Well-Being Interventions for Correctional Officers in a Prison Setting: A Review and Meta-Analysis

Abstract

This review and meta-analysis evaluated the effectiveness of well-being interventions offered to correctional officers within prison settings. A search strategy was developed and 11 databases were searched to identify eligible studies. Articles were screened against preset eligibility criteria, and quality was assessed using the Downs and Black Checklist. Nine studies were identified, including four single-group design and five quasi-experimental designed studies. Interventions comprised a mixture of crisis interventions, psychoeducational programs, and an exercise program. For those studies with sufficient data, a meta-analysis was conducted to examine intervention effects on stress and psychopathology. Fixed-effects meta-analysis showed that treatment had no effect on stress or psychopathology. Of those studies that could not be included in the meta-analysis, there were mixed outcomes. These findings highlight the need for more rigorous study designs and suggest further research is needed to examine the theoretical mechanisms in the development of interventions within correctional settings.

Keywords

well-being stress correctional officers interventions prison guard

The well-being of correctional officers (COs) in prison settings has long been a concern, given the complex and difficult environment in which they operate. Research has consistently demonstrated that COs experience serious and multiple adverse effects on their well-being as a result of their work (Dowden & Tellier, 2004; Trounson, Pfeifer, & Skues, 2019). The prison environment, with its extreme levels of control, minimal autonomy, and complex population, exposes COs to ongoing stressors across multiple domains (Finney, Stergiopoulos, Hensel, Bonato, & Dewa, 2013). More recently, research has suggested that COs’ experience of adversity in prisons exceeds that of many community occupations and is equivalent to those in other high-risk professions (e.g., emergency service personnel and police; Trounson, Pfeifer, & Critchley, 2016).

Recognition of the seriousness and magnitude of workplace adversity has led to increased research into stress experienced by COs in prison settings, with a number of stressors being identified as inherent to the corrections environment. The risk of violence is the most significant and feared occupational threat, with literature emphasizing the enormous strain of daily exposure to such behaviors and interactions with offenders who have a record of violence (Brower, 2013; Finn, 1998; Huckabee, 1992). The added pressure of crowding and violence between inmates further complicates and exacerbates officer stress (Finn, 1998; Misis, Kim, Cheeseman, Hogan, & Lambert, 2013). The prison setting itself, often characterized by regimentation, exposure to infectious disease, and a need for high levels of hypervigilance, is psychologically draining (Brower, 2013; Hartley, Davila, Marquart, & Mullings, 2013). Furthermore, the physical conditions of the prison environment, such as limited access to natural lighting, noise levels, cleanliness, dilapidation, and privacy, have been found to contribute to somatic and psychological distress, increased sick leave, and substance use in prison staff (Bierie, 2012).

In addition to these context-specific stressors, COs also deal with challenging organizational or administrative factors. These factors include poor communication between administration, limited autonomy, exclusion from decision-making processes, little opportunity for promotion, underutilization of knowledge and skills, and lack of positive recognition (Brower, 2013; Finney et al., 2013; Huckabee, 1992; Morgan, 2009; Schaufeli & Peeters, 2000). Within the prison context, role ambiguity, role overload, and role conflict are common, creating further stress (Brough & Williams, 2007; Brower, 2013; Morgan, 2009).

Finally, individual-level factors affect officer stress. Specifically, officer perception both of danger and of the inmates affects the severity of their stress, such that more negative perceptions predict higher levels of stress (Castle & Martin, 2006; Dowden & Tellier, 2004; Misis et al., 2013). In addition, CO attitudes toward help-seeking are often unhelpful. For example, the literature identifies common attitudes that promote a “macho” image and stigmatize stress and distress (Cheek & Miller, 1983; Dowden & Tellier, 2004). These attitudes are likely to have a negative effect on help-seeking behavior, thus reducing the likelihood that COs will access services to address and reduce their stress experiences. Interestingly, Cheek and Miller (1983) found that, in keeping with the “macho” image, COs denied their stress, yet objective markers (e.g., health measures) indicated that they were not only experiencing stress but were also negatively affected by it. Given the CO subculture, it is likely that officer reports of stress have been more globally underreported in the literature, and that services provided to target stress have not been taken up to the extent that they are needed.

Chronic workplace stress experienced by COs has been consistently associated with significant negative outcomes, both for corrections more generally and for COs personally. At an organizational level, CO stress can contribute to unsafe practices within the corrections facility, higher staff turnover, higher job absenteeism, and poor job performance (Brower, 2013; Finney et al., 2013; Trounson et al., 2018). At a personal level, workplace stress has serious deleterious effects for COs, with officers experiencing high levels of stress-related health problems, including cardiovascular disease, diabetes, gastrointestinal problems, and hypertension (Dowden & Tellier, 2004; Harvey, 2014; Morgan, 2009). In addition, research targeted specifically at the mental health of COs working in prisons has indicated that COs have high prevalence of anxiety, psychosomatic disorders, substance use, and psychological distress (Bourbonnais, Jauvin, & Dussault, 2007; Ghaddar, Mateo, & Sanchez, 2008; Morgan, 2009).

Investigations into the presence of psychiatric disorders in COs have found high rates of adjustment disorders, alcohol abuse (Tartaglini & Safran, 1997), mood and anxiety disorders, including depressive symptoms (Obidoa, Reeves, Warren, Reisine, & Cherniack, 2011), and posttraumatic stress disorder (PTSD; Spinaris, Denhof, & Kellaway, 2012). Disturbingly, CO stress has also been associated with shorter life span (Garland, 2002) and significantly higher rates of suicide in comparison with the general working population (Stack & Tsoudis, 1997).

