Successful Reintegration and Mental Health: An Examination of Gender Differences Among Reentering Offenders

Abstract

Men and women exiting the correctional system represent a population at high risk for mental health problems, and the body of research on the mental health needs of former prisoners is growing. These mental health problems pose challenges for individuals at every stage of the criminal justice process, from arrest to incarceration to reentry and reintegration. This article examines the mental health status and gender differences among a sample of 352 men and women leaving confinement and the role that mental health problems played in shaping their reentry outcomes using data collected between 2002 and 2005. In the year after leaving prison, men and women with mental health problems reported worse health indicators and less satisfactory social factors, such as employment, housing, and family support. The article concludes with a discussion of recommendations for improved policy and practice for assisting former prisoners with mental health problems during reintegration.

Keywords

mental health gender prisoner reentry recidivism

A recent report by the Bureau of Justice Statistics documents significant mental health problems for men and women being held in U.S. prisons and jails. The vast majority of these individuals are incarcerated for an average of 2 to 3 years and then are released to the community to face a myriad of challenges. The mental health status of persons involved in the criminal justice system is receiving increased attention as criminal justice professionals from law enforcement to probation officers struggle with limited options for managing individuals with mental health problems. Greater awareness of the impact of these mental health problems may assist policy makers and practitioners in developing solutions. In this article, we use data gathered from men and women leaving correctional facilities before and after their release to examine their mental health status and the role that mental health problems play in shaping a variety of reentry outcomes.

Background

Individuals exiting the correctional system whether from prison or jail represent a population at high risk for mental health problems, and a growing body of research exists on their mental health needs. Data collected from the Bureau of Justice Statistics of more than 100,000 prisoners and jail inmates revealed that half of prisoners and 64% of jail inmates either reported symptoms of serious psychological distress or had been told in the past by a mental health professional that they had a mental disorder (Bronson & Berzofsky, 2017). A larger percentage of women than men reported serious psychological distress in the last 30 days. The most commonly reported diagnoses were major depressive order, bipolar disorder, personality disorder, or posttraumatic stress disorder (PTSD). Also, those with some indication of mental health problems were more likely to be unmarried, to be incarcerated for a violent or property crime, and somewhat more likely to have some college or a college degree (Bronson & Berzofsky, 2017). The percentage of prisoners who reported current serious psychological distress (14%) was more than 3 times that of adults in the U.S. general population (5%; Bronson & Berzofsky, 2017). Moreover, prisoners’ self-reports of a mental disorder of over 40% are significantly higher than the estimated 25% of U.S. adults with a mental illness (Centers for Disease Control and Prevention [CDC], 2011). Thus, it is clear that men and women in the criminal justice system have substantial rates of mental health problems.

Given the high rates of behavioral health problems among men and women leaving prison, the need for treatment during and after incarceration for those that experience mental health problems is essential. Yet, in-prison services do not appear to be sufficient to meet treatment needs. In fact, between 1988 and 2000, prison mental health services declined and those services that are available are concentrated in the most secure facilities (Manderscheid, Gravesande, & Goldstrom, 2004). In the national Bureau of Justice Statistics survey, only one third of inmates with mental health problems were currently receiving treatment (Bronson & Berzofsky, 2017). In a multi-site longitudinal study of men and women exiting prison, 41% of women and 18% of men reported receiving some type of mental health treatment while incarcerated, but one quarter of men to over half of women felt that they currently needed treatment (Broner, Lattimore, & Steffey, 2010).

Many of the men and women with mental health problems confined to jail or prison have only a short stay in prison or jail and are then released to the community, which may involve some type of postrelease supervision. Unfortunately, many do not make a successful transition. Among men released from prison from 30 states in 2005, 69% were arrested for a new crime within 3 years, and among women, 58% were arrested within 3 years (Durose, Cooper, & Snyder, 2014). Moreover, about half were returned to prison within 3 years either for a conviction or parole violation. This revolving door phenomenon is costly in terms of criminal justice system expenditures, not to mention the human and social costs borne by crime victims, returning prisoners, their families, and the communities in which they reside. What is less well-known is whether those with mental health problems are at a greater likelihood of arrest or reincarceration after release, whether they face greater reentry challenges, and if differential outcomes exist among men or women with mental health conditions (see Snedker, Beach, & Corcoran, 2017).

These mental health problems pose problems for people at every stage of the criminal justice process, from arrest to incarceration to community reentry and reintegration. Individuals leaving prison face multiple, often simultaneous tasks as they embark on the process of reestablishing their lives outside prison—finding housing, getting a job, earning enough money to support themselves, reconnecting with children and family—and these intermediary steps influence the ability to live a drug-free and crime-free life. Unresolved mental health and substance use problems likely complicate an already challenging transition (Harding & Roman, 2016; Mallik-Kane & Visher, 2008; Travis, 2005). Returning prisoners with health problems may be unable to engage in work or other activities, and their families may be unwilling or unable to support them. They are additionally confronted with the tasks of managing their health problems, such as accessing health care and keeping up with medications and appointments. Individuals with severe or unmanaged health problems face an increased risk of adverse outcomes, including physical illness, relapse into drug use, or, particularly in the case of mental illness, inappropriate behavior that provokes a law enforcement response (Gido & Dalley, 2008; Hammett, Roberts, & Kennedy, 2001).

It stands to reason that successful treatment of returning prisoners’ health conditions could increase their chances of reentry success by improving their ability to work, support themselves, and abstain from substance use, all of which have been shown to contribute to desistance from criminal activity. However, after release, receipt of mental health treatment services declined among those who reported mental health problems in prison and remained low throughout a 15-month follow-up period in one study (Broner et al., 2010). Continuity of prescription medication also often suffers after release, especially among women (Mallik-Kane & Visher, 2008), due to treatment access and lack of insurance.

