Esuba

Abstract

This study examined the effects associated with a trauma and abuse-focused psychoeducation group for incarcerated women on decreasing trauma symptoms. A total of 320 women participated in 34 groups in five prisons that followed a manualized intervention. A one-group pretest–posttest design was used to examine changes on 10 Trauma Symptom Inventory subscales (Anxious Arousal, Depression, Anger/Irritability, Intrusive Experiences, Defensive Avoidance, Dissociation, Sexual Concerns, Dysfunctional Sexual Behavior, Impaired Self-Reference, and Tension Reduction Behavior). Analyses indicated significant improvement on all 10 trauma subscales. Small effect sizes were found on all of the subscales. The findings of this study are encouraging and are the first step toward more rigorous evaluation of this pilot program. These findings provide initial support for the use of group psychoeducation intervention to address traumatic stress symptoms with incarcerated women.

Keywords

abuse trauma incarcerated women posttraumatic stress psychoeducation

Introduction

Female offenders are the fastest growing inmate population in the United States, but in 2011, only 6.5% of all inmates in state and federal prisons were women (Bureau of Prisons, 2011). Therefore, it is not unexpected that the majority of prison programming and treatment that have been studied are for men (Lewis, 2006). The treatment and programming needs of incarcerated women have received increased research focus in the past 15 years due to a number of important factors: a steady annual increase of women being incarcerated (Sabol, West, & Cooper, et al., 2009), the increased awareness of their mental health and psychiatric needs and how they differ from males (Baldwin & Jones, 2000), and the understanding that they will reenter their communities at some point in their lifetimes. Incarcerated women have consistently self-reported significantly higher rates of mental health problems than males (73% vs. 55%, Bureau of Justice Statistics, 2006). Due to limited opportunities to treat the problems that women enter prison with, they are often unprepared to reenter into their communities (Petersilia, 2001; Thompson, 2004; Visher, La Vigue, & Travis, 2004). Many female offenders relapse and continue to be challenged by the posttraumatic stress, addiction problems, and poor self-esteem that existed upon prison entry (Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995; Kubiak, 2004; Zlotnick, Najavitz, Rohsenow, & Johnson, 2003).

Incarcerated women are often disadvantaged economically and have considerable mental health needs (Lewis, 2006). Incarcerated women are more likely than male inmates to report a history of victimization (Lewis, 2006). Amnesty International (2004) stated that between 44% and 88% of female inmates reported experiencing posttraumatic stress disorder related to experiences of physical or sexual abuse prior to entering prison. Female offenders have been found to go to prison with higher rates of prior trauma and abuse (Chesney-Lind & Pasko, 2008; McDaniels-Wilson & Belknap, 2008), serious mental health issues, and low self-esteem often resulting in self-injurious behavior when compared with male offenders (P. Bloom, 2003; Kubiak, 2004). Strong correlations between the female offenders’ history of childhood violence and abuse and involvement in criminal activity have been found in the research (Brown, Miller, & Maguin, 1999; Chesney-Lind & Pasko, 2004; Steffensmeier & Allen, 1996). Steffensmeier and Allen (1996) suggested that many women engage in criminal behavior to survive poverty and abuse combined with their substance use and dependency. Mental health problems suffered by incarcerated women often result from lifetime abuse histories and include major depression, posttraumatic stress disorder, and substance use/dependence (Chesney-Lind & Pasko, 2004; Kessler et al., 1995; Lewis, 2006; Zlotnick et al., 2003). It has been suggested that because of these violent and abuse-filled histories, female inmates require and utilize the medical and mental health systems in prison more than twice as often as their male counterparts (incarcerated males) and nonincarcerated females (Lindquist & Lindquist, 1999; Young, 1998). Providing programming to address the symptoms of depression and posttraumatic stress disorder related to lifetime abuse histories of female inmates may improve their ability to focus on the future while in prison and on their return to the community. Utilizing this “window of opportunity,” female-focused programming to address traumatic experiences and resulting trauma symptoms are necessary. This study evaluates one such program named Esuba (abuse spelled backward), which offers inmates the opportunity to discuss and receive intervention for their trauma and abuse histories and resulting trauma symptomology.

There is a paucity of research on the incidence and treatment of posttraumatic stress disorder among incarcerated women. Teplin, Abram, and McClelland (1996) found that posttraumatic stress disorder was the most common mental health disorder after substance dependence among jailed women. Thirty-three percent were found to have lifetime posttraumatic stress disorder, whereas 22.3% had current posttraumatic stress disorder (Teplin et al., 1996). Kubiak (2004) found out of 60 female inmates, 60% met the criteria for posttraumatic stress disorder.

There are few treatments that have been developed and evaluated to specifically target posttraumatic stress in incarcerated populations (Zlotnick, 2002; Zlotnick et al., 2003). Due to the limited resources given toward treatment in incarceration settings, intensive treatments like cognitive-behavioral interventions may be successful in addressing posttraumatic symptoms (Zlotnick et al., 2003) but requires skilled clinicians and has limited feasibility in most jail and prison settings.

The timing of when treatment is initiated for women with traumatic stress related symptoms has not been well explored in the research literature. Foa, Keane, and Friedman (2000) in their guidelines for the treatment of posttraumatic stress disorder found no conclusive data that age of treatment affected treatment outcomes. Stalker, Palmer, Wright, and Gebotys (2005) found in their study of inpatient treatment of adults abused as children that participants of 29 years and younger had an abrupt increase in posttraumatic stress disorder symptoms after 3 months, but after 12 months did better than all of the older participants.