Overall, the literature clearly indicates that officer well-being should be a target for intervention (Trounson & Pfeifer, 2017). However, there are ongoing and considerable variations across the literature in defining and operationalizing well-being, with such wide variations indicating the complexity and elusiveness of the construct (Pollard & Lee, 2003). A fast-growing body of research specifically investigating well-being across many domains has contributed to a more sophisticated and comprehensive understanding of the construct, but a simple and definitive conceptualization is unlikely (MacKian, 2009). Given this expansive approach to well-being, it is unsurprising that there is no standardized measurement of well-being (Pollard & Lee, 2003). Practically, this means that well-being tends to be defined and operationalized contextually, with each study measuring it differently, and often doing so without an explicit discussion of the concept itself (Dodge, Daly, Huyton, & Sanders, 2012). In acknowledging this inconsistency across the literature, Dodge et al., using historical and theoretical contributions to the development of the concept of well-being, proposed a useful, if somewhat generalized, definition of well-being as “the balance point between an individual’s resource pool and the challenges faced” (p. 230). That is, well-being is that state when individuals consistently have a stable set of resources—psychological, social, and physical—that allows them to successfully traverse psychological, social, and physical challenges.

In the context of occupational settings, well-being is again broadly conceptualized, with three categories emerging: psychological well-being, specifically in reference to processes and practices in the workplace; physical well-being as associated with or as an outcome of the workplace; and social well-being, including meaningful social connections within the workplace, and perceptions of equality within the workplace (Brunetto, Teo, Shacklock, & Farr-Wharton, 2012). In addition, well-being within the workplace has been positively correlated with job satisfaction and job performance (Grant, Christianson, & Price, 2007). Unsurprisingly, there is no clear agreement on the definition of well-being within a corrections environment. This ambiguity has been recognized more broadly in the relevant literature, with acknowledgment that, despite this lack of clarity, the importance of employee well-being is undiminished, both for the employer and the employee (MacKian, 2009).

The Present Study

Given the serious, dangerous, and substantial effects of job-related stress on officer well-being, there is a need for programs that can effectively reduce COs’ distress levels and increase their well-being (Trounson et al., 2018). Unfortunately, there are very few such programs available to COs (Trounson & Pfeifer, 2017). The most commonly provided intervention, Employee Assistance Programs, is typically staffed by service providers who are not specialists in the prison context and, therefore, fail to understand the complex and multifaceted nature of issues faced in the field of corrections (Brower, 2013).

Although there have been a small number of programs that target officer well-being, there is currently no systematic examination of the effectiveness of these programs. Corresponding to the broad range of well-being definitions and measures in the literature, this article aimed to collectively examine the type and effectiveness of any intervention that specifically targeted any facet of officer well-being in a prison environment. Both narrative review and meta-analysis were used to report effectiveness.

It should be noted that throughout the studies included in the review, there has at times been limited theoretical discussion around key concepts. For example, aggression, hostility, and violence have often been used interchangeably, without clearly defining or differentiating the terms. As such, this review has used whichever term was employed in the original study.

Method

Data Sources and Search Strategy

We conducted a review of the literature following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009). Between July and September 2016, 11 electronic databases (ProQuest, PsycInfo, EBSCOhost, Informit, SAGE Journals, Web of Science, CINAHL Plus, Criminal Justice Abstracts with Full Text, MEDLINE, Social Work Abstracts and Springer Link) were searched to identify relevant interventions published in peer-reviewed journals, government reports, and other gray literature, such as theses, organizational reports, and professional publications (e.g., Corrections Today published by the American Correctional Association). In addition, Internet searches were conducted to capture available gray literature. Searches were developed using database-specific index terms in relation to the subject of the review. Search terms included using subject headings (e.g., stress, intervention, well-being, CO) and keywords (e.g., correction* OR correction* OR prison AND officer OR guard AND program OR intervention OR training). In addition, hand searching of reference lists of included studies for relevant articles was conducted. An updated search was undertaken in February 2018 to ensure that the review included any new articles that met the criteria.

Study Selection (Inclusion and Exclusion Criteria)

Preliminary searches indicated a paucity of methodologically rigorous studies, leading us to develop broad inclusion criteria that would capture any literature published on the topic. As such, the search included studies that utilized a quantitative methodology, thus allowing for some assessment of intervention effectiveness. Furthermore, studies without a control group were included if there were outcomes comparing pre- and postdata of the experimental group. Beyond this, the following eligibility criteria were applied:

The article is published in full text in English.

The article focuses on evidence regarding psychological and well-being interventions for COs who worked in a prison environment, including youth and adult prisons. Studies where COs worked in both community and prison locations were excluded if the results were not reported separately. Studies which reported interventions that targeted CO well-being specifically within the context of workplace trauma were included given the daily risk of such exposure experienced by COs.

The article reports outcomes that are theoretically related to officer well-being, including but not limited to markers of stress, employment satisfaction, and other psychological outcomes. Studies that did not provide sufficient detail regarding the intervention or outcomes were excluded. Studies with outcomes that were provided exclusively in qualitative format were not included.

Assessment of Quality and Risk of Bias

The Checklist for Measuring Quality (Downs & Black, 1998) is one of the two quality assessment tools recommended by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins & Green, 2011). Given its facility in assessing both randomized and nonrandomized studies, the Downs and Black Checklist was used as a descriptive measure to evaluate the quality of the articles being included in this review. Two researchers independently applied the checklist to the included studies with 98% agreement. Differences were resolved by consultation between the two reviewers. Scores ranged from 10 to 23 on a 27-item scale, with higher scores indicating better quality (note that values actually range from zero to 28 due to one item containing three points instead of two). Summary scores are reported in Table 1, and full details of the quality assessment can be found in Supplemental Appendix 1 (available in the online version of this article).