Reentry scholars acknowledge that providing health care to men and women returning to the community can have profound public health and cost impacts on formerly incarcerated individuals, their families, and the communities to which they return (Greifinger, 2007). Research is essential to identify the specific health care problems of this population and the services they need (Mears & Cochran, 2012). Proper assessment of this need may require new screening tools. Sociodemographic variations in prevalence may suggest the importance of targeting resources more intensively to specific locations or subgroups. Ultimately, this line of research may reveal strategies to reduce the overrepresentation of individuals with mental health problems in prison.

Thus, mental health problems are a serious issue for men and women as they exit prison, and given the lack of community treatment and other social, cultural, and medical problems they face upon reentry, the chances of encountering significant reentry challenges, including recidivism, are high. The impact of mental health problems on reentry outcomes among women leaving prison has received some research attention (e.g., Bloom, Owen, & Covington, 2003; Mallik-Kane & Visher, 2008; Visher & Bakken, 2014). However, few studies have compared men’s and women’s experiences after release and the role of mental health in their reentry outcomes. It is widely recognized that the reentry experiences of men and women differ on important dimensions (Spjeldnes & Goodkind, 2009; Van Voorhis, Wright, Salisbury, & Bauman, 2010). Yet, historically, prison rehabilitation and community reentry programs have focused on men’s needs and have not been gender focused, although this is changing (Cobbina, Huebner, & Berg, 2012; Wright, Van Voorhis, Salisbury, & Bauman, 2012; see also, National Resource Center on Justice Involved Women [NRCJIW], 2018). As mentioned previously, mental health concerns emerge as a prominent concern for reentering women. In this study, we specifically examine the role of mental health in reentry pathways by comparing the experiences of men and women, controlling for other reentry risk factors.

In the sections that follow, a description of men and women leaving prison in the study sample is presented, along with a comparison of reentry experiences of men and women with and without mental health problems. Following this discussion, we use multivariate logistic regression to examine the impact of self-reported mental health while in prison on housing, employment, substance use, and recidivism at 9 to 12 months following release, controlling for several demographic, family, and criminal history indicators. We pay particular attention to gender differences in the impact of mental health on these reentry outcomes. The article concludes with a discussion of the role of mental health problems among men and women leaving prison in the reintegration process and offers empirically based recommendations for improved policy and practice.

Method

Sample and Data Collection

The data for this project come from the Urban Institute’s Returning Home project, a multistate, longitudinal study designed to explore the process of prisoner reentry, the challenges that returning prisoners and their families face, and the pathways to successful reintegration (La Vigne & Kachnowski, 2005; Visher, 2007). The data gathered through Returning Home provided a comprehensive look at individuals’ life circumstances prior to and during prison, and over the course of a year after release from prison. Roughly 30 days prior to release, respondents completed a self-administered survey examining areas related to dimensions of family life, physical and mental health, and experiences and behaviors preincarceration. The prerelease samples of prisoners were generally representative of all prisoners being released to the study areas in the previous year in terms of race, sentence length, and time served. After each prisoner’s release, experienced interviewers conducted up to three personal interviews within 15 months, including interviews with those who were in jail or had been returned to prison. The current data are comprised of male and female offenders incarcerated in the state of Texas, which includes a total of 676 participants (414 males and 262 females). Only respondents who had completed both the prerelease and the last postrelease interview (roughly 12 months postrelease) were included in the current study, providing a final sample of 352 respondents (210 males, 142 females).¹

Dependent Variables

The dependent variables in this project included four dichotomous measures that examined both positive and negative experiences of reintegration as reported at the last postrelease interview. First, housing was measured by asking respondents if they had trouble keeping housing since their release from prison (1 = had trouble keeping housing; 0 = did not have trouble keeping housing). Roughly 20% of respondents reported having trouble maintaining housing upon release from prison. Second, employment was measured using the average number of hours the respondent reported working per week in the last 30 days. This was dichotomized to specify if the respondent had any legal employment at the postrelease interview. Roughly 55% of all respondents reported being employed in the month prior to the last postrelease survey and worked an average of 24 hr a week. Third, drug use was measured by reports of any illegal drug use in the past 30 days. Finally, criminal activity was measured using a dichotomous indicator of any reports of 10 different crimes committed by the respondent since release (1 = any criminal activity, 0 = no criminal activity). Thirty percent of respondents reported past month drug use, while 24% reported any criminal behavior since release. To maintain consistency throughout these dichotomous measures, items were coded such that higher scores were indicative of an affirmative orientation to the question. The descriptive statistics for these and other variables are presented in Table 1.

Table 1:

Descriptive Statistics

Variable	M	SD	Minimum	Maximum	Factor loading
Dependent variables
Housing trouble	0.22	0.41	0.00	1.00	—
Employment (in hours)	24.34	25.11	0.00	84.00	—
Drug use	0.30	.46	0.00	1.00	—
Criminal behavior	0.24	0.43	0.00	1.00	—
Independent variables
Mental health condition	0.19	0.39	0.00	1.00	—
Family Support (scale)	3.37	0.60	1.00	4.00	—
Tangible Support (scale)	3.26	0.65	1.00	4.00	—
Family Conflict (scale)	2.11	0.67	1.00	4.00	—
Lived with kids	0.29	0.27	0.00	1.00	—
Cohabitate	0.27	0.44	0.00	1.00	—
Prior convictions	3.14	3.67	0.00	19.00	—
HS/GED	0.59	0.49	0.00	1.00	—
Age	36.20	9.93	18.00	67.00	—
Prison vs. Jail	0.46	0.50	0.00	1.00	—
Bad neighborhood factors
Neighborhood is safe	1.89	0.83	1.00	4.00	.74
Hard to stay out of trouble	1.98	0.87	1.00	4.00	.79
Nervous seeing people	1.92	0.87	1.00	4.00	.67
Problem with drug selling	2.27	1.04	1.00	4.00	.79
Hard to stay out of prison	1.86	0.89	1.00	4.00	.79
Deviant peers factor
Friends in prison	1.07	0.84	0.00	3.00	.73
Friends committed theft	1.20	0.94	0.00	3.00	.80
Friends committed assault	0.87	0.84	0.00	3.00	.77
Friends use illegal drugs	1.52	0.99	0.00	3.00	.82
Friends sold drugs	1.09	0.93	0.00	3.00	.82

