Elsevier

Journal of Criminal Justice

Volume 42, Issue 6, November–December 2014, Pages 527-537
Journal of Criminal Justice

Dissecting the relationship between mental illness and return to incarceration

https://doi.org/10.1016/j.jcrimjus.2014.09.005Get rights and content

Highlights

  • We examine how mental illness is related to reincarceration.

  • We find a statistically significant relationship between mental illness and recidivism.

  • Substance-related disorders largely account for that relationship.

  • There are important variations by type of mental disorder.

Abstract

Purpose

We examine all releases to parole supervision in a single state over a period of four years to consider how a diagnosis of mental illness is associated with return to incarceration.

Methods

We use survival methods and Cox regression to understand patterns of and influences on return to prison. Our measure of mental illness is based on in-prison clinical diagnoses. Data include a rich set of administrative variables with demographic, criminal history and institutional controls.

Results

Our findings suggest that (1) there is a statistically significant relationship between having a DSM diagnosis and reincarceration, (2) substance-related disorders account for most of that relationship, and (3) there are some important variations among types of disorders examined.

Conclusions

Research that examines mental illness and recidivism without controlling for substance use disorders/problems is likely to be uninformative and misleading. Findings provide qualified support for the notion that programming addressing criminogenic risks and needs may be as important, or more so, than therapeutic programming focusing on mental illness when recidivism reduction is the goal.

Introduction

The American social experiment in mass incarceration has contributed to, if not created, a correctional system that is overburdened, unwieldy, and places overwhelming demands on governmental resources at the expense of other social investments (e.g., Austin and Irwin, 2012, Jacobson, 2005, Raphael and Stoll, 2009, Tonry, 2004). On top of these growing demands is the unfortunate, but inevitable fact that correctional systems are not well-situated to meet the needs of large proportions of those incarcerated (e.g., Acquaviva, 2006, Wilper et al., 2009). One special-needs population that places additional burdens on the correctional system and that the system is not well-suited to help are those with mental illness. Despite this, there is considerable evidence that individuals with mental illness are overrepresented among correctional populations (e.g., Fisher et al., 2006, Frank and McGuire, 2011, Lurigio, 2011, Skeem et al., 2011). As a direct result, increased attention has recently been devoted to the issue of mental disorder and corrections by a variety of governmental and professional organizations (e.g., Bureau of Justice Assistance, 2012, CSG (Council of State Governments), 2012, NIC (National Institute of Corrections), 2012).

The incarceration of mentally ill individuals has reached such a large scale that prisons and jails have been referred to as “America’s New Mental Hospitals” (Torrey, 1995). At year-end 2011, nearly 7.2 million people in the US were under some form of correctional supervision (prison, probation, parole or jail), approximately 1 of every 34 US adults (Carson and Sabol, 2012, Glaze and Parks, 2012, Minton, 2012). Based on estimates derived from a variety of sources (see primarily Ditton, 1999; but also see Fazel and Danesh, 2002, Lurigio, 2001, Lurigio, 2011, Lurigio and Swartz, 2000, Lurigio et al., 2003, Steadman et al., 2009; among others), at any given point in time there are likely to be more than one million seriously mentally ill (SMI) individuals under some form of correctional supervision, and perhaps as many as 3.5 million mentally ill individuals coming in contact with the correctional system in the course of a given year – many of these are repeat contacts and represent multiple opportunities for intervention.2 To put this in context, these numbers are orders of magnitude larger than the nearly 50,000 residents in, and 189,000 admissions to, state and county psychiatric hospitals each year (SAMHSA, 2012; also see Torrey, Kennard, Eslinger, Lamb, & Pavle, 2010).

There is ample evidence that incarcerating the mentally ill, especially in such large numbers entails a variety of social, economic and human costs. The California prison system, for example, which is currently under federal court oversight, provides an exemplary illustration of the potential consequences both for the individuals and the institutions that incarcerate them. Onishi (2013) reports that in 2009, a federal appeals court found that prison overcrowding in California had compromised the delivery of adequate mental and medical health care and the state was mandated to reduce prison populations – an order which the US Supreme Court upheld in 2011 after the state appealed. Currently spending $400 million a year on mental health care in its prisons, and despite building or starting construction on a dozen new treatment facilities valued at a total of $1.2 billion in the past three years, federal courts have reaffirmed that the system does not meet constitutionally mandated standards of care, and have denied the state’s attempt to end federal court oversight. More generally, state prison systems have been estimated to spend $5 billion annually to incarcerate non-violent mentally ill offenders (Frances, 2013).