Table 1:

Overview of Studies Included in the Review

Authors	Reference	Quality score	Study design	Sample	Intervention	Outcome measures	Main results/conclusions
Backer (1990)	[1]	19/27	Quasi-experimental	62 prison officers (federal U.S. prison)	Stress inoculation training 4 × 2 hr sessions	Symptom Checklist-90-R; absenteeism	Improved outcomes on Symptom Checklist-90-R by those who had completed the SIT intervention when compared with the control group. However, there was no significant difference in rates of absenteeism. Effect size (psychopathology): d = 0.63 in favor of tx
Booth (2009)	[2]	19/27	Quasi-experimental	62 CO trainees, (employed to work in Broward County Correctional facilities postprogram completion)	Cognitive Behavioral Stress Management program 2.5 hr over 6 months	Symptom Checklist-90-R; Job Stress Survey; Balanced Inventory of Desirable Responding, V6; informal assessment of learning from intervention program.	Increase in job stress severity at 6 months for experimental group as compared with control group. Experimental group demonstrated decreases on some job stress measures and increases in phobic anxiety and depression. Effect size (psychopathology): d = 0.05 in favor of tx Effect size (stress): d = −0.08 in favor of tx
Bravo-Mehmedbasic et al. (2009)	[3]	10/27	Case series	122 prison guards (three prisons in Bosnia and Herzegovina).	Psychoeducation Two sessions implied (insufficient detail provided)	Index of reaction, STAI questionnaire, SAMAČA questionnaire.	Differences between first and second measuring of subjects from one prison indicated statistically significant reduction of stress reactions, improvement of coping strategies and communication skills. Effect size (psychopathology): r = .20
Dignam and Fagan (1996)	[4]	12/27	Case study	95 COs (Atlanta Penitentiary COs had been held hostage for 11 days)	Critical Incident Stress Debriefing/Family and Employee Assistance Team Length of intervention not specified	PTSD symptoms (informal, incomplete checklist used), some Holmes–Rahe Scale questions, short- and long-term need identification	Decrease in PTSD symptoms, with the exception of intrusive thoughts. Effect size (PTSD symptoms): Cohen’s h = .53
Engelmann (1997)	[5]	19/27	Quasi-experimental	111 correctional staff (Connecticut Department of Corrections)	Critical incident stress debriefing Length of intervention not specified	Impact of Events Scale; Center for Epidemiological Studies—Depression; Critical Incident Stress Debriefing Survey	Participants who had undergone CISD post involvement in a critical incident reported increased levels of depression in comparison with those who had not undertaken CISD post involvement in a critical incident. Effect size (psychopathology): d = 0.53 in favor of control
Johns (1986)	[6]	14/27	Quasi-experimental	55 correctional officers (maximum-security prison in Pennsylvania)	Stress inoculation training package delivered in varying designs to four groups, plus one control group which received no treatment. 2 × 1 hr sessions	General Stress Measures: Derogatis Stress Profile; Job-Related Tension Index; Maslach Human Services Inventory; systolic/diastolic blood pressure; pulse rate. Internal skills measures: STAXI Form Y1 and Y2; irrational beliefs test. External skills measures: Novaco Advocation Inventory; Assertion Inventory; Affective Sensitivity Scale. Standard Treatment Control Measure: Jenkins Activity Survey	There were no significant differences between any treatment groups on any of the dependent measures. The study concluded that a comprehensive stress inoculation training package does not reduce CO stress levels. Anxiety and anger levels also remained unaffected regardless of which intervention group the officer had been assigned to. Effect size: insufficient data
McCraty, Atkinson, Lipsenthal, and Arguelles (2009)	[7]	18/27	Quasi-experimental	94 COs (Northern California Youth Correctional Center)	Power to Change Performance program Two days	Health Risk Assessment; biological markers of stress (lipid panel, blood pressure, heart rate variability, cortisol, DHEA, and S-IgA); psychological measures (Jenkins Activity Survey, Personal and Organizational Quality Assessment)	Psychological improvement: Significant reductions in fatigue and anger along with increased positive outlook, gratitude, productivity, motivation, goal clarity, and manager support (POQA). Significant reductions in interpersonal sensitivity, hostility, and overall psychological distress (BSI), and significant reductions in the global Type A behavior pattern and in speed and impatience (JAS). Physiological markers: Total cholesterol, glucose, and blood pressure were reduced in both groups, whereas LDL cholesterol fell in the experimental group Effect size (psychopathology): d = −0.15 in favor of treatment Effect size (stress): d = −0.22 in favor treatment
Ruck, Bowes, and Tehrani (2013)	[8]	15/27	Quasi-experimental	91 COs exposed to trauma (HM prisons in the United Kingdom)	Crisis Intervention Debriefing 1 × 2 hr session	GAD; IESE	Experimental group had significantly decreased scores on all subscales of the IESE (avoidance, emotional arousal, and reexperience) in postassessment. Control group showed no change. GAD scores were unaffected. Effect size (psychopathology): d = 0.06 in favor of control
Sidney and Jette (1987)	[9]	14/27	Case series	25 male COs (Leclerc Institution—medium security prison in Canada)	46-day on-site exercise program comprising individualized strength and aerobic training.	Correctional Service Health and Stress Questionnaire; Attitudinal Questionnaire × 2 (20 bipolar adjectives on each describing direction and intensity of feelings)	All participants reported significantly increased positive feelings toward the prison environment and their roles, relaxation, and positive stimulation and reduced hostility, anxiety, and rigidity. Overall, the program was associated with positive changes in view of self, mood and attitudes toward work. Effect size (positive attitudes): d = 0.20 in favor of T2 Effect size (attitudes toward work): d = −1.09 in favor of T2

Note. CO = correctional officer; STAI = State-Trait Anxiety Inventory; CISD = crisis intervention stress debriefing; STAXI = State-Trait Anger Expression Inventory; DHEA = dehydroepiandrosterone; POQA = Personal and Organizational Quality Assessment; BSI = Brief Symptom Inventory; JAS = Jenkins Activity Survey; GAD = General Anxiety and Depression Scale; IESE = Impact of Event Scale-Extended; PTSD = posttraumatic stress disorder; LDL = low-density lipoprotein; HM prison = Her Majesty’s prison.

In addition to the quality assessment undertaken, each study was assessed for risk of bias using the Risk of Bias Tool in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins & Green, 2011). This assessment was conducted by the primary author to specifically evaluate the methodological rigor of each study. The checklist addressed selection bias, performance bias, detection bias, attrition bias, and reporting bias. Multiple criteria in each category were assessed to determine whether their presence or absence affected risk of bias, and then the category was ranked accordingly, as either low or high. The assessor could also rate an item as unclear if there was insufficient information provided in the article to determine the level of risk. While risk of bias was assessed, it should be noted that studies were not excluded on the basis of methodological quality so that we could ensure a comprehensive assessment of available studies. With regard to publication bias, because fewer than 10 studies were included in the meta-analysis, funnel plots were not analyzed (Higgins & Green, 2011).