The Esuba Program

The Esuba program was a psychoeducation intervention that was provided in a group format. The tenets of the groups are education, support, and empowerment. Esuba is manual based and utilizes the sharing of information and narration of experiences to address issues such as how those experiences affected the participants in the past, how they affect them currently, how to make changes, and how to deal with the thoughts and feelings that go along with their trauma (anger, sadness, suicidal ideations, self-harm, worthlessness, guilt).¹

The intervention manual consists of 10 topics, including identifying violence and abuse, stereotypes, cultural and historical abuse, sexual battery and sexual abuse, abuse in families, child abuse, elder abuse, abuse of the disabled, perception versus reality, and self-abuse. The last 2 weeks are devoted to releasing the past and healing the future when the participants discuss transition, new coping skills, and how they plan to take positive steps to improve their futures.

Each session format is structured to provide the participants with (a) information (definitions, research findings, vignettes) about specific types of abuse and traumatic experiences while supporting the group members to reflect on the information presented with their opinions, personal experiences, and feedback for the other group members. (b) The group leaders facilitate discussion related to the topic and are led in role-plays, read topic-focused poems (written by the participants and/or from the manual), and are led in activities, including life history exploration. Group members are encouraged to focus on the future and to see things from a view of self-awareness and power.

This intervention appeared to be promising as found in a preliminary study of the Esuba program that targeted self-esteem (Bedard, Pate, & Roe-Sepowitz, 2003). The preliminary study explored the pretest and posttest scores on the Rosenberg Self-Esteem Scale (RSE; Rosenberg, 1979) of 157 male and female inmates provided with the Esuba group. Findings included improvement on scores related to self-esteem, self-satisfaction, faith in people, and a decrease in sensitivity to criticism. Ward and Roe-Sepowitz (2009) conducted a second pilot study on the Esuba group in 2009 with a sample of prostituted women. The study compared the impact of the Esuba psychoeducation group on the trauma symptoms of two groups of prostituted women; one group was in a targeted residential program (n = 18) focused on the exiting process of women in prostitution and the other group was in a female prison (n = 11; Ward & Roe-Sepowitz, 2009). The Esuba group intervention was found to have significant impact in the decrease of trauma symptoms for both groups with the prison group showing the most significant improvements (Ward & Roe-Sepowitz, 2009). After the initial study of data collected from 1997 to 2000, the study authors changed the focus to be more specific to addressing trauma symptoms experienced by incarcerated women. A revised manual was developed with attention to information exchange and facilitating participants to narrate their thoughts, feelings, ideas, and experiences.

Psychoeducation Groups and Theoretical Foundation

Psychoeducation groups have been studied involving children and mood disorders (Cummings & Fristad, 2007; Fristad, 2006; Mendenhall, Fristad, & Early, 2009; West & Pavuluri, 2009), and their parents and families (McClendon, Pollio, North, Reid, & Johnson-Reid, 2007). Other psychoeducation group research has explored adult individuals (Pomeroy, Rubin, Van Langingham, & Walker, 1997), couples (Pomeroy, Green, & Van Laningham, 2002), and families (Pomeroy, Rubin, & Walker, 1995) affected by HIV. In a study exploring a psychoeducation group focused on decreasing depression, anxiety, and trauma symptoms among female juveniles incarcerated as adults, Pomeroy, Green, and Kiam (2001) found the intervention to significantly affect depression and trauma symptoms. The psychoeducation group was for 18 sessions over 9 weeks and was a time-limited manual-based intervention with three components: education, didactic, and supportive.

Herman (1992) identified three stages of treating trauma survivors who are experiencing traumatic symptoms (depression, anxiety, and posttraumatic stress disorder). The stages are (a) safety and stability, (b) remembrance and mourning, and (c) reconnection to ordinary life (Herman, 1992). The Esuba program is based on the first stage of safety and stabilization. This includes working with the participants on feelings of safety within their own bodies and within their relationships to others. Addressing feelings of safety may allow their trauma symptoms to decrease or stabilize. This symptom reduction/stabilization work is the key to allowing the participants to change. This group intervention applied the work of safety and stabilization with the participants’ relationships with themselves (self-worth, self-concept) and with others (supporting each other, trusting others). The intervention aims to address the issues of depression, anger, and trauma symptoms such as dissociation, avoidance, poor coping skills, and a weak sense of self as identified by Lewis (2006) as some of the key mental health problems of incarcerated women.

This group applied the psychoeducation approach to group intervention. Psychoeducation interventions combine supportive therapy with education to address the participant’s emotional needs and empowering them through increased knowledge and awareness (Dziegielewski, 1991). The intent is through the education and processing (narrating their experiences) to affect levels of trauma symptoms through accessing coping strategies, and integration of their past experiences into their current decisions, and feeling a sense of support and community from the group structure.

The information-based, manual-based curriculum of Esuba focuses on the statistics, definitions, and case examples of a topic and is then followed by a discussion about how traumatic experiences can manifest in each individual. Group members are asked to describe how they feel that their experiences have affected them as a person, their choices, and the people around them. When participants describe a symptom (such as feeling depressed a lot), the leader encourages the other group members to talk about if they had ever felt a similar way and how they coped with those feelings.

The purpose of this study was to examine the effectiveness of a psychoeducation group intervention aimed at providing the participants with opportunities to gain knowledge about their experiences as well as addressing their trauma symptoms. The aims of this intervention were to provide a supportive learning and narration group environment as well as to ameliorate some of the group members’ trauma symptoms. The primary goal of this study was to determine the effectiveness of a pilot group intervention in reducing the amount of trauma symptoms reported at the end of 12 weeks of intervention among incarcerated adult female inmates. The research hypotheses were as follows:

Hypothesis 1: There will be a significant decrease in trauma symptom subscale score means from pretest to posttest with incarcerated female adult participants in an in-prison, closed format, time-limited trauma, and abuse-focused psychoeducation group.