Note. N = 352. Percentages are based on nonmissing responses to each item. All variables are 0,1 except where noted. Employment is reported in hours per week for descriptive purposes and was dichotomized for analyses. HS = high school; GED = general educational development.

Independent Variables

To gain a better understanding of how mental health and pre- and postrelease experiences are related to important reentry outcomes, several independent variables were incorporated that can broadly be categorized into domains related to the social and personal characteristics of the respondents both before and after release from prison. To account for the status of respondent mental health, a self-report measure from the prerelease survey was used regarding whether the individual had ever been diagnosed with depression or another mental health condition by a doctor or nurse.² Depression and PTSD were estimated using symptom scales that were included into the baseline interview (Foa, Riggs, Dancu, & Rothbaum, 1993; Radloff, 1991). “Any self-reported or estimated mental illness” was comprised of those who reported a mental health diagnosis or reported symptomatology consistent with the diagnostic criteria for depression or PTSD. To establish the varying role that mental health may have on reentry outcomes for both men and women, a parsimonious set of independent variables measuring contextual and social support constructs from previous reentry research was selected.

Five multi-item scales measured at the final postrelease interview related to the family, neighborhood, and peer influences were included that are known to be linked to dimensions of a successful reentry experience. The family-related scales include Family Support (feeling close to your family, wanting family to be involved in your life, considering yourself a source of support for family, and considering family a source of support), Family Tangible Support (someone in family provides help/advice on finding a place to live, provides help finding a job, support for dealing with a substance abuse problem, provide transportation to work or appointments, provide financial support), and Family Conflict (fighting a lot, disappointing family, being criticized a lot). Two additional multi-item scales captured both Neighborhood Disorder (neighborhood is a safe place to live, hard to stay out of trouble in neighborhood, nervous about seeing certain people in neighborhood, living in neighborhood makes it hard to stay out of prison, drug selling is a major problem in neighborhood) and Antisocial Peer Influences (of your close friends, how many have ever been in prison, committed a theft, assaulted someone, used illegal drugs, sold illegal drugs). All scales have moderately strong reliability (α > .70).

Finally, multiple demographic and control measures known to be associated with reentry outcomes were used, including if the respondent was married or living as married, had lived with their child(ren) prior to imprisonment, the number of prior convictions, if the respondent had earned their high school diploma or GED prior to incarceration, age (in years), and whether they served their time in a state prison or a state jail (prison = 1).³ As mentioned earlier, these measures were correlated with mental health status in the nationwide survey of prisoners and jail inmates conducted by the Bureau of Justice Statistics.

Analytic Strategy

The current study seeks to examine the difficulties, similarities, and differences that former male and female prisoners with mental health problems face upon their release from incarceration. First, bivariate relationships were examined to understand and illustrate the broad range of social, criminal, and health-related problems associated with reentry based on the mental health status of these men and women. Second, separate multivariate logistic regression models were estimated to individually evaluate the influence of selected independent variables documented to affect reentry on multiple outcomes commonly associated with a successful reentry experience.⁴ Logistic regression is suitable for the present inquiry because the dependent variables were assigned a dichotomous measurement. Logistic regression is also useful because it provides an odds ratio (OR) to gauge the effect size of the independent variables. The data were weighted to adjust for sample attrition and all the independent variables were concurrently entered into the models and all measures were retained. Finally, the test for the equality of regression coefficients (Paternoster, Brame, Mazerolle, & Piquero, 1998) was used to test the significance of the difference between the regression coefficients in the male and female models.⁵

Results

Results from the bivariate cross-tabulation analysis (Table 2) demonstrate the many difficulties individuals with mental health problems face that complicate success upon reentry. While similar trends appear for both men and women, some differences should be noted across multiple domains related to health and social factors and criminal involvement.⁶ When examining health-related factors, both men and women with a mental health condition were significantly more likely to report experiencing poor health, hallucinations, suicidal thoughts, and trouble with violence. Women, however, were also roughly 3 times as likely to report being hospitalized upon release, while no significant difference existed for the men.

Table 2:

Reentry Experiences of Male and Female Offenders by Mental Health Status (%)

	Males		Females
Variable	No MH condition	MH condition	No MH condition	MH condition
Health factors
Poor health	25.7	44.8**	25.3	44.9**
Hospitalization	15.3	16.7	11.5	30.6**
Hallucinations	4.0	26.7**	2.3	26.5**
Suicidal thoughts	1.0	20.0**	3.5	33.3**
Trouble with violence	5.1	20.0**	4.7	30.6**
Social factors
Employment	63.4	43.3**	39.1	22.4**
Family support	68.8	50.0	73.9	66.7
Housing trouble	19.0	26.7	23.0	39.6**
Criminal involvement
Substance use	26.3	38.7	27.6	39.6
Criminal behavior	21.1	33.3	20.5	43.8**
Arrest	24.4	32.1	26.7	42.9**
Reincarceration	13.8	10.7	24.1	29.8

Note. MH = mental health.

