Abstract
The purpose of this study was to examine the infusion of a Toolkit for Enhancing End-of-Life Care in prisons, as well as the outcome and impact on the quality of prison end-of-life care. A total of 74 front-line staff and administrators were in attendance across two post-Toolkit-infusion evaluation visits. Applying qualitative analysis, co-researcher outcome findings were related to activities, community outreach and relations, multidisciplinary team, quality improvement approach, and participatory action research team effects. Organizational outcomes included barriers and challenges, cost, organizational features, sphere of influence, readiness (for change), and sustainability.
Introduction
The number of older inmates in state and federal prisons in the United States has been steadily increasing (Pew Charitable Trusts, 2014b). Between 1999 and 2013, the number of inmates of age 55 and older increased by 234%, from a population of 43,300 to 144,500 (Pew Charitable Trusts, 2014b). The “graying” of the prison population, combined with mandatory minimum lengthy sentence terms and little chance for parole (Anno, Graham, Lawrence, & Shansky, 2004), has resulted in many inmates not only aging in prison but also dying in prison. Health care spending levels are higher in those states with an aging prison population (Pew Charitable Trusts, 2014a). One recent action that will impact many nonviolent drug offenders incarcerated in U.S. federal prisons is the retroactive sentence reductions, which is anticipated to result in thousands of federal inmates being released early beginning on November 1, 2015 (U.S. Sentencing Commission, 2014). Also, 30 state prisons have recently shown decreases in their incarcerated populations (Montagne & Inskeep, 2015). In contrast, many state and federal prisoners are incarcerated for violent offenses and, as such, do not qualify for retroactive sentence reductions or transfers to local jails. Taken together, there remains a need for fiscally responsible solutions to caring for inmates who will age and die behind bars.
Beyond cost considerations, there is also a need for prison staff competent in end-of-life (EOL) care measures who can humanely address the needs of inmates with advanced chronic conditions approaching EOL. One notable milestone for improving EOL care in prison was the GRACE (Guiding Responsive Action in Corrections at End of Life) Project (Ratcliff & Craig, 2004). The overarching aim of the GRACE Project was to bring community quality EOL care to the prison setting (Ratcliff & Craig, 2004). Subsequent to the GRACE project, guidelines for EOL care in prisons were developed by the National Hospice and Palliative Care Organization (2009). The prison environment presents unique challenges to implementing EOL care initiatives, specifically, addressing care needs, while concomitantly ensuring the safety of the public, prison staff, and prisoners (Penrod, Loeb, & Smith, 2014). Despite such challenges, “a number of U.S. state prisons have implemented programs to care for inmates at end of life . . . ” (Cloyes, Berry, Martz, & Supiano, 2015, p. 299). However, the manner in which EOL care guidelines are interpreted and implemented varies (Hoffman & Dickinson, 2011). That is, care provided to dying inmates can vary across prisons throughout the United States (Hoffman & Dickinson, 2011; Wion & Loeb, 2016), as well as within prisons within the same state Department of Corrections (Penrod et al., 2014).
Recent research findings have described prison EOL care perspectives and experiences shared by chronically ill and dying inmates (Loeb, Penrod, McGhan, Kitt-Lewis, & Hollenbeak, 2014), as well as symptom prevalence in inmate patients cared for in a long-standing prison hospice (Cloyes et al., 2015). The values, beliefs, and perceptions of inmate volunteers and/or paid workers who provide EOL care to their dying peers also have been explored (Cloyes, Rosenkranz, Wold, Berry, & Supiano, 2014; Loeb et al., 2013; Supiano, Cloyes, & Berry, 2014; Wright & Bronstein, 2007). Yet, an unmet research need is the evaluation of EOL care interventions in prison settings from the perspective of prison administrators and front-line workers.
In an earlier phase of the authors’ broader study titled, “Enhancing End-of-Life Care in Complex Organizations: The Prison Study” (National Institutes of Health/National Institute of Nursing Research: R01 NR011874), a Toolkit for Enhancing End-of-Life Care was developed using a quality improvement approach (see Table 1 for Table of Contents) and infused in collaboration with key stakeholders (i.e., front-line staff) from six State Correctional Institutions (SCIs) in a Mid-Atlantic state using participatory action research (PAR) methods (Penrod, Loeb, Ladonne, & Martin, 2016). PAR is a methodological approach that is characterized by active collaboration between researchers and participants throughout the study—“ . . . participating with people is the way to move forward toward sustainable services that evoke human flourishing” (Koch & Kralik, 2006, p. 2). The purpose of this research article is to examine how the Toolkit for Enhancing EOL Care in Prisons was received and implemented by prison stakeholders. Specifically, the key areas of focus are policies and procedures; values, beliefs, and perceptions across stakeholders; and perceived impact on the quality of prison EOL care.