Hypothesis 2: Findings will show significant effect of the intervention on the 10 trauma symptom subscale from pretest to posttest controlling for group membership, pretest, age, race, and childhood emotional, sexual, and physical abuse.

Hypothesis 3: There will be significant mean differences on pretest to posttest trauma subscale scores between young participants (ages 18-24) and older participants (ages 25-61) with younger participants showing more reduction in symptoms.

Method

Overview of the Study and Design

The design of this study was a one-group pretest–posttest design. A time-limited psychoeducational group intervention was provided to volunteer participants affected by abuse and trauma in five women’s prisons in Florida. The length of the group was 2-hr weekly sessions over 12 weeks. Groups were provided at times when inmates would not be working or in class and varied by prison. Although there are limitations of this design, pilot intervention studies with only a treatment sample can provide valuable information regarding pursuit of replication using more experimental research designs. Although it was difficult to account and control for threats to internal and external validity, efforts were made in the effect size analysis to account for the cluster effects of group/facility, pretest scores, age, race, and childhood emotional, physical, and sexual abuse by using a random effects regression to control for the confounding impact on internal validity of statistical regression (Cambell & Stanley, 1963; Cameron & Trivedi, 2009).

This study involved collecting data from voluntary participants from 34 Esuba groups provided in five women’s prisons in Florida from June 2003 to May 2009. The groups were provided in the prisons by group leaders who were prison and community-based master’s level counselors/social workers or community volunteers trained and supervised by a master’s level counselor.

Participation in the groups was voluntary for the inmates. In four of the facilities, participants signed up for the groups on sheets placed in the classification officers’ offices (akin to a case manager) that they meet with monthly. The sign-up sheets were posted 30 days before the group starting date. In the fifth facility, the Chaplain notified inmates about the group and maintained an ongoing list with the top 25 entering the next group. Participants were informed during the first Esuba group meeting about informed consent and permission to use the collected data for research purposes. The Institutional Review Board of Florida State University and Arizona State University reviewed and approved this study. The Florida Department of Corrections (FDC) and the first author have an established Memorandum of Agreement that outlines the transfer of Esuba data for analysis and research procedures. The FDC provided financial support to this program from 2003 until 2005. In 2005, two members of the research team became employed at FDC and continued to administer and supervise the program in that capacity.

Prior to data collection, an Esuba facilitator’s manual was created with a concentration on addressing domestic violence, childhood abuse, and traumatic stress symptoms. The manual included 12 weeks of materials, including educational information about each topic, weekly group activities, and poems; guidelines for the facilitators regarding how to discuss confidentiality, group structure, and data collection instructions; and guidelines of how to facilitate the discussion for each group. An experienced social work PhD and two master’s level clinicians (counseling and social work) ran groups and trained and supervised group leaders. Group leaders were trained on group processes as well as the Esuba curriculum. The training protocol includes an overview of the program objectives, discussion of FDC procedures, dress code, conduct in a group setting, interaction with inmates, inmate accountability, personal safety, ethics, and standards. Group leaders then shadowed one of the experienced group leaders through a complete offering of the intervention group prior to coleading a group.

Group leaders were instructed to remain within the guidelines of the session outlined in the manual to ensure treatment fidelity. Criteria for inclusion of participants in this study included participation and completion of 10 weeks of the 12-week Esuba trauma and abuse-focused psychoeducation group and completion of the pretest, Esuba survey, and posttest. Pretests were administered along with the consent forms during the first group meeting, the Esuba survey was provided to participants on Week 6, and posttests were administered at the end of the last group session (Week 12).

Participants

Participants were adult women in prison, ages 18 or above, all of whom voluntarily attended a 12-week trauma and abuse-focused psychoeducation group called Esuba.

Attrition Numbers and Reasons

A total of 415 women signed-up and began the 12-week program during the study period. Of them, 65 women did not complete a posttest, 2 did not complete a pretest, and 9 did not complete an Esuba survey. Reasons for the noncompletions included beginning a new prison job (with different hours), transfer to another prison, early release, being ill on the last session, or having been placed in administrative or disciplinary confinement. In all, 29 cases were removed from analysis of the Trauma Symptom Inventory (TSI) subscales due to having higher than acceptable scores (75 or above) on the Inconsistent Response scale (17 cases at pretest, 12 cases at posttest), which indicates inconsistent responses that may reflect random endorsement, language issues, or concentration problems (Briere, 1995). The final sample was made up of 320 women. When compared with the participants not included in the study, there were no significant differences on any of the TSI pretest subscales, age, sentence length, and history of prior incarceration, but the two groups did differ significantly on race, χ²(3, n = 425) = 12.1, p < .01, with African American women dropping out more than expected when compared with White or Hispanic women. The final sample of 320 women ranged in age from 19 to 61 years (M = 33.8, SD = 9.84). The majority (198, 62%) were White, 106 (33%) were African American, 12 (3.8%) were Hispanic, and 4 (1.2%) were Native American.