Differences with a p value ≤.05.

In terms of social factors, men with mental health conditions were significantly less likely to report being employed postrelease; however, significant differences did not exist in terms of obtaining financial support from family members and in the ability to find and maintain adequate housing. Women with mental health conditions were similarly significantly less likely to report postrelease employment; however, they were also more likely to report having trouble finding housing with no significant differences being found in terms of familial financial support.

Interestingly, while men with mental health problems were more likely to report substance use, criminal engagement, and rearrest, none of these relationships reached the level of significance. Women with mental health problems, however, were more likely to report substance use, criminal behavior, being rearrested, and reincarceration, while only criminal behavior and being arrested reached the level of significance. Notably, women with mental health problems were twice as likely to report criminal behavior and roughly 1.5 times more likely to be rearrested compared with those without mental illness.

The results from the multivariate logistic regression analyses examining housing for men and women are presented in Table 3. For men, mental health (OR = 1.04, 95% confidence interval (CI) = [0.32, 3.28]) did not demonstrate a significant effect on their ability to obtain housing; however, men receiving tangible familial support (OR = 0.40, 95% CI = [0.17, 0.93]) were significantly less likely to have trouble finding housing. Having a mental health condition (OR = 3.05, 95% CI = [1.04, 8.97]) was a significant predictor of reported difficulty in finding housing for women, where these women were roughly 3 times more likely to report having trouble finding a suitable place to live upon release than those without a mental health problem. Increased family conflict (OR = 2.94, 95% CI = [1.14, 7.58]) also significantly inhibited the ability to find adequate housing for women, while those women who lived with their children prior to their incarceration (OR = 0.17, 95% CI = [0.04, 0.69]) had significantly less difficulty. Overall, the male and female models explained 28% and 42% of the variance in housing, respectively.

Table 3:

Logistic Regression Results Examining Trouble Finding Housing

	Males			Females
Variable	b	SE	Odds ratio	b	SE	Odds ratio
Mental health condition	0.04	.59	1.04 [0.32, 3.28]	1.12	.55	3.05 [1.04, 8.97]**
Employed	−0.40	.47	0.67 [0.27, 1.70]	−1.39	.84	0.25 [0.05, 1.30]
Family support	0.21	.48	1.24 [0.48, 3.16]	−0.40	.59	0.67 [0.21, 2.15]
Tangible support	−0.91	.43	0.40 [0.17, 0.93]**	0.44	.59	1.56 [0.49, 4.95]
Family conflict	0.54	.40	1.73 [0.78, 3.81]	1.10	.48	2.94 [1.14, 7.58]**
Bad neighborhood	0.25	.32	1.28 [0.69, 2.37]	−0.13	.32	0.88 [0.47, 1.63]
Deviant peers	0.38	.27	1.46 [0.86, 2.45]	0.30	.25	1.35 [0.83, 2.22]
Lived with kids	−0.26	.56	0.77 [0.26, 2.33]	−1.78	.72	0.17 [0.04, 0.69]**
Cohabitate	−0.33	.57	0.72 [0.24, 2.12]	0.33	.71	1.39 [0.34, 5.60]
Prior convictions	0.07	.07	1.07 [0.94, 1.23]	−0.02	.07	0.98 [0.86, 1.14]
High school/GED	−0.83	.46	0.44 [0.18, 1.08]	−1.01	.58	0.36 [0.12, 1.13]
Age	0.02	.02	1.02 [0.97, 1.06]	0.02	.03	1.02 [0.96, 1.07]
Prison vs. Jail	−0.42	.48	0.65 [0.26, 1.66]	−1.10	.70	0.33 [0.08, 1.33]
Nagelkerke R²	.28			.42

Note. Values are given as odds ratio [95% confidence interval].

p ≤ .05.

Overall, there were few significant predictors of postrelease employment for both men and women (see Table 4). However, having a mental health condition was a significant predictor of employment for both men (OR = 0.32, 95% CI = [0.12, 0.84]) and women (OR = 0.28, 95% CI = [0.09, 0.90]). Specifically, men and women reporting a mental health condition were less likely to report being employed at 1 year after release from incarceration. In addition, men who spent time in a state prison (OR = 3.95, 95% CI = [1.85, 8.41]) were significantly more likely to report postrelease employment compared with those serving their time in a state jail. Among women who had earned their high school diploma or GED (OR = 3.67, 95% CI = [1.21, 9.10]), they were more than 3 times more likely to report postrelease employment. The male and female models explained 25% and 35% of the variance in employment, respectively.

Table 4:

Logistic Regression Results Examining Postrelease Employment

	Males			Females
Variable	b	SE	Odds ratio	b	SE	Odds ratio
Mental health condition	−1.14	.49	0.32 [0.12, 0.84]**	−1.27	.59	0.28 [0.09, 0.90]**
Family support	0.16	.38	1.18 [0.55, 2.50]	−0.33	.68	0.72 [0.19, 2.73]
Tangible support	0.22	.35	1.25 [0.63, 2.49]	0.25	.68	1.28 [0.34, 4.88]
Family conflict	0.54	.34	1.72 [0.89, 3.32]	−0.53	.40	0.59 [0.27, 1.30]
Bad neighborhood	−0.41	.24	0.67 [0.42, 1.07]	−0.41	.30	0.67 [0.37, 1.19]
Deviant peers	0.11	.22	1.12 [0.73, 1.72]	−0.44	.28	0.64 [0.37, 1.12]
Lived with kids	0.05	.46	1.05 [0.42, 2.60]	0.92	.58	2.51 [0.80, 7.84]
Cohabitate	0.76	.48	2.15 [0.84, 5.50]	0.74	.64	2.10 [0.60, 7.37]
Prior convictions	0.09	.07	1.09 [0.96, 1.24]	0.09	.08	1.09 [0.93, 1.28]
High school/GED	0.40	.39	1.49 [0.69, 3.19]	1.30	.57	3.67 [1.21, 9.10]**
Age	−0.03	.02	0.98 [0.94, 1.01]	0.02	.03	1.02 [0.96, 1.08]
Prison vs. Jail	1.37	.39	3.95 [1.85, 8.41]**	0.63	.55	1.88 [0.64, 5.47]
Nagelkerke R²	.25			.35

Note. Values are given as odds ratio [95% confidence interval].

p ≤ .05.