Data Collection and Analysis

A meta-analysis was conducted to explore the effect of interventions on officer well-being. Data were extracted using a standardized form to enter intervention information including method, participants, interventions, and outcomes. For the outcome variable of interest, we extracted pre- and posttreatment scores, sample sizes, means, and standard deviations for the intervention and any control groups. We defined posttreatment as the first assessment after the completion of treatment. Given the heterogeneity of the studies and largely poor data collecting and/or reporting, not all of the included studies contained the data necessary for meta-analysis. Although these studies were not analyzed in the meta-analysis, they have been included as a narrative synthesis to complement the meta-analysis. Of the four studies excluded from meta-analysis, it was possible to calculate effect sizes for Studies 3, 4, and 9. Study 9 did not provide sufficient data whereby effect size could be calculated. See Table 2 for an overview of main effects.

Table 2:

Summary of Risk of Bias

Study	Random sequence generation (selection bias)	Allocation concealment (selection bias)	Blinding of participants and personnel (performance bias)	Blinding of outcome assessment (detection bias)	Incomplete outcome data (attrition bias)	Selective reporting (reporting bias)	Other bias
Backer (1990) [1]	+	–	?	?	+	–	?
Booth (2009) [2]	–	?	?	?	+	+	?
Bravo-Mehmedbasic et al. (2009) [3]	?	?	?	+	+	–	?
Dignam and Fagan (1996) [4]	–	?	?	–	–	–	+
Engelmann (1997) [5]	–	–	+	?	?	–	?
Johns (1986) [6]	+	+	?	?	+	–	+
McCraty, Atkinson, Lipsenthal, and Arguelles (2009) [7]	+	?	?	?	+	?	+
Ruck, Bowes, and Tehrani (2013) [8]	–	–	+	+	?	?	+
Sidney and Jette (1987) [9]	–	?	?	+	?	–	–

Note. “+” means low risk; “–” means high risk; “?” means unclear risk of bias.

A meta-analysis was conducted using Review Manager 5 software (Version 5.3; The Nordic Cochrane Centre, Copenhagen). To test for overall effect, a fixed-effects model for standardized mean difference was used due to the small number of studies being analyzed (Borenstein, 2009). Chi square was conducted to test for heterogeneity.

Results

Search Results

The initial search returned a total of 669 articles plus two articles that were located through hand searching. In the first screen, 112 duplicates were identified and removed, followed by screening of titles and abstracts, which resulted in exclusion of 488 articles that did not meet eligibility criteria. The selection process was guided by a PICOS tool (Higgins & Green, 2011) that provided details for inclusion, with all decisions being justified and recorded (available upon request). These results were verified with 100% agreement by a second research team member. A total of 71 articles remained for evaluation on the basis of their full text, which resulted in a total of nine articles for inclusion, with the other 62 articles failing to meet eligibility criteria. A summary of the search process is illustrated in Figure 1.

Figure 1:

Flowchart of Study Selection for the Review

Study Characteristics and Quality

All studies were conducted within a 27-year period (1986-2013). The features of the selected studies, including author, quality assessment score, study design, sample, intervention, outcome measures, and main results, are provided in Table 1. Each study was assigned a number and is referred to throughout the article by its number when mentioned.

All studies focused on officers working in adult prison settings, with the exception of Study 7, which was based in a youth correctional center. The number of participants ranged from 25 to 122. Six studies (1, 2, 4-7) included officers in correctional facilities in the United States; the three remaining studies included officers at prisons in Bosnia and Herzegovina (3), the United Kingdom (8), and Canada (9). Three studies specifically targeted officers who had experienced trauma within the prison context (4-5, 8). None of the studies compared gender differences. Other demographic information, such as age, education, tenure, position, and amount of contact with prisoners, was largely unreported.

Overall, the quality of the studies was assessed as average, with issues being identified in relation to weaker study designs, failure to address confounding variables, and limited reporting of information needed to assess risk of bias. Only four studies (1, 5-7) employed some form of randomization, which negatively affected the methodological rigor. Of these, Study 1 provided no detail regarding the random sequence generation. Study 5 selected every 10th person from an employee list. Studies 6 and 7 stratified participants into groups and then randomly allocated subjects to conditions, but provided limited (7) or no (6) information regarding initial recruitment and randomization methods used. Other potential sources of bias were allocation concealment (unclear or lacking), blinding (largely unclear), and selective reporting (under-reporting across most studies of information likely to have been collected, including, for example, descriptive statistics [4], means and/or standard deviations [3, 6]). A full summary of risk of bias is provided in Table 2.

Intervention Types

Studies 1, 2, and 6 delivered multisession group-format psychoeducation programs that focused on stress management or stress reduction. Similarly, Study 3 delivered brief, group-format stress management training, but provided only vague details, reporting there were a “couple” of sessions, without indicating their length, frequency, or specific content. Study 7 delivered the “Power to Change Performance” program, a five-module group-format stress reduction program delivered across two consecutive days. This program incorporated emotion-focusing and reframing techniques, as well as provided biofeedback training to assist participants in their learning and implementation of the training.

Studies 4, 5, and 8 delivered crisis intervention stress debriefing (CISD), varying in extent and duration. Study 4 described a comprehensive CISD response, reporting the inclusion of individual, family, and group interventions. Individual and family sessions were characterized as crisis counseling, and group sessions provided typical debriefing as well as brief therapeutic interventions for PTSD symptoms. Information about number, duration, and frequency of sessions was not provided. The CISD intervention employed in Study 5 was delivered by a specialized response team and guided by four principles: (a) an initial psychoeducational intervention informing participants about posttrauma effects and coping skills; (b) consistent support; (c) ventilation, that is, being able to talk freely about the event; and (d) coping skills development. Again, information regarding number, duration, and frequency of sessions was not provided. Study 8 reported a CISD intervention in the form of a single 2- to 3-hr group-based session comprising seven stages exploring the incident (roles in the incident, facts about the incident, thoughts during the incident, feelings during the incident, reactions since the incident, planning around future issues, and closure).