Data Collection

Procedures and Measures

Three instruments were used: a life experience survey called the Esuba Survey, the modified Parental Psychological Maltreatment Scale (PYS; Briere & Runtz, 1990), and the TSI (Briere, 1995). All instruments collected self-report data. The Esuba survey was created by Roe-Sepowitz, Bedard, and Pate (2007) to obtain information regarding childhood abuse and maltreatment (e.g., have you ever been molested), family issues (e.g., parental drug use problems, observing domestic violence), risk-taking behaviors (e.g., drinking alcohol excessively, having sex with strangers, prostitution), and criminal behaviors (e.g., have you ever driven a car drunk). It is a self-administered questionnaire with primarily dichotomous answer options. Questions are categorized into six sections: demographics, family history, health, substance use, behavior, and justice system contact. Sexual abuse was determined by answers of “yes” to either of the two questions asking about molestation or rape prior to age 18. Physical abuse was determined by a “yes” response to any of the three questions about whether a caregiver had ever made them bruise, bleed, or had broken a bone. There has been no exploration of reliability or validity on the Esuba survey.

The 7-item modified PYS (Briere, 1992) was used to assess childhood emotional abuse. The PYS stem asks, “Prior to age 15, how often did the following occur during an ‘average’ year?” followed by questions asking “How often did a parent, stepparent, foster-parent, or adult in charge of you ‘yell at you,’ ‘insult you,’ ‘criticize you,’ ‘try to make you feel guilty,’ ‘ridicule or humiliate you,’ ‘embarrass you in front of others,’ and ‘make you feel like you were a bad person’?” (Briere & Runtz, 1990). Each item was scored on a 6-point scale ranging from 0 = never to 6 = more than 20 times per year. The PYS was modified by the authors to increase the time span of “childhood” to include all years prior to age 18 and to minimize the requirement to recollect specific incident frequencies. Thus, the response format was changed to a 5-point scale (never, rarely, sometimes, often, and always). The scores were then dichotomized such that responses of “often” or “always” on all 7 items were categorized as childhood emotional abuse. This was done to minimize reporting errors as participants may not recall events or, if they do recall them, they may remember without specificity (Leonard-Barton, 1990). Previous studies have shown that the original format had reasonably good alpha reliability (Briere & Runtz, 1988, 1990) and Briere (1992) reported that the successful use of the PYS in various studies suggests predictive and construct validity. Using the data from this study with the modified PYS scoring, Cronbach’s alpha coefficient was .96, supporting strong internal consistency.

The TSI was created by Briere (1995) to assess traumatic stress symptoms. The TSI measures specific symptoms, including dissociation, posttraumatic stress, and related psychological symptoms (Briere, 1995). The full 100-item measure was used in pre- and posttest format. The TSI asked respondents about how often specific experiences had occurred during the past 6 months for the pretest and then the instructions were changed on the posttest to state that the participant should report how often those specific experiences they were currently experiencing. The TSI scale has been shown to have coefficient alpha reliability coefficients ranging from .80 to .90 and exhibit reasonable convergent, discriminant, and predictive validity in normative and clinical samples (Briere, 1995). The TSI subscales reliability estimates were calculated with data from the present study and each demonstrated strong reliability. Cronbach’s alpha reliability estimates are anxious arousal (.83), depression (.88), anger/irritability (.88), intrusive experiences (.83), defensive avoidance (.80), dissociation (.85), sexual concerns (.86), dysfunctional sexual behavior (.90), impaired self-reference (.81), and tension-reduction behavior (.84). The clinical scales of anxiety, depression, anger, intrusive experiences, defensive avoidance, dissociation, sexual concerns, dysfunctional sexual behavior, impaired self-reference, and tension-reduction behaviors reflect symptoms in each of these areas.

Results

Sentence lengths ranged from 4 to 418 months (M = 55.3, SD = 53.4) with three participants with life sentences. Seventy-six (23.8%) of the participants had been in prison prior to this incarceration. The crimes committed for the current incarceration for the participants included (many had more than one charge) drug related, 108 (33.8%); theft, 78 (24.4%); fraud, 40 (12.5%); burglary, 38 (11.9%); battery, 30 (9.4%); murder/homicide, 30 (9.4%); robbery, 29 (9.1%); trafficking in stolen property, 26 (8.1%); possession of a weapon, 22 (6.9%); child abuse and neglect, 18 (5.6%); driving under the influence (DUI) manslaughter, 16 (5%); assault, 15 (4.7%); serious driving offense, 11 (3.4%); both DUI and resisting arrest, 10 (3.1%); and below 10: prostitution, stalking, attempted murder, obstruction of justice, sexual offense, kidnapping, arson, and exploitation of the elderly. See Table 1 for descriptive information about the participants regarding childhood abuse experiences, lifetime drug use, and family details.

Table 1.

Esuba Participant Characteristics.

Variables	Participants (N = 320)	%
Childhood abuse (prior to age 18)
Sexual abuse	229	71.6
Molested	204	63.8
Raped	193	60.3
Physical abuse	152	47.5
Emotional abuse	75	23.4
Witnessed domestic violence as a child	166	51.9
Mother hit father	94	29.4
Father hit mother	156	48.8
Parents with drug/alcohol problem	210	65.6
Witnessed drug use in home	157	49.1
History of running away	107	33.4
Received social services as a child	111	34.7
Adult experiences
Has children	175	54.7
Adult victim of rape	109	34.1
Family member history of jail/prison	218	68.1
Risk behaviors
Self-mutilation	104	32.5
Drinking excessively	166	51.9
Using drugs	267	83.4
Having sex with strangers	132	41.3
Attempted suicide (lifetime)	141	44.1
Believe they are addicted to drugs/alcohol	101	31.6
Prostitution	165	51.6
Sold drugs	213	66.6
Questions added in 2006	Participants(n = 180)
Adult domestic violence
Victim of domestic violence	156	86.7
Perpetrator of domestic violence	129	71.6