The results examining postrelease substance use are presented in Table 5. Surprisingly, while both men and women with mental health conditions were more likely to report substance use, these effects did not reach the level of significance in multivariate analyses. For men, those receiving tangible support from their families (OR = 2.26, 95% CI = [1.04, 4.92]) and those who reported higher levels of neighborhood disorder (OR = 1.71, 95% CI = [1.05, 2.79]) and antisocial peer influences (OR = 1.86, 95% CI = [1.18, 2.95]) were significantly more likely to report postrelease substance use. Among women, employment (OR = 0.16, 95% CI = [0.03, 0.84]), increased family support (OR = 0.19, 95% CI = [0.05, 0.75]), and older age (OR = 0.94, 95% CI = [0.88, 1.00]) were significantly associated with a lower likelihood of substance use, while women residing in areas of high neighborhood disorder (OR = 2.18, 95% CI = [1.16, 4.10]) were more likely to report substance use. The male and female models explain 27% and 50% of the variation in postrelease substance use.

Table 5:

Logistic Regression Results Examining Substance Use

	Males			Females
Variable	b	SE	Odds ratio	b	SE	Odds ratio
Mental health condition	0.56	.52	1.75 [0.63, 4.86]	0.70	.56	2.00 [0.66, 6.06]
Employed	0.03	.43	1.03 [0.44, 2.41]	−1.84	.85	0.16 [0.03, 0.84]**
Family support	−0.29	.43	0.75 [0.32, 1.73]	−1.68	.71	0.19 [0.05, 0.75]**
Tangible support	0.82	.39	2.26 [1.04, 4.92]**	0.38	.65	1.46 [0.41, 5.20]
Family conflict	0.23	.34	1.26 [0.64, 2.48]	−0.15	.43	0.87 [0.38, 2.00]
Bad neighborhood	0.54	.25	1.71 [1.05, 2.79]**	0.78	.32	2.18 [1.16, 4.10]**
Deviant peers	0.62	.23	1.86 [1.18, 2.95]**	0.36	.27	1.43 [0.84, 2.43]
Lived with kids	−0.49	.53	0.61 [0.22, 1.75]	−0.46	.65	0.63 [0.18, 2.26]
Cohabitate	−0.26	.49	0.77 [0.29, 2.03]	−0.64	.77	0.53 [0.12, 2.38]
Prior convictions	−0.05	.06	0.95 [0.84, 1.07]	−0.10	.11	0.91 [0.73, 1.13]
High school/GED	−0.53	.42	0.59 [0.26, 1.34]	−0.18	.55	0.83 [0.28, 2.46]
Age	−0.03	.02	0.97 [0.93, 1.01]	−0.06	.03	0.94 [0.88, 1.00]**
Prison vs. Jail	−0.48	.42	0.62 [0.27, 1.42]	−0.20	.64	0.82 [0.23, 2.87]
Nagelkerke R²	.27			.50

Note. Values are given as odds ratio [95% confidence interval].

p ≤ .05.

The models examining criminal behavior are presented in Table 6. While the mental health status of men did not adversely affect their criminal behavior, women reporting mental health conditions (OR = 4.20, 95% CI = [1.41, 12.5]) were over 4 times more likely to report criminal engagement postrelease than those without a mental health condition, a significant effect. Increased familial conflict (OR = 4.75, 95% CI = [2.10, 10.7]) and antisocial peer influences (OR = 2.12, 95% CI = [1.24, 3.58]) were significant predictors of criminal behavior for men. Among women, those receiving more tangible support from their family (OR = 3.73, 95% CI = [0.99, 14.1]) were unexpectedly more likely to report criminal behavior, as were women reporting more neighborhood disorder (OR = 2.07, 95% CI = [1.08, 3.94]). The models explain 36% and 38% of the variation in male and female criminal behavior, respectively.

Table 6:

Logistic Regression Results Examining Criminal Behavior

	Males			Females
Variable	b	SE	Odds ratio	b	SE	Odds ratio
Mental health condition	−0.11	.55	0.89 [0.30, 2.63]	1.43	.56	4.20 [1.41, 12.5]**
Employed	−0.56	.46	0.57 [0.23, 1.43]	−1.31	.85	0.27 [0.05, 1.42]
Family support	−0.42	.45	0.66 [0.27, 1.59]	−1.08	.65	0.34 [0.10, 1.23]
Tangible support	0.29	.42	1.35 [0.60, 3.08]	1.32	.68	3.73 [0.99, 14.1]**
Family conflict	1.56	.42	4.75 [2.10, 10.7]**	−0.25	.43	0.78 [0.34, 1.82]
Bad neighborhood	−0.30	.29	0.74 [0.42, 1.31]	0.73	.33	2.07 [1.08, 3.94]**
Deviant peers	0.75	.27	2.12 [1.24, 3.58]**	0.18	.27	1.20 [0.70, 2.04]
Lived with kids	−1.17	.66	0.31 [0.84, 1.14]	−0.81	.66	0.45 [0.12, 1.62]
Cohabitate	0.01	.55	1.01 [0.34, 3.01]	−0.87	.76	0.42 [0.10, 1.87]
Prior convictions	0.03	.07	1.03 [0.90, 1.19]	−0.09	.10	0.91 [0.74, 1.12]
High school/GED	−0.41	.45	0.67 [0.28, 1.61]	0.03	.54	1.03 [0.35, 3.00]
Age	−0.02	.02	0.95 [0.94, 1.02]	−0.02	.03	0.98 [0.88, 1.00]
Prison vs. Jail	−0.06	.46	0.94 [0.38, 2.34]	−1.05	.65	0.35 [0.10, 1.26]
Nagelkerke R²	.36			.38

Note. Values are given as odds ratio [95% confidence interval].

p ≤ .05.