However, those with mental illness are often excluded from a variety of in-prison services other than direct mental health interventions, and often do not receive mental health or other services when needed (e.g., see Acquaviva, 2006, Austin, 2001, Wilson and Davis, 2006). And despite a strong inclination to provide mental health treatment as an intervention in the form of pre- and post-release services, the record of such services for reducing involvement with criminal justice systems is weak (Epperson et al., 2011, Skeem et al., 2009). Although there are still advocates of the mental health treatment model, a growing chorus is providing both theoretical arguments and empirical evidence suggesting that the mentally ill may benefit from programs that address the criminogenic needs that have robust support in the literature for reducing recidivism (e.g., Andrews and Bonta, 2010, Epperson et al., 2011, Skeem et al., 2009, Wong et al., 2012).

Frank and McGuire (2011) rightly note that mental illness should be a focus in programming considerations for recidivism reduction only if mental disorder is a significant factor in recidivism. And a number of recent analyses do implicate mental illness in recidivism by reporting that mentally-disordered individuals are likely to reoffend or return to prison at higher rates than non-disordered individuals. As an example, in an analysis based on a relatively large sample of prison releases (N = 14,621), Cloyes, Wong, Latimer, and Abarca (2010:182) found that “77% of the SMI (seriously mentally ill) group returned to prison within 36 months, compared to 62% for the non-SMI population”, and the median time to return for SMI offenders was about half that for non-SMI offenders. Grattet, Petersilia, and Lin (2008: 78) showed that parolees with a record of mental health problems violate parole more than non-disordered offenders, with a 70% greater risk for technical violations and a 36% higher risk for all violations. Similarly, Eno Louden and Skeem (2011) reported that mentally ill parolees were twice as likely as non-mentally ill parolees to return to prison within the first year of release. Differences of these magnitudes are substantial, the associations suggest that mental disorder is related to recidivism, and given these findings, one might reasonably conclude that focusing efforts on mental illness poses a prime opportunity to intervene in the cycle of release and return that characterizes US prison populations (Beck and Shipley, 1989, Durose et al., 2014, Langan and Levin, 2002).

However, it is not so clear that mental illness is as strongly implicated in recidivism as these and other findings suggest. Gendreau et al., 1996, Bonta et al., 1998 showed that indicators of mental illness fared poorly when compared to other predictors of criminal recidivism in a meta-analysis of prior studies. Andrews and Bonta (2010) review the psychopathological model and find it limited in its predictive ability. Gagliardi, Lovell, Peterson and Jemelka (2004) observe that mentally ill offenders are no more likely to be reconvicted than the non-mentally ill. And in a recent assessment of mental health based programs, Skeem et al. (2011) indicate that the efficacy of such programs for reducing recidivism is weak. These findings, in contrast to the previous work, suggest that targeting and treating mental illness in order to reduce recidivism may be misdirected.

How does one reconcile these different findings in terms the relationship between mental disorder and recidivism? First, the causal relationship between the two has yet to be established, despite the often-reported associations. Skeem et al., 2011, Skeem et al., 2014, for example, put forth a conceptual framework and compelling argument that there are several potential direct and indirect links between mental illness and recidivism. They begin by suggesting that for a very small subgroup of individuals, mental illness is directly and causally related to criminal behavior. For a second group, mental illness is not associated with differences in criminal behavior, but can trigger a differential response on the part of the criminal justice system which may result in higher recidivism. And finally, they note that indirect evidence supports the notion that the predictive factors for criminal behavior, such as substance use, criminal attitudes and peers, and criminal history, are the same for both mentally ill and non-mentally ill offenders, and that mentally ill individuals tend to have more of these risk factors than do non-disordered individuals. In short, individuals with mental illness are at greater risk of recidivating due to criminogenic risk factors other than mental illness, but that are highly correlated with mental illness.

In addition, many analyses of the association between mental disorder and recidivism treat mental illness as an aggregate or homogenous construct when it is thought to be much more variable. This is potentially problematic for a variety of reasons. Major mood and schizophrenic-psychotic disorders are often considered to signal the greatest impairment or dysfunction, and thus qualify as “Serious Mental Illnesses” (SMI) most likely to be associated with recidivism (e.g., Frank and McGuire, 2011, Lurigio, 2011). Yet, some evidence suggests that schizophrenic and/or psychotic disorders have a negative relationship with recidivism and are actually “protective factors” in terms of criminal conduct or recidivism (Andrews and Bonta, 2010, Quinsey et al., 2005).