Analyses

All 34 psychoeducation groups were combined for the analysis. Because we expect the entire sample to have high trauma symptom scores due to volunteering to participate in a group focused on abuse and trauma, the mean scores were centered and when exploring effect size, the pretest scores (i.e., the original effects of the abuse and trauma) were controlled to better understand the impact of the intervention. The Bonferroni method was used to control for Type I errors and we set our significance level at .01 to avoid Type I error. To explore the pretest and posttest score changes, t tests were conducted. We then clustered the data by group using a form of generalized least squares regression called random effects (Cameron & Trivedi, 2009). The random effects regression estimated the random effects and standard errors appropriate for clusters. The random effects regression allowed for each group’s intercept to randomly vary with the resulting estimated average effects across all of the groups and an appropriate standard error (Cameron & Trivedi, 2009). For our regression coefficients to reflect an effect size that takes into consideration the ceiling effect of most participants entering the group with traumatic experience histories, Glass’s D effect size was calculated by subtracting from each posttest the pretest average and divided by the standard deviation of the pretest (Glass & Hopkins, 1996). Cohen (1988) labeled effect sizes as small (.2-.49), medium (.5-.79), and large (greater than .8).

Results indicated a statistically significant difference between pretest and posttest for all of the 10 trauma subscales on the TSI. On two of the subscales, tension-reduction behavior and dysfunctional sexual behavior, mean scores changed from a mean score above the clinically significant score (65) to below. None of the other pretest subscale mean scores were above the clinically significant score. A t test for each subscale indicated a statistically significant difference between pretest and posttest: tension reduction behavior, t(319) = 7.55, p = .001; impaired self-reference, t(319) = 10.85, p = .001; dysfunctional sexual behavior, t(319) = 2.82, p = .005; sexual concerns, t(319) = 6.82, p = .001; dissociation, t(319) = 7.83, p = .001; defensive avoidance, t(319) = 8.92, p = .001; intrusive experiences, t(319) = 8.59, p = .001; anger/irritability, t(319) = 7.59, p = .001; depression, t(319) = 9.62, p = .001; and anxiety, t(319) = 8.4, p = .001. All of the TSI subscales demonstrated a small effect size as a contribution of the intervention. Please see Table 2 for means, standard deviations, mean changes from pretest to posttest, and the effect size of the intervention on each trauma subscale. Please see Table 3 for adjusted effect size using the random effects regression to estimate average effects across all groups and controlling for pretest scores, age, race, and childhood abuse (sexual, emotional, and physical).

Table 2.

Pretest and Posttest Mean Scores, Mean Differences, and Effect Sizes.

	Pretest		Posttest
N = 320	M	SD	M	SD	Difference	Effect size	t value	p value
TSI subscales
Tension reduction behavior	65.7	16.5	60.2	16.2	5.41**	.33	−7.55	.001
Impaired self-reference	60.5	9.9	55.8	9.6	4.74**	.48	−10.85	.001
Dysfunctional sexual behavior	68.7	18.5	63.8	32.7	4.96*	.27	−2.82	.005
Sexual concerns	59.1	12.7	55.4	11.8	3.69**	.29	−6.82	.001
Dissociation	62.8	12.1	58.4	12.3	4.38**	.36	−7.83	.001
Defensive avoidance	63.8	8.3	60.1	9	3.68**	.44	−8.93	.001
Intrusive experiences	62.6	10.7	58.6	11.1	3.95**	.37	−8.59	.001
Anger/irritability	58.8	10	55.5	10.1	3.29**	.33	−7.59	.001
Depression	58.1	9.8	54	9.2	4.02**	.41	−9.63	.001
Anxious arousal	58.2	10	54.3	10.2	3.85**	.38	−8.4	.001

Note: Pretest and posttest means were centered prior to analysis.

p < .01. **p < .001.

Table 3.

Adjusted Effect Size Controlling for Pretest Scores, Age, Race and Childhood Abuse (Sexual, Emotional, and Physical).

	Pretest score			Posttest score
TSI subscale	M	SD	ICC	M	SD	ICC	Correlations between pre- and posttest
Tension reduction behavior	65.7	16.5	0.381	60.3	16.2	0.337	0.694
Impaired self-reference	60.5	9.9	0.256	55.8	9.6	0.105	0.68
Dysfunctional sexual behavior	68.7	18.5	0.282	63.76	32.72	0	0.349
Sexual concerns	59.1	12.7	0.226	55.4	11.8	0.202	0.689
Dissociation	62.8	12.1	0.254	58.4	12.3	0.144	0.663
Defensive avoidance	63.8	8.3	0.093	60.1	9.0	0.033	0.637
Intrusive experiences	62.6	10.7	0.19	58.6	11.1	0.194	0.716
Anger/irritability	58.8	10.0	0.092	55.5	10.2	0.058	0.703
Depression	58.1	9.8	0.327	54.0	9.2	0.136	0.691
Anxious arousal	58.2	10	0.313	54.3	10.2	0.125	0.672

Note: TSI = Trauma Symptom Inventory; ICC = Intraclass Correlation Coefficient.