Discussion

The results of this study fill a gap in our understanding about the longitudinal impact of mental health on reentry outcomes, including housing, employment, substance use, and criminal activity. The study demonstrated that mental health problems are common among men and women released from incarceration, with 15% of men and 36% of women reported ever being diagnosed with a mental health condition. Men and women with a history of mental illness experienced greater health problems after release, including general poor health, hallucinations, suicidal thoughts, and trouble with violence. Somewhat surprisingly, despite experiencing greater challenges and reporting more criminal behavior, men and women leaving prison with mental health conditions were not significantly more likely to report substance use or to be reincarcerated at higher rates during the first year after release.

Multivariate analyses showed that mental health problems affected three of four reentry outcomes for women (housing, employment, and criminal behavior), but only one reentry outcome for men (employment), after controlling for individual and social characteristics. Thus, a major and important finding of this study is that women with mental health problems are at greater risk for poor reentry outcomes than men. This finding supports suggestions in prior research that mental health problems are a particularly salient concern for women returning to the community (Wright et al., 2012). The implications of this finding are discussed shortly. Mental health problems did not affect substance use after release, possibly because substance use was rather common among those without a mental health condition (26%-40%). The models explain roughly 25% to 50% of the variance in the four reentry outcomes.

Among other findings, the personal characteristics that affected reentry outcomes varied for both men and women across all outcomes. Social factors were much more common in predicting reentry outcomes. For men, the two recurring characteristics were family tangible support and associating with deviant peers. For women, multiple social factors appeared in the models explaining reentry outcomes, notably recent employment, family conflict, family emotional support, family tangible support, and living in a bad neighborhood. For men, whether they were serving time in a prison or jail affected postrelease employment, but was not associated with the other outcomes. Family factors were more likely to emerge as predictors of reentry outcomes for women than for men, which is consistent with reentry research (e.g., Cobbina et al., 2012). However, family tangible support was particularly important in predicting housing trouble and substance use among men, possibly because men’s identity is often tied to financial security.

The differential outcomes for women with mental health problems compared with their male counterparts are both important and noteworthy. While contact with the criminal justice system has been found to have a deleterious effect on mental health (Sugie & Turney, 2017), the women in the current study were consistently found to be at a greater risk than men for poor health and social outcomes as well as criminal outcomes. For example, 30% of women with mental health problems were incarcerated during the follow-up period upon release from prison compared with 11% of the men with mental health problems. Prior research has pointed to the unique pathways to crime for women, including histories of physical, sexual, and emotional abuse; trauma; antisocial relationships; economic and social marginality; and homelessness (Belknap, 2001; Bloom & Covington, 2008; Bloom et al., 2003; Messina, Burden, Hagopian, & Pendergast, 2006; Richie, 2001), all of which are likely related and contribute to both the mental health and criminal involvement of female offenders. The distinctive pathways to incarceration for women may help explain the greater prevalence rates of mental health issues and the noted relationship between mental health and poor reentry outcomes for women compared with men in this study.

Reentry planning for offenders with mental health problems has been on the radar of policy makers and practitioners only rather recently, particularly given that states are beginning to utilize more gender-responsive policies and practices (Dlugacz, 2010; NRCJIW, 2018). As succinctly described by Dlugacz (2010), “the multiproblem nature of the population, coupled with a deep social disruption, lack of connection to treatment, and the means to pay for needed treatment” leads to a revolving door between prison and the community for these offenders (p. xv). A principal problem is the lack of coordination among service providers, including agencies responsible for community supervision, health care, treatment services, and housing. Without reentry planning, individuals with chronic mental health problems may experience homelessness and appear at the emergency department of the local hospital for health care. A greater utilization of prerelease screening and discharge referrals to treatment and specialized reentry programs, particularly for women, while also maintaining a continuum of care for these individuals would lead to more successful reentry outcomes and reduced recidivism. In addition, reentering prisoners may be limited in their ability to access mental health care in the community due to their lack of health insurance coverage. While offenders may have insurance prior to entering prison, these benefits are often suspended and/or terminated while incarcerated. As is the case with Medicaid, these benefits are suspended during incarceration and reinstatement can take several months. Given that the weeks immediately following release are often the most precarious for men and women reentering the community, a lack of insurance coverage could result in a situation where one’s mental health could deteriorate and negatively affect reentry success (Mallik-Kane & Visher, 2008).

Detailed guidance on developing evidence-based models for reentry planning and community reintegration, including for men and women with mental health problems, is available from the Reentry Policy Council (www.csgjusticecenter.org). Basic elements include (a) individualized transition planning, (b) risk assessment, (c) housing, (d) benefits, (e) release of medical information to community providers, (f) supply of medication, and (g) fixed appointments with health care provider, community supervision agent, and/or case manager. New approaches to discharge, reentry, and preparation of the former prisoner for life in the community are necessary. Of particular importance is linking individuals with mental health problems to treatment providers in the community to avoid lapses in treatment and medication. Such a strategy might involve prerelease screening of individuals within several months of discharge for mental health problems. This assessment would be followed by careful reentry planning to include an adequate supply of medication upon release (at least 30 days), treatment referrals, pre-arranged appointments with service providers within a week of discharge, and community case management for especially high-risk individuals. Such an orderly transition from prison to the community would alleviate the revolving door phenomenon that plagues many offenders with mental health problems, particularly for women.