In a departure from typical programs targeting officer well-being, Study 9 delivered a 46-day, on-site exercise program aimed at improving mood, attitudes toward work, and feelings about self. The program comprised individual fitness sessions targeting strength and aerobic training and was part of the officers’ daily duties within the prison. Participants were requested to exercise daily and provided with a schedule of workout activities, duration, and intensities.

Control groups employed across the studies differed widely. Studies 1 and 8 used a no-treatment control. Study 6 had two control groups: a no-treatment control and an intervention matched to the target invention. Study 2 used a treatment-as-usual control group, and Study 7 used a wait-list. Studies 3, 4, and 9 did not have a control group, limiting their usefulness in the meta-analysis.

Outcome Measures

As noted earlier, the studies utilized a wide and disparate range of measures to assess indicators of well-being, with few measures being shared across studies. This review, therefore, categorized measures into six themes to facilitate comparison of outcomes. The themes are as follows: physical health markers (e.g., pulse rate, blood pressure), measures of stress, measures of psychopathology (e.g., depression, anxiety, trauma symptoms), measures of positive markers (e.g., motivation, general positive attitudes, and communication skills), measures of negative markers (e.g., anger, hostility, absenteeism, and Type A behavior), and attitudes toward work. Table 1 provides a summary of the measures used in each study.

Effectiveness of Interventions

Two separate meta-analyses were undertaken to assess treatment effectiveness. The first meta-analysis, aimed to assess well-being outcomes that measured stress, analyzed the two studies (2 and 7) that provided sufficient data. Studies 1, 4, 5, and 8 did not include a stress measure. Studies 3 and 6 did not report data appropriate for meta-analysis. Study 9 measured stress preintervention, but not postintervention. These studies were, therefore, excluded from the meta-analysis. The second meta-analysis assessed well-being outcomes that measured psychopathology, with Studies 1 to 8 employing a measure that assessed some dimension of psychopathology. Of these eight studies, Studies 3, 4, and 6 did not contain the data needed for meta-analysis and thus were not included.

Analysis of effects of treatment on stress

No effect was found for treatment on stress (standard mean difference [SMD] = −0.15; 95% confidence interval [CI] = [−0.50, 0.20]; p = .40), compared with the control group (see Figure 2). This finding is largely confirmed by the studies not included in the meta-analysis. Of these, Study 3 reported mixed findings, with significant reductions of stress reactions in officers based at prisons where people had shorter sentences or were on remand. However, this trend did not hold for officers based at the prison where detainees had long sentences. It was not possible to calculate overall effect scores for this study, limiting the interpretation of these findings. In addition, the methodological information provided is so vague it is difficult to extrapolate meaning with any confidence. Nevertheless, the Study 3 investigators suggested the lack of improvement was because officers in such prisons are exposed to greater levels of stress. Similarly, Study 6 reported no significant outcomes on measures of stress, despite using several different measures to assess CO stress levels.

Figure 2:

Forest Plots for Effects of Treatment on Psychopathology and Stress

Analysis of effects of treatment on psychopathology

No effect was found for treatment on psychopathology (SMD = −0.01; 95% CI = [−0.22, 0.20]; p = .92), compared with the control group (see Figure 2). Although considerable heterogeneity was found (χ² = 11.81, df = 4 [p = .02], I² = 66%), it was ignored because a fixed-effect meta-analysis was used (Higgins & Green, 2011). Those studies not included in the meta-analysis partially align with the meta-analysis results. Study 4, which included specific measures of trauma symptoms, had mixed results. The study reported a decrease in PTSD symptoms, with the exception of intrusive thoughts, which increased after treatment. However, beyond percentages of participant symptom experiences, there was no statistical data or analysis included in the article, making it impossible to determine correlation or significance. Study 6, which assessed anxiety, found no significant differences. Finally, Study 3, which assessed state and trait anxiety, found significant reductions of anxiety in officers based both at the remand prison and at one of the long-term sentence prisons. Again, poor data reporting limits meaningful interpretation, and the study authors did not address this finding in their discussion.

Other Outcomes

As noted in the “Method” section, measures employed by the studies were divided into categories to facilitate assessment. It was not possible to conduct a meta-analysis of the remaining categories of measures—physical health markers, positive markers, negative markers, and attitudes toward work—due to lack of data appropriate for meta-analysis and incommensurable measures across studies. However, findings were again generally mixed.

Studies 6 and 7 included measures of physical health markers (e.g., pulse rate and blood pressure), with Study 6 finding no significant outcomes. Study 7 had mixed results, reporting significant improvements in total cholesterol, glucose, and blood pressure levels in the experimental group, but no change in other biological markers of stress, including heart rate variability, level of stress hormones, and an immunity marker. Studies 6, 7, and 9 assessed positive markers of well-being (e.g., motivation, positive outlook, relaxation), with Studies 7 and 9 reporting significant improvement postintervention. In contrast, but in keeping with its other outcomes, Study 6 reported no significant findings. Studies 1, 6, 7, and 9 also measured negative markers (e.g., Type A behaviors, absenteeism, and hostility), again with inconsistent findings. Studies 1 and 6 found no significant differences in outcomes, whereas Studies 6 and 9 reported significant reductions in each of the markers. Finally, Study 9 measured attitudes to work and reported a significant increase in positive feelings toward work among all COs who participated in the program, with greater effects present for officers with higher levels of compliance.

Discussion

This review and meta-analysis were conducted to provide some insight into the effectiveness of interventions that target CO well-being. The results of the meta-analysis indicate that overall, such programs do not significantly improve officer well-being, specifically in terms of stress and psychopathology. Those studies that were unsuitable for meta-analysis had mixed outcomes, both within and between studies. Of note, Study 6, despite a comprehensive assessment package, found no significant differences between treatment groups on any of the dependent variables. Study 5, which provided intervention to officers who had experienced trauma, reported negative outcomes for the experimental group, with participants reporting increased levels of depression postintervention. The authors of Study 5 proposed that engagement in the treatment program may have increased officers’ awareness of symptoms, resulting in higher levels of symptom reporting.