Impact of Intervention for Age

In the Florida prison system, most individuals ages 18 to 24 are adjudicated as a young offender and are housed separately from the general population. For this analysis, participant data were divided into two groups, ages 18 to 24 years (n = 72) and 25 years and older (n = 248) to determine whether the intervention was more efficacious for younger offenders or older offenders (age 25 and above). Chi-square tests were conducted to determine whether the two groups differed significantly (at the .05 level) on lifetime abuse (childhood sexual, physical and emotional abuse, adult rape, family involvement in substance abuse). The two groups differed only on reported experiences of adult rape, χ²(1, n = 275) = 4.62, p < .03, with the younger group reporting more than expected compared with the older group. Hypothesis 3 proposed that there would be significant mean differences on pretest to posttest trauma subscale scores between young participants (ages 18-24) and older participants (ages 25-61) with younger participants showing more significant change. Paired sample t tests were conducted to examine the mean differences between the two age groups from pretest to posttest. Both groups showed significant change (decrease) between pretest and posttest on all TSI subscale scores, but the younger group did not show significant score changes on the anxiety (M = −2.46, SD = 8.43), t(71) = 2.48, p = .016, and dysfunctional sexual behavior subscales, (M = 1.49, SD = 60.39), t(71) = −.209, p = .84. Overall change was greater for the older group. Please see Table 4 for mean changes from pretest to posttest by age group. Hypothesis 3 was not supported by the findings.

Table 4.

Within Group Mean Differences for Young Participants (18-24 Years Old) Compared With Older Participants (Age 25 and Above).

	Young group(n = 72)		Older group(n = 248)
N = 320	Difference	t	Difference	t
TSI subscales
Tension reduction behavior	−3.64*	2.84	−5.92**	7.01
Impaired self-reference	−3.19**	3.96	−5.19*	10.17
Dysfunctional sexual behavior	1.49	−.21	−6.83*	7.37
Sexual concerns	−3.10**	3.43	−3.87*	5.97
Dissociation	−5.07**	4.05	−4.19*	6.70
Defensive avoidance	−2.26*	3.24	−4.10*	8.36
Intrusive experiences	−4.17**	4.84	−3.89*	7.22
Anger/irritability	−3.42**	3.66	−3.26*	6.64
Depression	−3.07**	3.73	−4.30*	8.91
Anxious arousal	−2.46*	2.48	−4.26*	8.27

p < .01. **p < .001.

Discussion and Application to Social Work Practice

The findings of this study suggest preliminary support for this intervention as a promising approach to addressing trauma symptoms experienced by incarcerated women. The results suggest a small effect size for all of the TSI subscales ranging from .27 to .48. As none of these trauma symptom subscales are generally amenable to change, any finding of improvement or impact is positive.

The structure of the group—learning, narrating, and supportive listening—may have been an unusual experience for the participants and may have contributed to the positive changes on trauma symptom scores. Another possible explanation for the positive effect associated with the intervention may have been due to the lack of information many of the participants had about the connection between their traumatic and abusive experiences and their reactions in behavior and emotion. Many of the participants reported that they had not considered themselves having been abused until they heard stories similar to theirs from group members who identified as having been abused. This recognition appeared to be critical for some participants to then be able to deal with their sense of self and the coping strategies they had used to emotionally function throughout their lives. Putnam (1990) described the awareness of victimization as a method of processing life experiences and integration and development of a self-concept. This awareness, or lack thereof, directly affects their mastery or power over their environment (Kagan, 1965), the decisions they make regarding relationships, and the way they treat themselves and those around them.

Many of the participants had very poor self-views and weak self-images. Most centered their identity on their actions or their relationships with others. Coping skills were also very limited and included high rates of self-destructive behaviors such as self-mutilation, suicide attempts, risk taking, drug use, and explosive behavior. Exploring self-abuse as a topic as well as discussing new coping strategies during each session appeared to be important aspects of this intervention.

It is difficult to assess the impact of the prison environment except to acknowledge that it is constrictive, punitive, and does not facilitate self-exploration or disclosure due to the complex relationships among inmates and staff. Few participants reported having received mental health services except for psychotropic medication management but some were participating in or had participated in faith-based unit living, therapeutic drug communities, and substance abuse support groups (e.g., Alcoholics Anonymous and Narcotics Anonymous). A number of participants identified their prison time as the first time in their lives they could focus on themselves and not the chaotic relationships around them. They stated that this group felt like an opportunity to take the time they needed to make necessary decisions and work on themselves. The groups often resulted in new social networks for the participants with resulting friendships and supportive relationships after the group ended.

The null result of Hypothesis 3 was not surprising. The age at which an individual begins to deal with their abuse and traumatic experiences has not been determined to affect outcomes (Foa et al., 2000). We specifically explored these age groups because they are often housed separately and so little is known about the responsiveness to interventions among different age groups.

Findings from the current study may be of use to social workers, and programming and clinical staff in jails and prisons. As most services for incarcerated persons currently focus on education and vocational training rather than mental health services or treatment programs (Gendrau & Andrews, 1994), the creative use of resources (volunteers, internal clinical staff) to run groups that suggest promising associated results is necessary. Social workers and counseling staff within female prisons may find this program to be particularly useful for their clients. The group format is manual based and does not require advanced clinical training, but does require an initial training and supervision to facilitate fidelity. The social work staff may be able to utilize and supervise trained volunteers from outside the prison to provide this intervention. By providing an intervention that decreases traumatic stress symptoms, the participants may be able to use their prison time as a “window of opportunity” to make the changes they desire and practice for community reentry. For social workers in clinical practice, these findings introduce a new intervention that may also be useful to treat nonincarcerated abuse and trauma survivors.

Future Research

The next step in the evaluation of the effectiveness of the Esuba group intervention is to develop a wait-list control group study. Follow-up surveys at 3 and 6 months may also provide useful information to better evaluate the impact associated with this intervention. Lewis (2006) recommended that treatments addressing major depression or posttraumatic stress disorders should be integrated with substance dependence programs; therefore, adding aspects of substance use treatment and relapse prevention may be beneficial.