In addition, states should look to incorporate a greater utilization of mental health courts (MHC) as a way to divert offenders with mental health problems toward treatment and social service providers. MHCs allow defendants that meet the criteria for a mental health condition to receive community-based treatment while seeking to prevent any continued involvement with the criminal justice system. With over 300 MHCs currently in operation across the United States (The Council of State Governments [CSG], n.d.), the aim of these courts is to improve public safety by reducing recidivism, reducing correctional costs, and improving quality of life and treatment for those with mental illness. While more randomized controlled trials are needed, initial evidence suggests that MHCs have the capacity to be effective in reducing recidivism (Kim, Becker-Cohen, & Serakos, 2015; Rossman et al., 2012; Sarteschi, Vaughn, & Kim, 2011; Snedker et al., 2017). A study by Costopoulos and Wellman (2017) found that criminal defendants who graduated from a MHC in Florida were significantly less likely to be rearrested within 3 years after release from MHC and that participation in MHC was able to reduce recidivism regardless of the varying severity of criminal history. MHCs appear to be a promising approach which could have the capacity to introduce gender-responsive treatment while diverting mentally ill offenders from the criminal justice system; however, we are unaware of any evaluations of such approaches.

While this study adds to the literature regarding gender differences in reentry outcomes among men and women leaving prison with mental health problems, it does have important limitations. First, the measure of mental health in this study measured if the respondent had ever been diagnosed with a mental health condition in his or her life. This may not be fully indicative of how one’s current mental health status during the course of this study affected their reentry experience. Second, the data for this study came from one state (Texas) and it is therefore unknown if the findings generalize out to the larger population of other male and female offenders exiting prison with mental health conditions. Third, the samples (particularly for females) utilized were relatively small and may not yield enough statistical power to detect all consequential differences as statistically significant. Finally, the self-reported nature of these data can be limiting. Despite the moderately high self-reports of criminal behavior, substance use, and other negative outcomes that may indicate that the respondents were not misleading in their responses, it cannot be guaranteed that the all responses were indeed accurate or honest.

Understanding the extent and distinctions in which one’s mental health differentially affects the reentry experiences of men and women exiting prison is consequential on many fronts. This study found that female offenders with mental health conditions experienced more negative reentry outcomes than their male counterparts which points not only to the need for greater utilization of prehealth assessment, MHCs, and a continuum of care upon release but also the need for gender-responsive treatment within those entities that are working with returning offenders with mental health conditions. Future research should continue to examine the long-term consequences of mental health on the reentry process with an emphasis on continuing to develop evidence-based programming that focuses on the unique challenges that mental health places on the criminal justice system and our communities. In addition, future research should attempt to more fully examine the influence that co-occurring mental health and substance use disorders, which are prevalent among incarcerated offenders, have on offender recidivism and desistance (James & Glaze, 2006; Mallik-Kane & Visher, 2008). This study adds to the growing literature on offenders with mental health conditions and clearly indicates the need for more attention to developing and implementing best practices in terms of mental health assessment, programming, and potential diversionary or “exit points” within the criminal justice system.

Footnotes

Authors’ Note:

The authors would like to thank the men and women who shared their experiences, opinions, and expectations with them. Without their willingness to share the details of their lives, this study would not have been possible. This article stems from a larger research effort at The Urban Institute’s Justice Policy Center. This research was funded through the generous support of the John D. and Catherine T. MacArthur Foundation, Annie E. Casey Foundation, JEHT Foundation, Cleveland Foundation, Houston Endowment, Robert Wood Johnson Foundation, Rockefeller Foundation, George Gund Foundation, Smith Richardson Foundation, Ohio Department of Rehabilitation and Correction, Ohio Office of Criminal Justice Services, and The Urban Institute.

Notes

Nicholas W. Bakken is an associate professor in the Department of Sociology at the University of Wisconsin–La Crosse. His research explores offending across the life course, prisoner reentry, criminal desistance, and substance use and health.

Christy A. Visher is professor of sociology and criminal justice at the University of Delaware and director of the Center for Drug and Health Studies. Over the past three decades, her research has focused broadly on crime and justice topics, including prisoner reentry, crime prevention strategies, implementing evidence-based practices in the criminal justice system, incapacitation, and use of drug testing in the criminal justice system.

References

Belknap

(2001). The invisible women: Gender, crime, and justice. Belmont, CA: Wadsworth.

Bloom

Covington

(2008). Addressing the mental health needs of women offenders. In Gido

Dalley

(Eds.), Women’s mental health issues across the criminal justice system (pp. 160-176). Upper Saddle River, NJ: Prentice Hall.

Bloom

Owen

Covington

(2003). Gender-responsive strategies: Research, practice, and guiding principles for women offenders. Washington, DC: National Institute for Corrections.

Broner

Lattimore

P. K.

Steffey

D. M.

(2010). Mental health needs and services receipt of reentering offenders: A multi-site study of men, women and male youth. In Dlugacz

H. A.

(Ed.), Re-entry planning for offenders with mental disorders: Policy and practice (pp. 12-41). Kingston, NJ: Civic Research Institute.

Bronson

Berzofsky

(2017). Indicators of mental health problems reported by prisoners and jail inmates, 2011-12. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Centers for Disease Control and Prevention. (2011). Mental illness surveillance among U.S. adults. Atlanta, GA: Author.

