental health needs, relative to (0 ) having never received mental health treatment, as a function of time served, prior incarcerations, and controls. Model 1 demonstrates the main effects of duration of incarceration and prior incarcerations on the mental health treatment status of curren tl y incarcerated inmates. Results show that duration of time served is positively and significantly associated with mental health treatment. For inmates who have served a medium vs. short sentence, the odds of having new treatment is 2.09 times the odds of n ever being treated (95% confidence interval [CI] = 1.76, 2.48). Those with a long vs. short duration of time served have 3.38 the odds of having new treatment vs. having never been treated (95% CI = 2.83, 4.03). Inmates who have prior incarcerations have 1 .30 times higher odds of having new treatment c ompared to inmates who have never received treatment (95 % CI = 1.12, 1.51) Model 1 shows that inmates who served a m edium vs. a short period of incarceration are more likely to have met need compared to bei ng never treated. For inmates who have served a medium vs. short sentence, the odds of having met need is 1.41 times the odds of never being treated for any mental health disorder (95% CI = 1.21, 1.65). Those who have served a long vs. a short sentence hav e 1.20 times higher odds of having met need compared to never receiving mental health treatment ( 95% CI = 1.01, 1.44) Inmates who have a prior incarceration have 1.31 times higher odds of having met need compared to never being treated (95% CI = 1.12, 1.5 2). Model 1 shows that for those with a medium duration of time served vs. a short duration of time served the odds of having unmet need compared to never

PAGE 34

! ! %* Table 3. Multinomial Logistic Regression Models Predicting Mental Health Treatment Outcomes vs. Havi ng Met Treatment Needs (N=10,907) a Model 1 Model 2 Unmet Need New Treatment Unmet Need New Treatment OR OR OR OR (95% CI) (95% CI) (95% CI) (95% CI) Age 0.98*** 0.98*** 0.98*** 0.98*** (.97, .99) (.98, .99) (.97, .99) (.9 7, .99) Male 1.67*** 1.26** 1.66*** 1.27* (1.30, 2.14) (1.05, 1.51) (1.30, 2.14) (1.06, 1.52) Race (reference White) Black 1.10 1.34** 1.09 1.35** (.84, 1.43) (1.11, 1.63) (.84, 1.42) (1.11, 1.52) Hispanic 1.01 1.37 1.01 1.37* (.70, 1.47) (1.05, 1.52) (.69, 1.47) (1.05, 1.63) Other 0.61 0.91 0.61* 0.90 (.37, 1.00) (.65, 1.25) (.37, 1.01) (1.05, 1.80) Graduated 0.92 1.06 0.92 1.06 (.69, 1.23) (.85, 1.32) (.69, 1.23) (.85, 1.32) Month ly Income 1.02 1.02 1.01 1.00 (.98, 1.05) (1.00, 1.05) (.98, 1.05) (.99, 1.05) Sentence Length in Months 1.00 1.00** 1.00 1.00** (1.00, 1.00) (1.00, 1.00) (1.00, 1.00) (1.00, 1.00) Prior Incarcerations 0.91 1.00 0.81 1.06 (.70, 1. 18) (.82, 1.21) (.55, 1.20) (.72, 1.56) Time Served (reference short) Medium 0.54*** 1.48*** 0.59* 1.40 (.42, .71) (1.12, 1.82) (.35, .97) (.90, 1.1)7 Long 0.55*** 2.81*** 0.35*** 3.09*** (.40, .76) (2.23, 3.54) (.19, .64) (2.00, 4.77) PriorXMedium 0.89 1.08 (.49, 1.60) (.66, 1.79) PriorXLong 1.83 0.88 (.93, 3.60) (.54, 1.43) Constant 0.82 0.88 0.89 0.84 (.45, 1.51) (.56, 1.39) (.47, 1.70) (.50, 1.42) Confidence intervals in par enthesis p <.05,** p<.01, *** p<.001 a Omitted reference group is met need