Table of Contents of Toolkit for Enhancing End-of-Life Care.
Note. Table of contents has been edited in length to better fit the format of a manuscript table.
Method
Under principles of PAR, the early focus of the project was on understanding the context (i.e., the six SCIs in a Mid-Atlantic State) and forming a research team including prison insiders as co-researchers (i.e., nurses, chaplains, corrections officers [COs], and a group-labeled counseling and coordination, which was inclusive of psychological support staff as well as unit managers). Areas for improvement in EOL care were identified and prioritized by the collective research team. Intervention strategies were developed and compiled in the Toolkit for Enhancing EOL Care in Prison (Penrod et al., 2016). Team members were oriented to the content and processes addressed in the Toolkit prior to the roll-out for implementation in the six SCIs. Following the implementation phase, techniques of qualitative outcome analysis (Morse, Penrod, & Hupcey, 2000) were applied to evaluate outcomes related to the implementation of the Toolkit. A semistructured discussion guide (see Table 2 for abbreviated version) was used to collect data via group interviews with prison staff members at the study sites.
Post-Toolkit Evaluation Questions.
Note. The discussion guide has been edited in length to better fit the format of a manuscript table; however, the major points addressed have been retained. MDT = multidisciplinary team; EOL = end-of-life; SCI = State Correctional Institution.
Participants for this study were state prison employees who took part in earlier phases of the study (e.g., the qualitative descriptive study [Penrod et al., 2014] or the PAR collaborative team of co-researchers [Penrod et al., 2016]). Participants (N = 74) were involved in the two post-Toolkit-infusion evaluation visits at each of six SCIs. Some participants attended both evaluation visits. Data were recorded through detailed handwritten field notes by the University researchers. This approach was used since as prisons engaged in the study did not permit audio-recording equipment.
Data were analyzed using techniques of content analysis. Field notes were transcribed verbatim, verified by a member of the research team, and entered into Microsoft Word documents. The word documents were read and reread by the analytic team, which consisted of a principal investigator (Author 1) and two trained research assistants (Authors 4 and 5). Initial categorical findings were developed independently by each analyst. Then, through a series of team analysis meetings, insights were shared, collapsed, and refined, and a consensus regarding categorization of the data was reached. Categories presented below were confirmed by the analytic team, each of whom read the transcripts independently, and concurred with the resultant analysis.
Findings
Outcomes of implementation of the Toolkit were broadly categorized as oriented toward the team or the organization. Team outcomes refer to those responses related to the implementation efforts of the teams of co-researchers, including activities, community outreach and relations, multidisciplinary team (MDT), quality improvement approach, and PAR team effects. Organizational outcomes refer to systemwide effects related to implementation processes, including barriers and challenges, cost, organizational features, sphere of influence, readiness for change, and sustainability. Each category is discussed in greater detail below.
Co-Researcher Team Outcomes
Co-researcher team outcomes included activities, community outreach and relations, MDT, quality improvement, and PAR team effects.
Activities
Activities were defined as any actions or strategies that were reported as being initiated as part of the collaborative University researcher/Department of Corrections EOL care initiative. Specifically, implementation of any module from the Toolkit for Enhancing EOL Care in Prisons (with the exception of building an MDT, which is addressed later in its own individual category) was categorized within activities. In addition, other EOL care strategies employed or planned that were not from a Toolkit module as well as innovative approaches were deemed activities, as were discussions of where the Toolkit was strategically placed (i.e., its location in the prison).
Activities employed from the Toolkit modules included distributing the brochures around the housing units to raise awareness among inmates of the EOL care initiative and putting a photograph and announcement about the EOL initiative in an SCI’s newsletter. In addition, participants shared that selected inmates “ . . . are being released from work so that they can attend the class [on EOL care giving],” and staff are “working to bring sick inmates to the infirmary at night and allowed them to be out in population during the day . . . to maintain friendships with other inmates and not feel isolated.” Beyond activities from the Toolkit modules, there were innovations such as “they [nurses] had been . . . going to the cafeteria and getting a bowl of soup or apple sauce or some soft foods like ice cream that they [dying patients] needed.” In general, SCI staff were reportedly “ . . . much more aware of chronically ill inmates . . . ” and one participant shared that “ . . . there’s more communication between disciplines . . . ” since the EOL project started.