Cobbina

J. E.

Huebner

B. M.

Berg

M. T.

(2012). Men, women, and postrelease offending: An examination of the nature of the link between relational ties and recidivism. Crime & Delinquency, 58, 331-361.

Costopoulos

J. S.

Wellman

B. L.

(2017). The effectiveness of one mental health court: Overcoming criminal history. Psychological Injury and Law, 10, 254-263.

The Council of State Governments. (n.d). Mental health courts. Retrieved from https://csgjusticecenter.org/mental-health-court-project/

10.

Dlugacz

H. A.

(Ed.). (2010). Reentry planning for offenders with mental disorders: Policy and practice. Kingston, NJ: Civic Research Institute.

11.

Durose

M. R.

Cooper

A. D.

Snyder

H. N.

(2014). Recidivism of prisoners released in 30 states in 2005: Patterns from 2005 to 2010 (NCJ 244205). Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

12.

Foa

E. B.

Riggs

D. S.

Dancu

C. V.

Rothbaum

B. O.

(1993). Reliability and validity of a brief instrument for assessing post-traumatic stress disorder. Journal of Traumatic Stress, 6, 459-473.

13.

Gido

Dalley

(2008). Women’s mental health issues across the criminal justice system. Upper Saddle River, NJ: Prentice Hall.

14.

Greifinger

(2007). Public health behind bars. New York, NY: Springer.

15.

Hammett

T. H.

Roberts

Kennedy

(2001). Health-related issues in prisoner reentry. Crime & Delinquency, 47, 390-409.

16.

Harding

Roman

(2016). Identifying discrete subgroups of chronically homeless frequent utilizers of jail and public mental health services. Criminal Justice and Behavior, 44, 511-530.

17.

James

D. L.

Glaze

L. E.

(2006). Mental health problems of prison and jail inmates (NCJ 213600). Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

18.

Kim

Becker-Cohen

Serakos

(2015). The processing and treatment of mentally ill persons in the criminal justice system. Washington, DC: The Urban Institute.

19.

La Vigne

Kachnowski

. (2005). Texas prisoner’s reflections on returning home. Washington, DC: Justice Policy Center, The Urban Institute. Retrieved from urban.org/sites/default/files/alfresco/publication-pdfs/311247-Texas-Prisoners-Reflections-on-Returning-Home.PDF

20.

Mallik-Kane

Visher

(2008). Health and prisoner reentry: How physical, mental, and substance abuse conditions shape the process of reintegration. Washington, DC: The Urban Institute.

21.

Manderscheid

R. W.

Gravesande

Goldstrom

I. D.

(2004). Growth of mental health services in state adult correctional facilities, 1988 to 2000. Psychiatry Services, 55, 869-872.

22.

Mears

D. P.

Cochran

J. C.

(2012). U.S. prisoner reentry health care policy in international perspective: Service gaps and the moral and public health implications. The Prison Journal, 92, 175-202.

23.

Messina

Burden

Hagopian

Pendergast

(2006). Predictors of prison therapeutic communities treatment outcomes: A comparison of men and women participants. American Journal of Drug and Alcohol Abuse, 31, 7-28.

24.

Mowen

T. J.

Visher

C. A.

(2016). Ties that bind: How incarceration impacts family relationships. Criminology & Public Policy, 15, 503-528.

25.

National Resource Center on Justice Involved Women. (2018, October 15). Available from http://www.cjinvolvedwomen.org/

26.

Paternoster

Brame

Mazerolle

Piquero

(1998). Using the correct statistical test for the equality of regression coefficients. Criminology, 36, 859-866.

27.

Radloff

L. S.

(1991). The use of the center for epidemiologic studies depression scale in adolescents and young adults. Journal of Youth and Adolescence, 20, 149-166.

28.

Richie

B. E.

(2001). Challenges incarcerated women face as they return to their communities: Findings from life history interviews. Crime & Delinquency, 47, 368-389.

29.

Rossman

S. B.

Wilson

J. B.

Mallik-Kane

KiDeuk

Sherrill

S. D.

Downey

P. M.

(2012). Criminal justice interventions for offenders with mental illness: Evaluation of mental health courts in Bronx and Brooklyn, New York. Washington, DC: Office of Justice Programs, U.S. Department of Justice.

30.

Sarteschi

Vaughn

M. G.

Kim

(2011). Assessing the effectiveness of mental health courts: A quantitative review. Journal of Criminal Justice, 39, 12-20.

31.

Snedker

K. A.

Beach

L. R.

Corcoran

K. E.

(2017). Beyond the “revolving door?” Incentives and criminal recidivism in a mental health court. Criminal Justice and Behavior, 44, 1141-1162.

32.

Spjeldnes

Goodkind

(2009). Gender differences and offender reentry: A review of the literature. Journal of Offender Rehabilitation, 48, 314-335.

33.

Sugie

N. F.

Turney

(2017). Beyond incarceration: Criminal justice contact and mental health. American Sociological Review, 82, 719-743.

34.

Travis

(2005). They all come back. Washington, DC: The Urban Institute.

35.

Van Voorhis

Wright

Salisbury

Bauman

. (2010). Women’s risk factors and their contributions to existing risk/needs assessment: The current status of a gender-responsive supplement. Criminal Justice and Behavior, 37, 261-288.

36.

Visher

(2007). Returning home: Emerging findings and policy lessons about prisoner reentry. Federal Sentencing Reporter, 20, 93-102.

37.

Visher

Bakken

(2014). Reentry challenges facing women with mental health problems. Women & Health, 54, 768-780.

38.

Wright

E. M.

Van Voorhis

Salisbury

E. J.

Bauman

(2012). Gender-responsive lessons learned and policy implications for women in prison: A review. Criminal Justice and Behavior, 39, 1612-1632.