Different mental disorders are also thought to have different causal relationships with recidivism and criminal behavior. Individuals with severe depression, for example, are subject to distorted and maladaptive thinking which may lead to dysfunctional, and in some cases, criminal behavior (Ryan & Redding, 2004). Others have suggested that major depression is linked to increased hostility and intense angry outbursts which can lead to violence (Fava et al., 1996). The elevated moods associated with mania have been linked to increased risk-taking and sensation-seeking, traits that have been linked to criminal behavior (Ryan and Redding, 2004, Wood et al., 1997). The hallucinations and delusions associated with schizophrenia and psychotic disorders may be the source of violent outbursts and behavior, and are linked to poor anger management and emotional regulation (Fazel, Langstrom, Hjern, Grann, & Lichtenstein, 2009).

And as is well known, substance use tends to co-occur with mental illness at high rates and has been shown to influence the association between mental illness and crime (Swartz & Lurigio, 2007). Although those diagnosed with major mental disorder (e.g., schizophrenia or major mood disorder) were long thought to be at higher risk for violent behavior, recent epidemiologic studies suggest little increased risk without a corresponding substance use disorder (Fazel et al., 2009, Schubert et al., 2011, Volavka and Citrome, 2008). Substance disorders are thought to influence recidivism in multiple ways. Substance use can lead to impaired impulse control and lowered inhibitions, and feelings (Belenko and Peugh, 1998, Volavka and Citrome, 2008). Volavka and Citrome (2008) show that substance use disorders may also affect treatment non-adherence which may increase the risk of violence among those diagnosed with serious mental illness. Those with substance use disorders may also be more likely to trigger parole revocations through failed drug tests or failure to adhere to other conditions of parole (Skeem et al., 2011). But due to data limitations, many studies are unable to consider the relationship between substance use and mental illness and do not distinguish among the disorders that constitute mental illness. For these reasons among others, studies may often infer a stronger relationship between mental illness and criminal recidivism than is warranted.

In this analysis, we examine the relationship between mental illness (defined as having a diagnosis of a DSM-IV disorder) and return to incarceration, paying attention to the mediating role of substance disorders and the possibility that different mental health diagnoses have different associations with criminal recidivism. In general, our analysis is concerned with exploring several key questions. First, given the mixed evidence that we have already cited, is mental illness, defined as having a DSM diagnosis, related to recidivism, and specifically, reincarceration? Second, given that some evidence suggests variation in the associations between different mental disorders and recidivism, what are the size and directional relationships between different diagnostic categories and return to incarceration? And finally, does the inclusion of substance use disorders mediate the relationship between different disorders and recidivism?

Section snippets

Data and methods

A major limitation of previous studies on the relationship between mental disorder and recidivism is a reliance on relatively small and non-representative samples. For this analysis, we obtained data on all individuals (N = 14,148) released to parole supervision between January 1998 and December 2001 in Tennessee.3

Results

We begin by reviewing basic descriptive statistics for the population under study, disaggregated by mental health status. Table 1 provides details on key demographic, criminal history and institutional variables – all of these variables have been identified as relevant to criminal recidivism and we reference key aspects of that literature in our discussion of the table. Statistically significant differences exist for all three of the core demographic indicators. SMIs are slightly more than two

Discussion and conclusions

Using a unique data set that consists of the entire population of state prison releases to parole supervision over a four year period, and which includes in-prison clinical diagnoses of mental disorders, our analyses and results suggest the following key conclusions. First, our findings, with reincarceration as the outcome variable, adds support to the recent literature suggesting that most measures of mental illness absent an indicator of substance use disorder have no statistically

James A Wilson is Senior Program Officer at the Russell Sage Foundation in New York where he directs the Foundation’s program in Social Inequality. His research interests are focused on offender reentry, the policies and institutions that facilitate or impede successful return to the community, and how processes of educational attainment have changed over time.

References (88)

  • J. Austin et al.

    It’s About Time: America’s Imprisonment Binge

    (2012)
  • W.D. Bales et al.

    Inmate social ties and the transition to society: Does visitation reduce recidivism?

    Journal of Research in Crime and Delinquency

    (2008)
  • A.J. Beck et al.

    Recidivism of Prisoners Released in 1983

    NCJ 116261

    (1989)
  • S. Belenko et al.

    Fighting Crime by Treating Substance Abuse

    Issues in Science and Technology Online

    (1998)
  • A.S. Bhati et al.

    To Treat or Not to Treat: Evidence on the Prospects of Expanding Treatment to Drug-Involved Offenders

    (2008)
  • R. Blackburn

    “What Works” with mentally disordered offenders

    Psychology Crime and Law

    (2004)
  • H.P. Blossfeld et al.

    Event History Analysis: Statistical Theory and Application in the Social Sciences

    (1989)
  • A. Blumstein et al.

    Population growth in U.S. prisons, 1980-1996

    Crime and Justice: A Review of Research

    (1999)
  • J. Bonta et al.