Activities were also in the planning stage at some SCIs, as evidenced in the following exemplar quotes: “ . . . [we are] going to get the job-related duties of the corrections officers in the post orders . . . so that corrections officers know that the activities related to hospice care at their institution are supported” and “there was some talk about death notifications and the IT [information technology] department helped put together a calendar to insert names of inmates who get death notices and the idea here was that they would, on the one-year anniversary, they would send another sympathy card for the inmate.” Examples of locations of the Toolkit materials were in a Unit Manager’s office, which was located on one of the inmate housing units. Another location was in the administrative conference room.
Community Outreach and Relations
This category included any effort or strategy to engage community partners (i.e., free world people outside of the prison system) in EOL care activities for the prisoners. Examples included meeting with area community hospice providers, press releases, seeking volunteers, and securing equipment and/or materials. One participant noted, “ . . . Tuesday night they completed training for 11 community volunteers that signed up to be hospice volunteers. They are religious volunteers.” Roles of the community volunteers included sitting vigil, offering spiritual comfort, writing letters, and playing games with the dying inmates.
MDT
The category of MDT included all references to the leadership team (including those who were our co-researchers [i.e., the PAR team]), transitions toward an MDT, interdisciplinary team, or interdisciplinary group as different SCIs adopted different labels for this emerging group. Participants from one SCI noted that “they did not form a multidisciplinary team [however] they’re going to be working on this interdisciplinary group that the PAR RN [is] developing.” Another participant noted that they had “been in corrections for 19 years and that it’s not ‘how’ many people you have on a team, it is actually ‘who’ many and this is a really important insight as far as inmates too . . . ” In regard to members recruited to an MDT, one participant shared “as far as representatives on [the] MDT . . . [we] have counselors from the block, unit managers, corrections officers, nurses, dietary, a dentist, and a lab tech . . . no physicians volunteered . . . ”
Quality Improvement Approach
The quality improvement approach category was defined as all comments related to the quality improvement processes and/or the plan/do/study/act (PDSA) phases. One participant noted, “ . . . [I] especially like that with the PDSA paradigm . . . plans and execution don’t have to be perfect right away . . . can execute, evaluate, make changes and improve, and then re-execute a new, improved plan.” Another participant offered the opinion: “It’s [the quality improvement approach] intuitive but [we’re] not necessarily using it deliberately but . . . understand that the Toolkit’s organized around the principle and [we’re] using it as [we] go through the Toolkit.” Several participants noted that while they recognized that evaluations were important for the research project, they were not necessarily important for the institution and were sometimes viewed as being “burdensome.” There was a request to streamline the evaluation: “It would simplify things and make the Toolkit more user friendly.”
PAR Team Effects
The category titled “PAR team effects” is defined as all references to the academic partnership, the University research team, or any grant-related activities (e.g., PAR meetings/leadership development; communications with researchers). Exemplars representative of this category included “having the [specific University] project kind of legitimized what [we] were doing and gave . . . an opportunity to look at how [we] could improve things . . . working more like a team . . . that’s been really positive,” “ . . . glad [specific University ] was involved in this program . . . [this is] allowing for more formalization of what [we’re] doing,” and a report of being “very proud that they were one of the first SCIs to do hospice care and . . . that it’s been a positive relationship between them and [specific University].”
Organizational Outcomes
Organizational outcomes included barriers and challenges, cost, organizational features, sphere of influence, readiness [for change], and sustainability.
Barriers and Challenges
This category focused on all cited occurrences, attitudes, characteristics, and needs that impeded implementation of enhanced EOL care. Multiple responses in this category focused on COs (more commonly referred to in the community as prison guards). One respondent who himself was a CO indicated that “some of them [his peers] . . . worried if victims would get the type of mercy that we were proposing for inmates.” Another reported challenge “is that COs want to be seen as ethical, but not sensitive because that puts them at risk with inmates . . . if they come across as being soft, they may suffer from that.” Another barrier noted was a requirement in one SCI “in order to be a hospice patient . . . you need to sign a consent for the program, this also means agreeing to DNR [do not resuscitate] status which does not often happen.” Finally, a participant shared that “one of the inmates took the initiative to make a schedule [for inmate volunteers] for 6-2 and 2-10 shift. Unfortunately, a man died on the 10-6 shift overnight and nobody was there.”
Cost
Any insights regarding costs or resources needed to implement the toolkit (including willingness to pay estimates) constituted the cost category. Insights gained about costs of providing enhanced EOL care included “[doesn’t] feel that there has been a significant cost related to the project . . . too early to tell yet in terms of a return on the entire project . . . it’s been positive so far . . . ”; “in the long run, the cost savings of the decreased health care costs from better, more efficient inmate patient care, that . . . might actually be a big cost savings . . . ”; and “there were costs entailed with an initial day of training and maybe twice a year at most.”
Other discussions of cost focused on either staffing costs, printing costs, or perceived value of the Toolkit itself. Staffing costs were assessed as being low (e.g., time away from one’s regular work and over time). Time away from work was viewed as a nonsignificant problem, as staff attended meetings regarding Toolkit delivery at times when colleagues were available and able to cover their shift. Over time costs were typically none or very minimal across the six SCIs. Costs of paper and printing were described as being “negligible.” Finally, participants were asked how much they would pay for the Toolkit if it were commercially available. Responses ranged from US$1,000 to US$5,000. Prices were reportedly based upon the price paid for previous education materials with parallel amounts of content and materials. An insight was provided into why US$5,000 was the typical top dollar amount suggested—special authorization was needed to purchase any item costing greater than US$5,000. In contrast, a few participants reported they “couldn’t put a price on something like [the Toolkit]”; however, they went on to say that they “didn’t anticipate having trouble finding funds for it if needed.” All participating SCIs received full versions of the Toolkit, as well as supplementary resources as part of the study. Data regarding cost savings were not collected as part of this study.
On a related note, one participant viewed “participating in the program as outweighing any cost of investment.” Participants at several SCIs communicated the considerable positive effects of the EOL program with the following quote summing it up best: “[the program] has really uplifted all.”
Organizational Features
The organizational features category included any descriptions of the institution (e.g., availability of a print shop or technology support) or organizational climate (e.g., staff turnover, stability, contract negotiation, administrative support, available EOL support, and demand for services). Insights shared that fell into this category included both positive and negative organizational features. Positive exemplars included “actually have an in-house print shop where the inmates do printing and printed a lot of brochures and distributed those throughout the institution and they really did look quite nice” and “management has been very supportive . . . management pushed them . . . to keep focused on the Toolkit goals and held them accountable for moving forward.” Negative organizational features included “ . . . people have not had raises since 2008 and so, asking people to volunteer for these activities and step up and go above and beyond is very difficult . . . ” and “[we are] not sure administration is open to keeping folks who are dying in-house or whether they would prefer to continue to send them to [local community hospice].” Other responses were more neutral in nature: “You must be an infirmary volunteer for at least 6 months before you can be a hospice volunteer.”
Sphere of Influence
Any descriptors of changes in attitudes, networking, practices, or services that were attributed to involvement in the project were part of the sphere of influence category. Exemplar quotes included “ . . . we did notice posters had been put up throughout the areas [of the prison] . . . it’s generating interest . . . inmates are starting to ask about it . . . [we’re] getting volunteers”; “inmates got word out about inmates helping inmates, which was a good thing . . . ”; and “one of the inmate volunteers had suggested that they get a phone on wheels for the fellows that were infirm or approaching end of life . . . and the PAR CO took that suggestion . . . and brought that to fruition . . . ”
Readiness [for Change]
Characteristics perceived to facilitate initial uptake of the Toolkit constituted the category titled “readiness for change.” One participant shared that “more staff are becoming enlightened . . . the staff [are] really responding well to the PAR team and . . . each terminal case . . . the more . . . they can do . . . ” Another respondent indicated their SCI had experienced readiness a bit earlier: “ . . . attitudes were already here and the foundation was already here prior to [the EOL] initiative taking place so . . . the seeds of change had already been planted . . . ” A bit of a different perspective was noted in the following sentiment: “You need support from the top down. The inmate population of the institution will really dictate whether it’s [the EOL program] needed.”
Sustainability
Any insights, recommendations, and concerns regarding long-term engagement in EOL activities were included under the category of sustainability. One participant astutely noted, “ . . . the key to sustainability is that training can’t stop here . . . you must bring new people on board, on to the team . . . you need a dynamic leader who’s supportive of and invested in keeping the program going.” A contrasting perspective was noted in the following quote: “ . . . if it’s not laid out and spelled out, they’re probably not going to do it . . . really need a policy to keep things going on in some SCIs and the more concrete that you can be the better.”
Discussion
Qualitative outcome analysis provides critical insights to advance our understanding of implementation processes and outcomes in prisons. Findings from the current study highlight the importance of systemwide buy-in for successful EOL care initiatives. Top-down endorsement from the superintendent or warden is essential (Penrod et al., 2014; Ratcliff & Craig, 2004); however, it may not be sufficient for promoting sustainable change because inmate need, interest, and investment are also important. The latter of which is true particularly when inmate caregivers are part of the EOL services. Taking critical next steps such as implementing institutional policies that promote humane EOL care (Penrod et al., 2014) and communicating sanctioned EOL practices (Wright & Bronstein, 2007) in the COs’ post orders can promote more widespread adoption and help to overcome human barriers to care. At the same time as implementing the aforementioned “stick” type of initiatives, SCIs can also provide “carrots” to promoting EOL care, such as publicly acknowledging a prison’s MDT of EOL caregivers for safely and successfully adopting and/or refining EOL care provided to dying inmates. The sharing of real-life success stories (e.g., deaths without expensive futile care; engagement of inmate peer caregivers; community involvement) can bring a program to life for those who have not yet embarked on enhancing EOL care in their prison (Bronstein & Wright, 2007; Loeb et al., 2014). Stories should be shared across all relevant occupational categories (e.g., nurses/medical, COs, chaplains, and the group we labeled counseling and coordination) by individuals representing each of the constituent groups, as people relate better to those who are more, rather than less like themselves. Through this exposure, those not currently on an EOL team can get a better understanding of the importance and value of humane EOL care in prison (Loeb et al., 2014). Similarly, SCI leaders may be acknowledged in front of their peers at statewide superintendent meetings or correctional health care administrator meetings—further extending the sphere of influence, by exposing peers in leadership positions to the benefits of better caring for those dying under their charge.
The strategy of sharing real-life experiences also holds potential for engaging individuals from a particular occupational category that might be missing from an existing MDT (e.g., in the case of the study described herein—a physician). Beyond live meetings, perhaps recorded testimonials from dedicated individuals from a variety of occupational categories on the benefits to care recipients and the overall prison environment from their involvement in a prison EOL MDT could prove a beneficial resource for promoting prosocial motivation—the driving force for engagement being the potential benefit to others rather than for self (Hu & Liden, 2015). Similarly, if an MDT member was recruited but not particularly engaged in the MDT, expressions that are reflective of prosocial motivation for engaging in EOL care in prisons could be contagious throughout the MDT due to what Hu and Liden (2015) describe as the important social context of being a team member.
Knowledge gained from this study will inform future modifications to the intervention, Toolkit for Enhancing End-of-Life Care in Prisons, so that it is in the most user-friendly format for broad dissemination. Transformation of the Toolkit is currently underway to convert the paper and electronic file Toolkit into media-rich and interactive computer-based learning modules, which would eliminate the heavy reliance on prison human resources for delivering training and allow maximal flexibility in terms of individual employee (or inmate) completion of training.
Conclusion
Through this examination of outcomes, important lessons regarding implementation of a systemwide intervention program were highlighted. The value of careful assessment of need and readiness for change as important steps prior to implementation of new programming or services cannot be overestimated. The most successful teams have strong leaders, are continually evolving, and have the right people fulfilling the right roles. These insights emphasize that careful attention to leadership development and team building activities are warranted in systemwide intervention programs. Leveraging resources that are readily available and largely avoid additional institutional costs (e.g., inmates, community volunteers, enthusiastic and motivated staff, and prison industries) are key strategies for achieving fiscally responsible change in organizations with limited budgets. Finally, the reality of the context of corrections (i.e., a 24-hr operation) has a profound effect on implementation processes. Much happens during the overnight shift (e.g., 10:00 p.m.-6:00 a.m.); therefore, extending intervention strategies to address the needs of the 24 hr a day, 7 day a week operations of prisons is essential.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The National Institutes of Health/National Institute of Nursing Research grant # R01 NR011874. The views presented in this article are those of the authors and do not necessarily represent the views of the NIH/NINR.