Limitations

As with all studies, this study has a number of limitations and the findings must be interpreted with caution. The most significant limitation is the lack of a comparison group, therefore history, and regression to the mean and maturation cannot be ruled out as possible reasons for the changes found. In consideration of that, by using the random effects regression to cluster the data by group attended, we attempted to decrease the impact of the specific group, including leadership style, group size, and specific prison. The participants were from five prisons in Florida but do not represent all Florida inmates and the findings can only be generalized to the participants.

The self-report nature of the instruments is a limitation and strength of the study. Paulhus and Vazire (2007) found self-report data to be rich with details, but Paulhus (1991) wrote that self-report data can be susceptible to social desirability effects and cannot be verified, thus limiting validity. The Esuba survey was asked after 6 weeks of group attendance with the intention that the participants would feel a therapeutic alliance with the group leaders and would perhaps be more open to telling the truth about their experiences. All of the group leaders except for one were not prison employees and the prison staff did not see the instruments. Another limitation is the lack of follow-up data, thus we are unable to evaluate the durability of the changes from the intervention.

The findings of this study are encouraging and are the first step toward a more rigorous evaluation of this pilot intervention. These findings provide initial support for the use of group psychoeducation intervention to address traumatic stress symptoms with incarcerated women. The findings are encouraging and support the need for gender-specific, trauma-based prison programming. If society’s focus is on reentry, providing inmates with tools to deal with their trauma and abuse during incarceration can only assist them with making a healthy transition back into the community.

Regarding use of the Esuba manual, the research team is interested in collaborating with other groups to assist in training to provide the intervention and share collected data. Please contact the first author for details.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Amnesty International. (2004). Women in custody. Retrieved from www.amnestyusa.org/pdf/custodyissues.pdf

Baldwin

Jones

(2000). Health issues specific to incarcerated women: Information for state maternal and child health programs. US Department of Health and Human Services, Health Resources and Services Administration Maternal and Child Health Bureau.

Bedard

L. E.

Pate

K. N.

Roe-Sepowitz

D. E.

(2003). A program analysis of Esuba: Helping turn abuse around. International Journal of Offender Therapy and Comparative Criminology, 47, 579-607.

Bloom

(2003, May). Gender responsive strategies: Research, practice and guiding principles. Paper presented at the American Jail Association 22nd Conference, Albuquerque, New Mexico.

Briere

(1992). Child abuse trauma: Theory and treatment of the lasting effects. Newbury Park, CA: Sage.

Briere

(1995). Trauma symptom inventory: Professional manual. Lutz, FL: Psychological Assessment Resources.

Briere

Runtz

(1990). Differential adult symptomatology associated with three types of child abuse histories. Child Abuse & Neglect: The International Journal, 14, 357-364.

Briere

Runtz

(1988). Multivariate correlates of childhood psychological and physical maltreatment among university women. Child Abuse & Neglect: The International Journal, 12, 331-341.

Briere

Runtz

(1990). Differential adult symptomology associated with three types of child abuse histories. Child Abuse & Neglect: The International Journal, 14, 357-364.

10.

Brown

Miller Maguin

E. B.

(1999). Prevalence and severity of lifetime physical and sexual victimization among incarcerated women. International Journal of Law and Psychiatry, 22, 301-322.

11.

Bureau of Justice Statistics. (2006, September). Mental health problems of prison and jail inmates. Bureau of Justice Statistics, US Department of Justice.

12.

Bureau of Prisons. (2011, July). Quick facts about the Bureau of Prisons. Retrieved from http://www.bop.gov/news/quick.jsp

13.

Cambell

Stanley

(1963). Experimental and quasi-experimental designs for research. Boston, MA: Houghton Miffin.

14.

Cameron

A. C.

Trivedi

(2009). Microeconometrics using Stata. College Station, TX: Stata Press.

15.

Chesney-Lind

Pasko

(2004). The female offender: Girls, women, and crime (2nd ed.). Thousand Oaks, CA: Sage.

16.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

17.

Cummings

Fristad

(2007). Medications prescribed for children with mood disorders: Effects of a family-based psychoeducation program. Experimental and Clinical Psychopharmacology, 15, 555-562.

18.

Dziegielewski

(1991). Social group work with the family members of elderly nursing home residents with dementia: A controlled evaluation. Research on Social Work Practice, 1, 359-370.

19.

Foa

Keane

Friedman

(2000). Guidelines for treatment of PTSD. Journal of Traumatic Stress, 13, 539-588.

20.

Fristad

(2006). Psychoeducational treatment for school-aged children with bipolar disorder. Developmental and Psychopathology: Special Issue: Developmental Approaches to Bipolar Disorder, 18, 1289-1306.

21.

Gendrau

Andrews

(1994). Tertiary prevention: What the meta-analysis of offender treatment literature tells us about “what works.” Canadian Journal of Criminology, 32, 173-184.

22.

Glass

G. V.

Hopkins

K. D.

(1996). Statistical methods in education & psychology (3rd ed.). Boston, MA: Allyn & Bacon.

23.

Herman

(1992). Trauma and recovery: The aftermath of violence-from domestic violence to political terror. New York, NY: Basic Books.

24.

Kagan

(1965). The concept of identification, In Mussen

P. H.

Conger

J. J.

Kagan

(Eds.), Readings in Child Development and Personality (pp. 315-326). New York, Harper and Row.

25.