PAGE 35

! ! %+ receiving treatment are .77 times lower (95% CI = .61, .97). For inmates who have served a long sentence vs. a short sentence there is .67 times the odds o f having unmet need compared to never receiving treatment (95% CI = .50, .87). Model 2 shows no significant effect of the interaction between number of prior incarcerations and duration of time served, suggesting that duration of time served and number of priors have only significant independent effects. Table 3 presents results from a multinomial model comparing the odds of having (1) unmet mental health treatment needs, (2) newly treated mental health needs, relative to (3) consistently treated mental he alth needs, as a function of time served, prior incarcerations, and controls. Model 1 demonstrates the main effects of duration of incarceration and prior incarceration s These results show that inmates are more likely to have met need compared to unmet ne ed. The odds of having met need compared to having unmet need are .54 times lower for inmates serving a medium sentence vs. a short sentence (95% CI = .42, .71). For inmates with a long vs. short duration of time served, the odds of having met need compare to unmet need are .55 times lower (95% CI = .40, .76). Results in Table 3 show that duration of time served is significantly associated with mental health treatment inmate populations in inmates with new treatment compared to those with met need I nmates with who have served a medium sentence vs. a short sentence have 1.48 times the odds of being newly treated compared to having met need (95% CI = 1.12, 1.82). For those inmates who have served long sentences vs. short sentences, the odds are 2.81 times hi gher for receiving new treatment group compared to having met need (95% CI = 2.23, 3.54).

PAGE 36

! ! %, Model 2 in Table 3 includes the interaction between time served and number of prior incarcerations. There are no significant effects of the interactions between dura tion of time served and prior incarcerations in this model. Figure 2 summarizes the findings from Tables 2 and 3 in the form of fitted predicted probabilities of membership' in each of the mental health treatment categories by duration of incarceratio n. The fitted probabilities are adjusted for the mean effect of all controls in the fully adjusted models. Results indicate that the probability of being in the never treated' category was highest for those with short durations of incarceration ( 82 %), fo llowed by those with medium durations of incarceration ( 75 %), and lowest among those with the longest intervals of incarceration ( 72 %). For those inmates in the met need' category, results demonstrate the probabliltity for being in this category were highest for those with medium duration of incarceration (10%). For those who had served a short sentence duariton or a long sentence duration the probablility of being in the met need' category decreased to 8%. The proabablity of being in 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Never Treated Met Need Unmet Need New Treatment Notes: adjusted for age, sex, race, income, education, and sentence length Figure 2. Predicted Probability of Mental Health Treatment Catagories by Duration of Incarceration Short Medium Long

PAGE 37

! ! &! the unmet need category decreased over time. Results show that for inmates who have a short duration of incarceration the probablitly of being in the unmet need category is highest (5%), but as time served increases the proability of being in the unmet need category de creses. For those who have served a meduim or long sentence the probablity of being in this category decreased by almost half (3%) Results specify that the proablility of being in the new treatment' category increased as duration of time served increase s. The probablilty of being in this category is lowest for those who have served a short sentence (6%), followed by those with medium duratino n of incarcertion (11%) and is highest among inmates who have served a long sentence duration (16%).

PAGE 38

! ! &$ CHAPTER V D ISCUSSION Re cent studies have shown that time spent incarcerated is correlated with declines in the mental health of inmates ( Kim, Beck Cohen and Remster 2012; Lamb, Weinberger and Gross 2014 ). Research has shown that inmates have a much higher prevalence of mental health disorder compared to the general population. While incarcerated, more than half of inmates will seek some type of treatment for a mental health disorder (James and Glaze 2006), regardless of prior history of mental disorder. Several conse quences related to the prison environment have been associated with poor mental health outcomes of inmates such as increased stress, environmental risk factors, and post release consequences (Massoglia 2008b, Schnittker and John 2007). With over 2 million inmates currently incarcerated, understanding the association between mental health and duration of time served is important for both scholars and policymakers. Although much research has examined the strains associated with being incarcerated (Binswanger Kruger and Steiner 2009; Turney, Wildeman and Schnittker 2012; Kim, Beck Cohen and Remster 2012) few studies have examine d the effects of both number of prior incarcerations and duration of time served in one's current sentence to assess the consequence s of duration of incarceration on the mental health of currently incarcerated inmates limiting our u nderstanding of how duration of incarceration is associated with receiving treatment for new and previously diagnosed mental health disorders in currently incarcerated inmates. Moreover, studies on the mental health of inmates often use a subsample from general population data to assess the hazards of incarceration making it difficult to tease out differences within the inmate population.

PAGE 39

! ! &% The current study examined these gaps in the literature by using multinomial logistic regression to analyze the odds of receiving new or continued treatment and discontinuing treatment or never having received treatment for mental health disorders while incarcerated as a function of the independent and joint effects of duration of current incarceration and prior incarcerations Data from the 2004 National Survey of Inmates in State Correctional F acilities were used to assess mental health treatment patterns in currently incarcerated populations The four category typology considered here includes inmates who have never been treated, inmates with met need, inmates with unmet need, and inmates who are currently receiving new treatment. Using this typology I was able to addr ess the flowing research aims: (1) assess the relationship between duration of time served and mental health outcomes, (2) examine the association between prior incarcerations and mental health treatment, and (3) evaluate the joint effect of number of prio rs and duration of time served. Duration of time served was significant across all treatment typologies An interesting finding showed that prior incarcerations were only significant when comparing inmates with met need and new treatment to inmates who ne ver received treatment A history of prior incarcerations therefore, was positively associated with receiving mental health treatment while incarcerated in subsequent incarcerations An unexpected result demonstrated that i n no case was the interaction bet ween number of prior incarcerations and duration of time served significant However, this finding shows that although there are significant main effects for both variables, these measures are independent and not inter dependent. The four category treatme nt typology included two origin patterns : those with a history of treatment prior to incarceration and those with no history of incarceration. The difference in origin correlates to distinct patterns in expected treatment of inmates based on

PAGE 40

! ! && duration of t ime served. Inmates with met and unmet need both share a common history of mental health treatment prior to their current incarceration. Inmates with met need demonstrate continuity of treatment whereas unmet need shows interrupted treatment suggesting th at these inmates have different selection pathways once incarcerated. This study finds that inmates who have a history of mental health treatment are more likely to have met need compared to having unme t need the longer incarcerated. C onsistent with Total Institutions theory, this finding suggests that inmates who have a history of mental health treatment and who spend more time incarcerated are more likely to have their mental health treatment n eeds met. This may be due to the accumulation of institutiona l knowledge gained through prolonged exposure to the prison environment. Total Institutions theory posits that repeated or extended lengths of time spent in a total institution reduces ones autonomy leading to utilizing coping mechanisms through the use of adaptation and institutional knowledge (Goffman 1961). Inmates with who are incarcerated for longer durations are more likely to develop specialized skills such as learning how to access and take advantage of services provided within the prison, including access to mental health care treatments. Additionally, the results in this analysis demonstrate that individuals with a history of multiple incarcerations are more likely to access mental health treatment compared to individuals who are serving their firs t sentence, which shows that those who have repeated institutional exposure s may be more capable or willing to use prison resources. Moreover, the results show that longer duration of time served is associated with met need compared to unmet need suggestin g that inmates who are incarcerated for greater lengths develop coping mechanisms not limited to seeking treatment but perhaps also relating to personal autonomy helping to reduc e stressors associated with the prison environment.

PAGE 41

! ! &' The second origin patter n illuminated in the current study examines inmates who have no history of mental health treatment prior to their current incarceration. Results from the analysis show that the longer inmates are incarcerated the more likely they are to receive new treatme nt and the less likely they are to have been never treated This pattern suggest s that factors associated with the prison environment contribute to increased mental health anguish the longer one spends incarcerated as well as with repeated exposures throu gh multiple incarcerations In accordance with stress process theory major changes with in ones life are associated with increased stress and difficulty coping (Pearlin 1989) Being incarcerated is a major shock to one's life, and the longer duration of in carceration or the more times one is incarcerated, the more challenging one may find coping with such a change. These findings support stress process theory and suggest that incarceration is highly correlated with increased stress ors and chronic strain, as well as the manifestation of new or under diagnosed mental health disorder s that are associated with time spent incarcerated. Inmates who have never received treatment for a mental health disorder prior to or during their current incarceration are most l ikely to have served a short sentence compared to inmates who have received new treatment It is possible that inmates who serve short sentences are less likely to seek treatment simply due to the limited time spent incarcerated. Moreover, i t is possible t hat i nmates with shorter sentence duration lack the extended exposure to the stressors of the prison environment and therefore may not suffer the same mental health consequences as those who are incarcerated for longer durations. It is also plausible that those who serve longer sentences experience different or cumulative stressors compared to inmates who served short sentences resulting in a greater need for mental health treatment while incarcerated.

PAGE 42

! ! &( Strengths and Limitations This study has several streng ths. The first strength of this study is the large nationally representative data ( N =10,907), which provides enough power to detect subtle patterns and differences among mental health outcomes in incarcerated populations. In order to provide an objective m easure of model fit and parsimony, this study used AIC scores to select the best fit ting model for the data. Second, this study uses a stratified sample of only currently incarcerated inmates instead of using a control sample from the general public, which reduces selection bias in estimating the consequences of incarceration on mental health treatment. Comparing only inmates significantly reduces selection bias within the sample, and allows us to infer pathways between incarceration and mental health treat ment in inmate populations. Third this is one of few studies to examine actual duration of time served using data driven percentile ranks (i.e. tertiles ) as a predictor for mental health treatment in incarcerated populations. Finally, the current study co nsiders an outcome that includes categories for pre incarceration treatment experiences as well as experiences during one's current incarceration. This innovation provides more detail on the history and current status of one's interaction with mental heal th professionals, and better identifies those positively selected into prison based on prior mental health problems from those who develop new treatment needs while incarcerated. This study has several noteworthy limitations. Currently, there is no perfe ct measure for total cumulative duration of incarceration over the life course. While this research generated a new approach to assessing cumulative duration of incarceration by examining both number of prior incarcerations and duration of time served, fut ure studies should produce more precise measures of total cumulative duration of incarceration over the life

PAGE 43

! ! &) course. Improving this measure would more accurately capture the consequences of the cumulative exposure of incarceration allowing for a greater un derstanding of the specific mechanisms contributing to poor mental heal th in incarcerated populations. A second limitation of this study is that it used cross sectional data, which limits the observations to a single observation for each participant and ma kes determining causality difficult. However, by using a stratified model to help adjust for bias based on prior mental health treatment the study was able to partially attenuate for this limitation Using longitudinal data would increase the ability to ex amine changes in the mental health within individuals over time allowing for a better understanding of precise pathways to mental health outcomes among inmates. A third limitation to this study was that based on the results, it is difficult to make claims regarding the primary motivation for mental health treatment seeking among inmates. Both increased stressors and acclimation of the institutional environment may contribute to treatment seeking in currently incarcerated populations Conclusion This resea rch provides clear evidence of the consequences to mental health based on duration of incarceration using a typology of mental health status in currently imprisoned populations However, m ore research is needed to explore how cumulative duration of incarce ration over the life course is associated with mental health outcomes in inmate populations and what factors contribute to seeking mental health treatment while imprisoned Future studies should work to improve the measure of cumulative lifetime duration o f incarceration allowing for greater causal inference of the specific pathways leading to poor mental health among inmates while simultaneously allowing for the observation of changes in the mental health of currently incarcerated inmates over time. Overal l, the results

PAGE 44

! ! &* demonstrate a complex association between declines in the mental health of inmates and duration of time served, and can aid both correctional facilities and public health officials with understanding exposure risks, integration of treatment plans, and continuity of care post release.

PAGE 45

! ! &+ REFERENCES Alexander Michelle. 2010. "The New Jim Crow: Mass Incarceration in the Age of Colorblindness." New York, NY: The New Press. Bailey, Zinzi D., Cassandra Okechukwu, Ichiro Kawachi and David. R. Willi ams. 2015. "Incarceration a n d Current Tobacco Smoking Among Black and Caribbean Black Americans in the National Survey of American Life." American Journal of Public Health. 11: 2275 2282. Bales, William D., and Alex R. Piquero. 2011. "Assessing the impact of imprisonment on recidivism." Journal of Experimental Criminology, 8 (1), 71 101. doi:10.1007/s11292 011 9139 3. Binswanger, Ingrid A., Patrick M. Kruger, and JF Steiner. 2009. "Prevalence of Chronic Medical Conditions Among Jail and Prison Inmates i n the USA Compared With the General Population." Journal of Epidemiology and Community Health (1979 ). (63)11: 912 919. Binswanger, Ingrid A., Patrick J. Blatchford, Rebecca G. Lindsay and Marc F. Stern. 2011a. "Risk Factors For All Cause, Overdose and Ea rly Deaths After Release From Prison in Washington State." Drug and Alcohol Dependence. 117: 1 6. Binswanger, Ingrid A., Carolyn Nowels, Karen F. Corsi, Jeremy Long, Robert E. Booth, Jean Kutner and John F. Stiener. 2011b. "From the Prison Door Right to t he Sidewalk, Everything Went Downhill," A Qualitative study of Health Experiences or Recently Released Inmates. International Journal of Law and Psychiatry (34) 249 255. Carson, E. Ann. 2015. "Prisoners in 2014." Washington, DC: Bureau of Justice Statist ics. http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5387 Dannefer, Dale. 2003. "Cumulative Advantage/Disadvantage in the Life Course: Cross Fertilizing Age and Social Science Theory." Journal of Gerentology: SOCIAL SCIENCES. (58B)6 327 337. Donahue, John. 2014. "Coronary Artery Disease in Offender Populations: Incarceration as a Risk Factor and Point of Intervention." Journal of Correctional Health. (20)4: 302 312. Felson, Richard B, Eric S ilver and Brianna Remster. 2012. "Mental Disorder and Offending in Prison." Criminal Justice and Behavior. (39)2: 125 143. Glaser, Ronald and Janice K. Kiecolt Glaser. 2005. "Science and Society Stress induced Immune Dysfunction: Implications for Health Nature Reviews Immunology. (5)3243 251. Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates Garden City, NY.: Doubleday.

PAGE 46

! ! &, Harding, David J. 2014. "Making Ends Meet After Prison." Journal of Policy Analys is and Management. (33)2: 440 470. Houle, Brian. 2014. "The Effect of Incarceration on Adult Male BMI Trajectories, USA, 1981 2006." Journal of Racial and Ethnic Health Disparities 1: 21 28. Howell, James C., Feld, Barry C. and Daniel P. Mears. 2002. Pp 200 244 in From juvenile delinquency to adult crime: criminal careers, justice policy and prevention. Edited by Loeber, Rolf and David P. Farrington. New York: Oxford University Press. James, Doris J. and Lauren E. Glaze. 2006. "Mental Health Proble ms of Prison and Jail Inmates." Bureau of Justice Statistics. http://bjs.ojp.usdoj.gov/index.cfm?ty=pbdetail&iid=789 Kessler, Ronald C., Patricia Berglund, Wai Tat Chi, Olga Demier, Steven Heeringa, Eva Hiripi, Robert Jin, Beth Ellen Penneil, Ellen E. W alters, Alan Zaslavsky, and Hui Zheung. 2006. "The US National Comorbid ity Survey Replication (NCS R): Design and Field Procedures." International Journal of Methods in Psychiatric Research (1 3 ) :69 92. Kim, KiDek, Miriam Beck Cohen and Maria Serakos. 2015. "The Processing and Treat ment of Mental Ill Person s in the Criminal Justice System: A Scan of Practice and Background Analysis." Crime and Justice: Urban Institute Kim, Yujin. 2014. "The Effect of Incarceration on Midlife Health: A Life Course Appr oach." Population Research and Policy Review 34: 827 849. Kinner, Stuart A. and MJ Milloy. 2011. "Collateral Consequences of an Ever Expanding Prison System." Canadian Medical Association. (183)5 632. Lamb, Richard H., Linda E. Weinberger and Bruce H. G ross. 2004. "Mentally Ill Persons in the Criminal Justice System: Some Perspectives." Psychiatric Quarterly. (75)2: 107 126 Lamberti, Steven J. and Robert L. Wiseman. 2004. "Persons with Severe Mental Disorders in the Criminal Justice System: Challenges and Opportunities." Psychiatric Quarterly. (75)2: 151 164. Maruschak, Laura M. 2008. "Medical Problems of Prisoners." Washington, DC: Bureau of Justice Statistics. http://bjs. gov/content/pub/pdf/mpp.pdf. Massoglia, Michael. 2008a. "Incarceration, Healt h, and Racial Disparities in Health." Law and Society Review. (42)2: 275 306.

PAGE 47

! ! '! Massoglia, Michael. 2008b. "Incarceration as Exposure: The Prison, Infectious Disease, and Other Stress Related Illnesses." Journal of Health and Social Behavior. (49) March 56 71. Massoglia, Michael and William Alex Pridemore. 2015. "Incarceration and Health." Annual Review of Sociology. (41) 291 310. Mears, Daniel, Joshua C. Cochran and Francis T. Cullen. 2015. "Incarceration Heterogeneity and Its Implication for Assessing t he Effectiveness on Recidivism." Criminal Justice Policy Review. (26)7: 691 712. Mumola, Christopher J. 2007. Medical Causes of Death in State Prisons, 2001 2004 ." Washington DC: Bureau of Justice Statistics. http://www.bjs.gov/content/pub/pdf/mcdsp04.pd f. Munn, Melissa. 2011. "Living in the Aftermath: The Impact of Lengthy Incarceration on Post Carceral Success." The Howard Journal. (50)3: 233 246. Nagin, Daniel S., Francis T. Cullen and Cheryl Lero Jonson. 2009. "Imprisonment and Reoffending." Crime and Justice. (38)1: 115 200. Nowotny, Kathryn M., 2016. "Social Factors Related to the Utilization of Health Care Among Prison Inmates." Journal of Correctional Health Care. 22(2): 129 138. Nowo tny, Kathryn M., Alice Cepeda, Laurie James Hawkins and Jas on D. Boardman. 2015. "Growing Old Behind Bars: Health Profiles of the Older Inmate Population in the United States." Journal of Aging and Health. 1 22. Pelaez, Vicky. 2008. "The Prison Industry in the United States: Big Business or a New Form of Slavery? Global Research. http://www.globalresearch.ca/the prison industry in the united states big business or a new form of slavery/8289 Pager, Devah. 2003. "The Mark of a Criminal Record." American Journal of Sociology. (108)5: 937 995. Pearlin, Leonard I. 1 989. "The Sociological Study of Stress." Journal of Health and Social Behavior. (30) 3: 241 256. Peters, Roger H., Aurthor J. Lurigio and Harry K. Wexler. 2015. "Co Occurring Substance Use and Mental Disorder in the Criminal Justice System: A New Frontier of Clinical Practice and Research." Psychiatric Rehabilitation Journal. (38)1 1 6 Pew Cent. States. 2008. "One in 100: Behind Bars in America 2008".Washington, DC: Pew Cent. States.

PAGE 48

! ! '$ Porter, Lauren C. 2015. "Incarceration and Post release Health Behavior s." Journal of Health and Social Behavior. (55)2 234 249. Schnittker, Jason and Andrea John. 2007. "Enduring Stigma: The Long Term Effects of Incarceration." Journal of Health and Social Behavior. (48) 2:115 130. Schnittker, Jason, Michael Massoglia, and Christopher Uggen. 2012. "Out and Down: Incarceration and Psychiatric Disorders." Journal of Health and Social Behavior. (53)4: 448 464. Schnittker, Jason and Michael Massoglia. 2015. "A Sociological Approach to Studying the Effects of Incarceration." Wi sconsin Law Review. 350 374. Scott, Susie. 2010. "Revisiting the Total Institution: Performative Regulation in the Reinventive Institution." Sociology. (44)2: 213 231. Sered, Susan and Maureen Norton Hawk. 2008. "Disrupted Lives, Fragmented Care: Illnes s Experiences of Criminalized Women." Women and Health. (48)1: 43 61. Sung, Hung En and Jeff Mellow. 2011. "Criminalization of Disease, medicalization of Justice and Beyond." Corrections Compendium. 11 16. Thoits, Peggy A. 2010. "Stress and Health: Majo r Findings and Policy Implications." Journal of HelathHealth and Social Behavior. (51): S41 S53. Thomas, James C., Brooke A. Lavandowski, Malika Roman Isler, Elizabeth Torrone and George Wilson. 2007. "Incarceration and Sexually Transmitted Infections: A Neighborhood Perspective." Journal of Urban Health: Bulletin of the New York Academy of Medicine. (85)1 90 99 Turney, Kristen, Christopher Wildeman and Jason Schnittker. 2012. As Fathers and Felons: Explaining the Effects of Current and Recent Incarcerat ion on Major Depression. Journal of Health and Social Behavior. 53 (4) 465 481. Wakefield, Sara and Christopher Uggen. 2010. "Incarceration and Stratification." Annual Review of Sociology. (36): 387 406. Wakeman, Sarah E., Margaret E. McKinney and Josiah D. Rich. 2009. "Filling the Gap: The Importance of Medicaid Continuity for Former Inmates." Journal of General Internal Medicine. (24)7: 860 862. Western, Bruce, Anthony A. Braga, Jaclyn Davis and Catherine Sirois. 2015. "Stress and Hardship After Priso n." American Journal of Sociology. (120)5: 1512 1547. Williams, Brie A., James McGuire, Rebecca G Lindsay, Jacques Baillargeon, Ireana Stijacic Cenzar, Sel J. Lee and Margot Kushel. 2010. "Coming Home: Health Status of

PAGE 49

! ! '% Homelessness Risk of Older Pre relea se Prisoners." Journal of General Internal Medicine (25) 10 1038 1044. Yu, Sung suk Violet, Hung En Sung, Jeff Mellow and Carl J. Koenigsmann. 2015. "Self Perceived Health Improvements Among Prison Inmates." Journal of Correctional Health Care. (21)1: 59 6 9.