Of the nine studies, only four used a randomized controlled trial (RCT) design, thus limiting conclusions that can be drawn regarding the effectiveness of the interventions described in this review. Overall, while associations between treatments and outcomes can be suggested, we cannot exclude the possibility that outcomes were a function of other factors external to the intervention. Also, only six of the nine studies had a comparison group, thus limiting the inferences that can be made regarding the effectiveness of the interventions.

Further complicating the assessment of intervention effectiveness is the wide variety of assessment outcomes used. Across the nine studies summarized, 27 different measures were used, with only two measures being used in more than one study (Symptom Checklist-90-R and Jenkins Activity Survey). Such a variety of measures, combined with the often-limited data provided, served to severely restrict rigorous comparison of findings across studies. Such disparity of measurement highlights a particularly relevant question: What constitutes officer well-being in a prison setting? Predominantly, the studies included in this review did not clearly define or thoroughly elucidate the mechanisms affecting well-being, although such mechanisms could be implied (e.g., providing coping skills training to reduce stress).

In exploring the intervention outcomes, both positive and insignificant, no clearly identifiable patterns emerge. The lack of consistent explanatory principles could be due to several factors. First, most of the studies lacked a comprehensive theoretical model for the intervention that was specifically applied to the context of stress and well-being for COs. That is, while all studies discussed the nature and extent of stress or trauma experienced by COs, the interventions had been originally designed for other populations (usually a general population), and then applied to the CO population in the prison setting. This is problematic because of the unique nature of the setting and the stressors inherent to that setting, which do not exist in other contexts (e.g., the latent unpredictability and dangerousness, the distinctly limited autonomy, isolation, and confinement) and, consequently, are not specifically addressed by the interventions. This gives rise to the question of whether intervention failure is due to lack of theoretical development or misapplication of theory to the prison context. Equally, poor theoretical discussion or application limits possible explanations for success.

Second, none of the interventions included a planning phase where a baseline study with COs was conducted to determine the population needs prior to developing the intervention. Given the consistency across the literature in describing the stress experiences of COs, it could be argued that any intervention that targeted officer stress would be beneficial. Unfortunately, this has not proven to be the case, with some studies having either no effect or a deleterious effect on the COs involved in the study (2, 5, and 6). This inconsistency in outcomes could be due to the variation in CO stress levels when commencing treatment. For example, a new CO may be experiencing extreme levels of stress compared with a CO who has had longer tenure and more experience and lower levels of stress. Baseline measures would have provided data that would help explore any interaction between stress states and intervention outcomes.

Another benefit of a planning phase is the opportunity to determine the most effective approach to evaluation, which comprises the third factor that could explain the lack of consistency in intervention outcomes. The difference in number of measures used in each of the studies to evaluate program effectiveness was significant, ranging from two to 14. Interestingly, those interventions that reported positive outcomes used no more than three measures (excluding measures that did not require participant engagement, such as blood test results and absenteeism reports), whereas the interventions that reported no effect or a negative effect used three to 14 measures. This pattern of evaluation raises several possibilities: First, that in keeping with the complexity of defining and operationalizing well-being, there is an absence of a gold-standard measure that can be used to measure the well-being of COs in the prison context. Second, it is possible that studies that used fewer measures did not adequately capture the symptomatology, regardless of whether the intervention was successful or not. Finally, that longer evaluation forms created an effect on the participants, whereby the measures’ contextual cues, priming effects, and a consistency motif resulted in inflated or unchanged results. This effect would likely have been amplified by an intervention effect, resulting in participants becoming more aware of their emotional state and stress experiences.

Overall, the limited and widely varying approach to planning, theoretical application, and evaluation has resulted in restrictive explanations for both the successes and failures of the interventions. Without the guidance of clear, relevant theoretical frameworks, planned approaches to evaluation, appropriate implementation, and rigorous assessment of effectiveness, it is difficult to identify clear factors contributing to intervention success. Equally, it is more difficult to distinguish and measure factors that may have negatively affected or confounded outcomes.

Strengths and Limitations of this Review

This study was carried out in accordance with the systematic review method. The search strategy was generated in consultation with an expert librarian, and searches were carefully undertaken using relevant databases. To guard against subjectivity, inclusion criteria were developed and independently assessed by two researchers. However, there are several limitations that must be mentioned. First, because our study was limited to literature published in English, there is a language bias. Second, several of the studies were not published in peer-reviewed journals and as such potentially lacked the rigor that comes with such review publication.

Future Research

This review highlights the need for additional research into interventions for officer well-being. More specifically, it draws attention to the critical importance of methodological rigor. As this study demonstrates, lack of rigor results in research that lacks integrity, impact, and practical applicability. In the future, it is imperative that researchers design studies that reduce bias at every level and employ suitable, well-validated measures commensurate with treatment targets expected to be addressed by the intervention. Data should be comprehensively reported to allow delineation of effective interventions. It will also be important to establish baseline measures prior to program development, consulting the extant literature to determine factors that might function as mediators. Related to this, there needs to be a concerted effort to develop a context-specific definition of well-being, followed by the establishment of a standard measure thereof. In addition, future studies should seek to include objective measures, such as absenteeism, turnover, and sick/unplanned leave, to allow empirical and explicit exploration of well-being on observable outcomes.

Finally, there is a need for better intervention development, such that interventions are rigorously constructed, developed, and implemented. Interventions should be firmly grounded in theoretical research that applies the wider body of knowledge about well-being to the context of the prison, thus incorporating the unique stressors and organizational factors that specifically affect COs. At each point, there should be careful consideration of whether the intervention will meet the needs of the COs and of how applicable, useful, or helpful the skills learned will be in their day-to-day routine. Future work aiming to design, develop, implement, and assess interventions targeting COs should consider these findings to maximize the chances of successfully improving officer well-being within the prison environment.