    The prediction of criminal and violent recidivism among mentally disordered offenders: A meta-analysis

    Psychological Bulletin

    (1998)
  • Bureau of Justice Assistance

    Justice and Mental Health Collaboration Program (JMHCP)

    (2012)
  • E.A. Carson et al.

    Prisoners in 2011. NCJ 239808

    (2012)
  • T.R. Clear et al.

    The transition from being an inmate: Discharge planning, parole, and community-based services for mentally ill offenders

  • K.G. Cloyes et al.

    Time to prison return for offenders with serious mental illness after release from prison

    Criminal Justice and Behavior

    (2010)
  • CSG (Council of State Governments)

    Mental Health Courts: A Primer for Policymakers and Practitioners

    (2008)
  • CSG (Council of State Governments)

    Criminal Justice – Mental Health Consensus Project

  • P.M. Ditton

    Mental Health and Treatment of Inmates and Probationers

    Bureau of Justice Statistics Special Report, NCJ 174463

    (1999)
  • M.R. Durose et al.

    Recidivism of Prisoners Released in 30 States in 2005: Patterns from 2005 to 2010. NCJ 244205

    (2014)
  • E.B. Elbogen et al.

    The intricate link between violence and mental disorder: Results from the National Epidemiological Survey on alcohol and related conditions

    Archives of General Psychiatry

    (2009)
  • J. Eno Louden et al.

    Parolees with Mental Disorder: Toward Evidence-based Practice

    Bulletin of the Center for Evidence-Based Corrections

    (2011)
  • M. Epperson et al.

    The Next Generation of Behavioral Health and Criminal Justice Interventions: Improving Outcomes by Improving Interventions

    (2011)
  • Farrell, H. M. (2013). Is DSM-V Worth All the Hype?. Psychology Today. Retrieved July 5, 2013 from...
  • S. Fazel et al.

    Schizophrenia, substance abuse, and violent crime

    Journal of the American Medical Association

    (2009)
  • T.E. Feucht et al.

    Mental and substance use disorders among adult men on probation or parole: Some success against a persistent challenge

    SAMHSA Data Review

    (2011)
  • W.H. Fisher et al.

    Beyond criminalization: Toward a criminologically informed framework for mental health policy and services research

    Administrative Policy Mental Health and Mental Health Services Research

    (2006)
  • Frances, A. J. (2013). Prison or treatment for the mentally ill. Psychology Today. Retrieved April 4, 2013 from...
  • R.G. Frank et al.

    Mental health treatment and criminal justice outcomes

  • G.J. Gagliardi et al.

    Forecasting Recidivism in Mentally Ill Offenders Released from Prison

    Law and Human Behavior

    (2004)
  • P. Gendreau et al.

    A meta-analysis of the predictors of adult offender recidivism: What works!

    Criminology

    (1996)
  • L.E. Glaze et al.

    Correctional Population in the United States, 2011

    BJS Bulletin, NCJ 239972

    (2012)
  • R. Grattet et al.

    Parole Violations and Revocations in California

  • P. Guerino et al.

    Prisoners in 2010

    BJS Bulletin, NCJ 236096

    (2011)
  • M. Jacobson

    Downsizing Prisons: How to Reduce Crime and End Mass Incarceration

    (2005)
  • D.J. James et al.

    Mental health problems of prison and jail inmates

  • R.P. Jemelka et al.

    The mentally ill in prison: A review

    Hospital and Community Psychiatry

    (1989)
  • Cited by (0)

    James A Wilson is Senior Program Officer at the Russell Sage Foundation in New York where he directs the Foundation’s program in Social Inequality. His research interests are focused on offender reentry, the policies and institutions that facilitate or impede successful return to the community, and how processes of educational attainment have changed over time.

    Peter B. Wood is Professor in the Department of Sociology, Anthropology, and Criminology at Eastern Michigan University. His research includes the study of factors that motivate and maintain habitual offending, and issues associated with correctional policy and practice. His work has appeared in Criminology, Justice Quarterly, Journal of Research in Crime and Delinquency, Punishment and Society, The Prison Journal, Crime and Delinquency, Journal of Offender Rehabilitation, Journal of Criminal Justice, and Deviant Behavior.

    Direct all correspondence to James Wilson, Russell Sage Foundation, [email protected]. Prior versions of this manuscript were presented at the 2011 Academy of Criminal Justice Sciences meetings in Toronto, Canada, the 2011 American Society of Criminology meetings in Washington, DC, and the 2012 American Sociological Association meetings in Denver, CO. We also thank Jim Parsons @ the Vera Institute, and the anonymous reviewers who provided excellent comments and advice on an earlier version of this manuscript.

    1

    Tel.: + 1 734 487 8099.

    View full text