Kessler

Sonnega

Bromet

Hughes

Nelson

(1995). Posttraumatic national comorbidity survey. Archives of General Psychiatry, 52, 1048-1060.

26.

Kubiak

(2004). The effects of PTSD on treatment adherence, drug relapse, and criminal recidivism in a sample of incarcerated men and women. Research on Social Work Practice, 14, 424-433.

27.

Leonard-Barton

(1990). A dual methodology for case studies: Synergistic use of a longitudinal single site with replicated multiple sites. Organisation Science, 1, 248-266.

28.

Lewis

(2006). Treating incarcerated women: Gender matters. Psychiatric Clinics of North America, 29, 773-789.

29.

Lindquist

(1999). Health behind bars: Utilization and evaluation of medical care among jail inmate. Journal of Community Health, 24, 285-303. doi:10.1023/A:1018794305843

30.

McClendon

Pollio

North

Reid

Johnson-Reid

(2007). School-based groups for parents of children with emotional and behavioral disorders: Pilot results. Families in Society, 88, 124-129.

31.

McDaniels-Wilson

Belknap

(2008). The extensive sexual violation and sexual abuse histories of incarcerated women. Violence Against Women, 14, 1090-1120.

32.

Mendenhall

Fristad

Early

(2009). Factors influencing service utilization and mood symptom severity in children with mood disorders: Effects of multifamily psychoeducation groups. Journal of Consulting and Clinical Psychology, 77, 463-473.

33.

Paulhus

D. L.

(1991). Measurement and control of response bias. In Robinson

J. P.

Shaver

P. R.

Wrightsman

L. S.

(Eds.), Measures of personality and social psychological attitudes (pp. 17-59). San Diego, CA: Academic Press.

34.

Paulhus

D. L.

Vazire

(2007). The self-report method. In Robins

R. W.

Fraley

R. C.

Krueger

R. F.

(Eds.), Handbook of research methods in personality psychology (pp. 224-239). London, England: Guilford.

35.

Petersilia

(2001). Prisoner reentry: Public safety and re-integration challenges. The Prison Journal, 81, 360-375.

36.

Pomeroy

Green

Kiam

(2001). Female juvenile offenders incarcerated as adults: A psychoeducational group intervention. Journal of Social Work, 1, 101-115.

37.

Pomeroy

Green

Van Laningham

(2002). Couples who care: The effectiveness of a psychoeducational group intervention for HIV serodiscordant couples. Research on Social Work Practice, 12, 238-252. doi:10.1177/104973150201200203

38.

Pomeroy

Rubin

Van Langingham

Walker

(1997). “Straight Talk”: The effectiveness of a psychoeducational group intervention for heterosexuals with HIV/AIDS. Research on Social Work Practice, 7, 149-164.

39.

Pomeroy

Rubin

Walker

(1995). Effectiveness of a psychoeducational and task-centered group intervention for family members of people with AIDS. Social Work Research, 19, 142-152.

40.

Putnam

F.W.

(1990). Disturbances of ‘self’ in victims of childhood sexual abuse. In Kluft

R. P.

(Ed.), Incest-related syndromes of adult psychopathology (pp. 113-131). Washington, DC: American Psychiatric Press.

41.

Roe-Sepowitz

D. E.

Bedard

L. E.

Pate

K. N.

(2007). The impact of child abuse on dissociative symptoms: A study of incarcerated women. Journal of Trauma & Dissociation, 8, 7-26. doi:10.1300/J229v08n03_02

42.

Rosenberg

(1979). Conceiving the self. New York, NY: Basic Books.

43.

Sabol

West

Cooper

(2009). Prisoners in 2008 (Bureau of Justice Statistics Bulletin, NCJ 228417). U.S. Department of Justice, Office of Justice Studies.

44.

Stalker

Palmer

Wright

Gebotys

(2005). Specialized inpatient trauma treatment for adults abused as children: A follow up study. American Journal of Psychiatry, 162, 552-559.

45.

Steffensmeier

Allen

(1996). Gender and crime: Toward a gendered theory of female offending. Annual Review of Sociology, 22, 459-487.

46.

Teplin

Abram

McClelland

(1996). Prevalence of psychiatric disorders among incarcerated women. American Journal of Public Health, 84, 290-293.

47.

Thompson

(2004). Navigating the hidden obstacles to ex-offender reentry. Boston College Law Review, 2, 255-306.

48.

Visher

La Vigue

Travis

(2004). Returning home: Understanding the challenges of prisoner reentry: Maryland pilot study, findings from Baltimore. Urban Institute, Justice Policy Center. Retrieved from http://www.urban.org/center/jpc/returning-home/index.cfm

49.

Ward

Roe-Sepowitz

D. E.

(2009). Assessing the effectiveness of a trauma-oriented approach to treating prostituted women in a prison and a community exiting program. The Journal of Aggression, Maltreatment & Trauma, 18, 293-313.

50.

West

Pavuluri

(2009). Psychosocial treatments for childhood and adolescent bipolar disorder. Child and Adolescent Psychiatric Clinics of North America, 18, 471-482.

51.

Young

(1998). Health status and service use among incarcerated women. Family & Community Health, 21, 16-31.

52.

Zlotnick

(2002). Treatment of incarcerated women with substance abuse and posttraumatic stress disorder: Final report. Rockville, MD: NCJRS.

53.

Zlotnick

Najavitz

Rohsenow

Johnson

(2003). A cognitive-behavioral treatment for incarcerated women with substance abuse disorders and posttraumatic stress disorders: Findings from a pilot study. Journal of Substance Abuse, 25, 99-105.