Conclusion

Despite the serious concerns across the literature regarding the well-being of COs, there are few targeted well-being programs available to COs. Of those that exist, they lack a strong theoretical, context-specific basis, and none have been the subject of a robust assessment and review process. Overall, this review concluded that due to a variation of approaches, poor data reporting, and a lack of well-controlled intervention studies, there is little reliable evidence that interventions can significantly improve officer well-being in prison settings. The limited data suitable for meta-analysis showed no significant effect sizes, and while several of the studies indicated significant, positive results, their methodological weaknesses disallow definitive conclusions that the intervention itself affected officer well-being. These results highlight the need for robust planning, design, implementation, and evaluation of interventions based on sound models of psychological processes associated with well-being. Such an approach, while a challenging undertaking in a prison context, will ensure integrity of outcomes that can be built upon to make a positive, meaningful, and practical impact on officer well-being.

Supplemental Material

Supplemental_Appendix_1 – Supplemental material for Well-Being Interventions for Correctional Officers in a Prison Setting: A Review and Meta-Analysis

Supplemental material, Supplemental_Appendix_1 for Well-Being Interventions for Correctional Officers in a Prison Setting: A Review and Meta-Analysis by Trisha J. Evers, James R. P. Ogloff, Justin S. Trounson and Jeffrey E. Pfeifer in Criminal Justice and Behavior

Footnotes

Authors’ Note:

We are grateful to David Horne for his assistance with the article search process undertaken and to Ryan Veal for his assistance with quality assessment. This review was undertaken as part of Australian Research Council Linkage Project LP140100397: Enhancing wellbeing and resilience in prisons: A psycho-educational approach for the missing middle in partnership with G4S Australia and New Zealand and Swinburne University of Technology. The chief investigators are James R. P. Ogloff, Jeffrey E. Pfeifer, Michael Daffern, and Jason Skues. Dennis Roach is an associate investigator

ORCID iD

Trisha J. Evers

Supplemental Material

Supplemental Appendix 1 is available in the online version of this article at .

Trisha J. Evers is a doctoral student in the clinical forensic program at Swinburne University of Technology. Her research focuses on resilience and well-being of correctional officers in a prison environment.

James R. P. Ogloff, JD, PhD, is the foundation professor of Forensic Behavioural Science and director of the Centre for Forensic Behavioural Science at Swinburne University of Technology in Melbourne, Australia. He is also executive director of Psychological Services and Research and Research for Forensicare, the statewide forensic mental health service. He works as a researcher, clinician, educator, and administrator.

Justin S. Trounson, PhD, is a lecturer in clinical forensic psychology at the Centre for Forensic Behavioural Science (CFBS). He is a registered clinical psychologist and has worked in forensic mental health, adult correctional settings, and private practice.

Jeffrey E. Pfeifer, PhD, MLegSt, is an associate professor with the Centre for Forensic Behavioural Science at Swinburne University and sits on the Executive Board of the International Association of Corrections and Forensic Psychology. His research has been cited by the Supreme Court of Canada and Appellate Courts in Ontario and British Columbia, and he is the recipient of the 2004 and 2017 International Corrections and Prisons Association Research Awards. Most recently, he has been conducting a program of research on the use of technology and gaming as a platform for positively impacting the well-being of prison officers as well as offenders.

References

Backer

L. H.

(1990). The differential effectiveness of a stress inoculation training intervention as a function of coping style among correctional workers. Available from ProQuest Dissertations and Theses Global database. (Order No. 9029749)

Bierie

D. M.

(2012). The impact of prison conditions on staff well-being. International Journal of Offender Therapy and Comparative Criminology, 56, 81-95. doi:10.1177/0306624X10388383

Booth

(2009). Cognitive-behavioral stress management program for correctional officers. Available from ProQuest Dissertations and Theses Global database. (Order No. 3368969)

Borenstein

(2009). Introduction to meta-analysis. Chichester, UK: John Wiley.

Bourbonnais

Jauvin

Dussault

(2007). Psychosocial work environment, interpersonal violence at work and mental health among correctional officers. International Journal of Law & Psychiatry, 30, 355-368. doi:10.1016/j.ijlp.2007.06.008

Bravo-Mehmedbasic

Salcic

Kucukalic

Fadilpasic

Cakovic

Mehmedika-Suljic

Masic

(2009). Impact of psychoeducation on professional stress reduction among prison guards. Materia Socio-Medica, 21, 24-26. doi:10.5455/msm.2009.21.24-26

Brough

Williams

(2007). Managing occupational stress in a high-risk industry: Measuring the job demands of correctional officers. Criminal Justice and Behavior, 34, 555-567. doi:10.1177/0093854806294147

Brower

(2013). Correctional officer wellness and safety literature review. Washington, DC: U.S. Department of Justice. Retrieved from https://s3-amazonaws-com-443.web.bisu.edu.cn/static.nicic.gov/Public/244831.pdf

Brunetto

Teo

S. T. T.

Shacklock

Farr-Wharton

(2012). Emotional intelligence, job satisfaction, well-being and engagement: Explaining organisational commitment and turnover intentions in policing. Human Resource Management Journal, 22, 428-441. doi:10.1111/j.1748-8583.2012.00198.x

10.

Castle

T. L.

Martin

J. S.

(2006). Occupational hazard: Predictors of stress among jail correctional officers. American Journal of Criminal Justice, 31, 65-80. doi:10.1007/BF02885685

11.

Cheek

F. E.

Miller

M. D. S.

(1983). The experience of stress for correction officers: A double-bind theory of correctional stress. Journal of Criminal Justice, 11, 105-120. doi:10.1016/0047-2352(83)90046-6

12.

Dignam

Fagan

T. J.

(1996). Workplace violence in correctional settings: A comprehensive approach to critical incident stress management. In VandenBos

G. R.

Bulatao

E. Q.

(Eds.), Violence on the job: Identifying risks and developing solutions (pp. 367-384). Washington, DC: American Psychological Association.

13.

Dodge

Daly

Huyton

Sanders

(2012). The challenge of defining wellbeing. International Journal of Wellbeing, 2, 222-235. doi:10.5502/ijw.v2i3.4

14.

Dowden

Tellier

(2004). Predicting work-related stress in correctional officers: A meta-analysis. Journal of Criminal Justice, 32, 31-47. doi:10.1016/j.jcrimjus.2003.10.003

15.

Downs

Black

(1998). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. Journal of Epidemiology & Community Health, 52, 377-384.

16.

Engelmann

N. D.

(1997). The impact of critical incident stress debriefing on correctional officers. Southern Connecticut State University, Ann Arbor, MI. Available from ProQuest Dissertations and Theses Global database. (1386149 M.A.)

17.

Finn

(1998). Correctional officer stress: A cause for concern and additional help. Federal Probation, 62, 65-75.

18.

Finney

Stergiopoulos

Hensel

Bonato

Dewa

C. S.

(2013). Organizational stressors associated with job stress and burnout in correctional officers: A systematic review. BMC Public Health, 13, 82. doi:10.1186/1471-2458-13-82

19.

Garland

(2002). Prison treatment staff burnout: Consequences, causes and prevention. Corrections Today, 64, 116-121.

20.

Ghaddar

Mateo

Sanchez

(2008). Occupational stress and mental health among correctional officers: A cross-sectional study. Journal of Occupational Health, 50, 92-98. doi:10.1539/joh.50.92

21.

Grant

A. M.

Christianson

M. K.

Price

R. H.

(2007). Happiness, health, or relationships? Managerial practices and employee well-being tradeoffs. Academy of Management Perspectives, 21, 51-63. doi:10.5465/AMP.2007.26421238

22.

Hartley

Davila

Marquart

Mullings

(2013). Fear is a disease: The impact of fear and exposure to infectious disease on correctional officer job stress and satisfaction. American Journal of Criminal Justice, 38, 323-340. doi:10.1007/s12103-012-9175-1

23.

Harvey

(2014). Perceived physical health, psychological distress, and social support among prison officers. The Prison Journal, 94, 242-259. doi:10.1177/0032885514524883

24.

Higgins

J. P. T.

Green

(Eds.) (2011). Cochrane handbook for systematic reviews of interventions. The Cochrane Collaboration. Available from http://handbook.cochrane.org

25.

Huckabee

R. G.

(1992). Stress in corrections: An overview of the issues. Journal of Criminal Justice, 20, 479-486. doi:10.1016/0047-2352(92)90081-J

26.

Johns

R. E.

(1986). Occupational stress inoculation: The separate and combined effects of internal and external coping skills relevant to correctional officers at a maximum security prison. Available from ProQuest Dissertations and Theses Global database. (Order No. 8615204)

27.

MacKian

S. C.

(2009). Wellbeing. In Thrift

R. K.

(Ed.), International encyclopedia of human geography (pp. 235-240). Oxford, UK: Elsevier.

28.

McCraty

Atkinson

Lipsenthal

Arguelles

(2009). New hope for correctional officers: An innovative program for reducing stress and health risks. Applied Psychophysiology and Biofeedback, 34, 251-272. doi:10.1007/s10484-009-9087-0

29.

Misis

Kim

Cheeseman

Hogan

N. L.

Lambert

E. G.

(2013). The impact of correctional officer perceptions of inmates on job stress. SAGE Open, 3, 1-13. doi:10.1177/2158244013489695

30.

Moher

Liberati

Tetzlaff

Altman

(2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151, 264-269. doi:10.7326/0003-4819-151-4-200908180-00135

31.

Morgan

W. J.

(2009). Correctional officer stress: A review of the literature 1977–2007. American Jails, 23, 33-43.

32.

Obidoa

Reeves

Warren

Reisine

Cherniack

(2011). Depression and work family conflict among corrections officers. Journal of Occupational and Environmental Medicine, 53, 1294-1301. doi:10.1097/JOM.0b013e3182307888

33.

Pollard

Lee

(2003). Child wellbeing: A systematic review of the literature. An International and Interdisciplinary Journal for Quality-of-Life Measurement, 61, 59-78. doi:10.1023/A:1021284215801

34.

Ruck

Bowes

Tehrani

(2013). Evaluating trauma debriefing within the UK prison service. Journal of Forensic Practice, 15, 281-290.

35.

Schaufeli

Peeters

M. W.

(2000). Job stress and burnout among correctional officers: A literature review. International Journal of Stress Management, 7, 19-48. doi:10.1023/A:1009514731657

36.

Sidney

Jette

(1987). Attitudinal changes of correctional service officers following on-site physical exercise program. Canadian Journal of Sport Sciences, 12, 83-90.

37.

Spinaris

C. G.

Denhof

M. D.

Kellaway

J. A.

(2012). Posttraumatic stress disorder in United States corrections professionals: Prevalence and impact on health and functioning. Retrieved from http://desertwaters.com/wp-content/uploads/2013/09/PTSD_Prev_in_Corrections_09-03-131.pdf

38.

Stack

Tsoudis

(1997). Suicide risk among correctional officers: A logistic regression analysis. Archives of Suicide Research, 3, 183-186. doi:10.1023/A:1009677102357

39.

Tartaglini

Safran

(1997). A topography of psychiatric disorders among correction officers. Journal of Occupational and Environmental Medicine, 39, 569-573.

40.

Trounson

J. S.

Pfeifer

J. E.

(2017). Corrections officer wellbeing: Training challenges and opportunities. Practice: The New Zealand Corrections Journal, 5(1), 22-28.

41.

Trounson

J. S.

Pfeifer

J. E.

Critchley

(2016). Correctional officers and work-related environmental adversity: A cross-occupational comparison. Applied Psychology in Criminal Justice, 12, 18-35.

42.

Trounson

J. S.

Pfeifer

J. E.

Skues

(2019). Perceived workplace adversity and correctional officer psychological well-being: An international examination of the impact of officer response styles. The Journal of Forensic Psychiatry and Psychology, 30, 17-37. doi:10.1080/14789949.2018.1441427

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB