Currently, a large number of inpatient mental health settings in the United States provide services that are still based on a traditional model of care that is uninformed by recovery-oriented principles, trauma, person-directed care, or evidence-based practices (National Association of State Mental Health Program Directors [NASMHPD], 2014b; Wale, Belkin, & Moon, 2011). This traditional inpatient culture of care can be characterized by (a) paternalistic attitudes; (b) staff-to-consumer power struggles; (c) one-size-fits-all treatment practices; (d) a lack of voice from individuals being served; (e) indiscriminately applied rules resulting from attempts to control behaviors; (f) practices based on intuition rather than evidence; and (e) often blatant discrimination expressed through language, practices, and policies in inpatient settings (Anthony, Cohen, Farkas, & Gagne, 2002; Huckshorn, 2004; NASMHPD, 2014b). Traditional mental health practices include (a) talking or writing about patients as though they were a disease (e.g., schizophrenic, borderline); (b) providing documentation in medical records that reduces requests for help to adjectives, such as needy, manipulative, attention seeking, or intrusive; (c) enforcing idiosyncratic rules that expect all patients to go to bed at fixed times regardless of sleep patterns; (d) requiring mandatory wake-up times mostly for staff convenience; and (e) instilling expectations that individuals passively accept treatment team recommendations that they had no role or voice in developing (NASMHPD, 2014a; Robins, Sauvageot, Cusack, Suffoletta-Maierle, & Frueh, 2005).
Inpatient mental health service culture can be experienced as patronizing, shaming, and disrespectful, and such an environment can lead to patients’ distrust, which, in turn, results in poor clinical care (Bluebird, 2004; Huckshorn, 2007). Previous research studies have focused on patient experiences that documented the unnecessary use of force, a lack of respect from staff, feeling punished but not understanding why, and feeling not listened to by or having quality time with staff (Allen, Carpenter, Sheets, Miccio, & Ross, 2003; Ray, Myers, & Rappaport, 1996). This research topic is relatively unstudied, but it is the current author’s experience that these situations still widely exist. Therefore, it should not be surprising that staff practices in these treatment environments can result in conflicts, threatened or real violence, and staff responses that include the use of restraint and seclusion (R/S) (Callaghan, Nijman, Palmstierna, & Oud, 2007).
The use of R/S interventions are controversial and potentially dangerous staff practices that are used in most inpatient mental health care settings to control an individual patient’s behavior that is deemed to be dangerous (American Psychiatric Nurses Association [APNA], 2014; National Association of Psychiatric Health Systems & American Hospital Association, 2003; NASMHPD, 2014b). Seclusion is defined as the “involuntary confinement of a person in a room or an area where the person is physically prevented from leaving” (U.S. Department of Health and Human Services [USDHHS], 2006, 13[f]). A seclusion event begins when the individual is escorted, usually involuntarily, by staff to a room that is locked or unlocked where he or she is isolated from others and physically prevented from leaving; the event ends when the individual is informed that he or she can leave on his or her own volition.
Restraint is defined as a manual (i.e., physical) hold or mechanical device, material, or equipment attached or adjacent to an individual’s body that is not easily removed and restricts the individual’s freedom or normal access to one’s body (NASMHPD, 2014a; USDHHS, 2006). Mechanical restraint practices include leather or plastic cuffs for extremity immobilization to a bed or cot, wrist-to-waist restraints or ankle hobbles, lap belts, lap trays attached to chairs, restraint chairs, and Posey vests. Physical or manual holds are mostly, but not always, used on children and adolescents (NASMHPD, 2014a; USDHHS, 2006). It has been noted that only in health care settings is the human and civil right to freedom displaced without oversight by a law enforcement officer, judge, or jury (NASMHPD, 2014a).
Restraint and Seclusion Practices
R/S practices have received a much greater level of interest, oversight, and regulation by legislators and policy makers in recent years (APNA, 2014; USDHHS, 2006). In part, this interest initially emerged in response to hospitals not reporting serious injuries and deaths in inpatient mental health service settings. An exposé in The Hartford Courant (Weiss, Altimari, Blint, & Megan, 1998) and in other investigations (U.S. General Accountability Office [USGAO], 1999), coupled with subsequent research, have shown that R/S are in and of themselves aggressive and violent acts and have a palpably negative effect on the quality of care provided. In addition, they create unsafe conditions for both patients and staff, lower staff morale, and increase staff turnover (NASMHPD, 2014a). The use of coercive and invasive practices in settings expected to be safe and recovery-oriented interrupts and negates the necessary trust and treatment alliance that is fundamental to successful engagement with individuals in care (NASMHPD, 2014b).
R/S practices are most commonly initiated by inpatient mental health staff in response to unique clinical situations, which are generally described as dangerous patient behaviors that are experienced by staff as threatening (NASMHPD, 2014a; USGAO, 1999; Weiss et al., 1998). Situations in which R/S are used often have similar origins: the emergence of a conflict between an individual patient and staff member, or between patients themselves, that escalates and results in a staff member deciding that only R/S can control the situation. Although regulatory language exists that is intended to limit the use of R/S to situations that are characterized by imminent danger, this language is highly subjective and leaves decision making up to individual direct care staff, many of whom are relatively inexperienced and hold 2-year college degrees or less (NASMHPD, 2014a; USGAO, 1999; Weiss et al., 1998). Overall, the policy and practice of R/S use is not supported by credible science-based evidence, except as a safety measure of last resort (Institute of Medicine, 2005; NASMHPD, 2014b).
The principle investigator (K.A.H.) of the current study explored the described experiences of leaders and staff in mental health inpatient facilities who had directed or participated in successfully reducing the use of R/S in their facilities. Due to a lack of empirical knowledge about the culture of care that supports the use of R/S, the current study used a phenomenological methodology instead of a quantitative method, as this was believed to be the best method to capture the lived experiences of leaders and staff who successfully reduced the use of R/S in their hospital work environments.
The research questions for the current study were:
- How did inpatient mental health hospital leaders and staff describe the experience of successfully reducing the use of R/S?
- How did mental health hospital leaders and staff change their organizational culture from one that used R/S to one that was able to implement new practices to avoid the use of R/S?
- What strategies did mental health hospital leaders and staff use to implement successful organizational change, and how did they communicate these strategies to staff?
The current study’s data collection began with the selection of two state-operated mental health facilities, which were chosen based on the state mental health office’s agreement to participate in the current research project and because they were geographically accessible for the author. The facilities provide services to different populations: one facility provides child, adolescent, and adult services, and the other facility provides only adult services. The primary key informants in each hospital included leadership staff holding executive, senior, or middle management positions, in addition to direct care staff. The phenomenon of interest and focus of the current study, common to all participants, was their personal experiences in either directing or participating in an organizational change process that resulted in a successful reduction in the use of R/S (i.e., a ≥65% reduction from each hospital’s own baseline). Participant recruitment was based on purposeful sampling, a strategy that researchers seeking to understand a specific experience use and one that only includes participants who have first-hand knowledge of the phenomenon under study (Patton, 2002).
The interview questions were initially broad-based and further refined during the research process (Mertens, 2005). Thirty-three semistructured, open-ended questions emerged through the researcher’s literature review and experiences. This semistructured interview protocol was field tested with three chief operating officers (COOs) and one medical director in other hospitals who significantly reduced the use of R/S in their respective facilities. Phenomenological methods require an active researcher role, beginning with gathering, audiotaping, and transcribing interviews; reviewing data from individuals who have lived the phenomenon of interest; and analyzing and interpreting these data. This analysis includes both “dwelling with the data” and bracketing personal beliefs and biases during the research process (Patton, 2002; Speziale & Carpenter, 2006).
The researcher, with help from each hospital’s chief executive officer (CEO), identified senior leaders, middle managers, and direct care staff (N = 21) who had the most indepth information regarding the focus of the current research. The senior leaders identified for inclusion included the CEO, COO, medical director, chief nursing officer, director of consumer affairs, performance improvement director, and other senior leaders. A variety of middle managers and direct care staff were also interviewed. Because facilities may designate different titles for these positions, the terms senior leader, middle manager, and direct care were used to designate staff levels. Many of those interviewed held nursing staff positions or were licensed nurses.
The next step was to locate general concepts or statements within the participant responses that spoke directly to the lived experience (Patton, 2002). These key phrases were then studied as to their meanings from the informed stance of the researcher. To complete this step, each participant’s response was carefully analyzed in regard to every interview question.
Participant responses to each question, also called thematic excerpts, were analyzed to identify commonalities. Agreement among at least three of the 21 participants was considered a response commonality. Commonalities that were discovered were then further analyzed against the original thematic excerpts from the transcripts. In the end, 115 initial common themes were identified.
The researcher engaged the services of a statistician and used a kappa coefficient statistical process to measure the level of agreement between two external raters who were asked to review the work to date. The external expert raters found 98 common themes (of the original 115) on which they agreed. The kappa coefficient process is a conservative statistical measure of interrater reliability or agreement that is used to assess qualitative documents and determine agreement between raters (Trochim, 2000).
Of the agreed on initial 98 potential themes, 32 themes that reached at least fair (i.e., 21% to 60%) agreement were synthesized into five “meaning theme” statements, which were used to write textural, structural, and composite descriptions of the research phenomenon (Creswell, 2007, p. 62). These three descriptions are focused on the common experiences of the interviewed participants through the development of the underlying structure of the lived experience. These descriptions include three long paragraphs, from which a reader should walk away feeling, “I understand better what it is like for someone to experience…”(Creswell, 2009, p. 62).
The last activity of the data analyses constituted synthesizing the five meaning themes and narrative descriptions into six significant findings, which represented the final analysis of what the participants in the current study experienced to successfully reduce the use of R/S. This final step provided a rich and indepth understanding of the phenomenon under study, as experienced by several or more key informants (Creswell, 2007).
The two state hospitals that provided the inpatient settings for study are more similar than different. Both hospitals serve clients in civil and forensic settings, are publicly funded and managed by the state government, hold accreditation from the Joint Commission, and are certified by the federal Centers for Medicare & Medicaid Services (CMS). Both hospitals demonstrate highly similar staffing patterns, types of staff, patient–staff ratios, lengths of stay (from 6 months to 1 year), and admission and discharge activity. Differences include capacity and type of clients. Hospital B serves a larger number of clients (N = 192), including adolescents, and Hospital A has the capacity to serve 169 adults. Both hospitals employ a similar number of staff (Hospital A, N = 451; Hospital B, N = 532), given Hospital B’s slightly larger size.
Thirty-two themes were extracted from the raters’ review of the original 98 identified themes (, available in the online version of the article). These themes were further analyzed and reduced to five meaning themes, including:
- State and hospital leaders took on critical roles in successful R/S reduction projects in their respective hospitals.
- Leaders and key hospital staff had to change their beliefs and behaviors about the use of R/S throughout the project.
- Leaders and key hospital staff had to identify and operationalize new interventions to prevent the use of R/S throughout the project.
- Leaders and key hospital staff needed to identify and resolve key challenges to the reduction effort.
- Hospital leaders and key staff were able to report important lessons learned as a result of this process and indicate what they would do differently next time.
A textural description of the phenomenon of interest is a description of “what the participants in the study experienced with the phenomenon,” (Creswell, 2007, p. 159) as reported to the researcher. This description also helps answer research Question 1 (i.e., “How do inpatient mental health hospital leaders and staff describe the experience of reducing successfully the use of R/S?”).
The participants in the current study described their experiences in successfully reducing the use of R/S as a “project that needed to be led by state- and hospital-level executive leaders who were able to change the way that seclusion and restraint were viewed by staff.” Staff described the change in the use of R/S from a baseline of being an “unquestioned, culturally based, practice norm” to being an “event to be avoided if at all possible” by reducing opportunities for staff-to-client conflicts resulting from hospital rules and old beliefs.
Hospital staff in both facilities met the initial announcement of the goal to reduce R/S with mixed reactions, ranging from negative, to skeptical, to being welcomed. Over time, most hospital staff learned new skills to avoid R/S use and demonstrated new beliefs, as evidenced by the data that show that R/S are now rarely used and only for dangerous behaviors (Figure 1 and Figure 2). Key challenges, such as a lack of resources, communication issues, staff uncertainty in practicing new ways of working, and an initial negative reaction to change, are common barriers to implementing organizational changes.
Hospital A data for years 2003 to 2008 regarding restraint/seclusion and use of involuntary medicine by (A) patients, (B) hours, and (C) episodes. FY = fiscal year.
Hospital B data for years 2003 to 2008 regarding restraint/seclusion and use of involuntary medicine by (A) patients, (B) hours, and (C) episodes. FY = fiscal year.
A structural description “reflects on the setting and context in which the phenomenon was experienced” (Creswell, 2009, p. 161), as reported by study participants. This statement also helped answer research Questions 2 and 3 (i.e., “How did mental health hospital leaders and staff change their organizational culture from one that used R/S to one that was able to implement new practices to avoid the use of R/S?” and “What strategies did mental health hospital leaders and staff use to implement successful organizational change, and how do they communicate these strategies to staff?”).
At the beginning of the R/S reduction project, participants characterized the two hospitals’ organizational cultures as believing that the use of R/S was a normal practice, part of usual staff practices, a way to efficiently control clients, a way to keep the units safe, and the only available option for staff. The introduction of a new approach that engendered new thinking about R/S occurred initially through specific and credible external training; this training was seen as a key change agent. As the project progressed, it became a performance-improvement process, in which leaders and staff could carefully analyze events and learn how to improve their responses to potential conflicts. Challenges were identified during the project, including a lack of resources, difficulty in communicating in a timely way with all staff, delayed responses to staff concerns, staff uncertainty in trying new approaches with clients, attempts to include consumers in the change process, and leaders acting in a way that made staff believe they were being blamed or criticized. Success in both hospitals occurred as a result of the involvement of key leaders at the state and hospital executive level and direct care staff who led by example (i.e., core group of staff who were willing to risk and change their behaviors and help other staff change their behaviors). Lessons learned included the need to involve clients sooner and manage staff concerns more effectively and rapidly.
A composite description synthesizes the “textural and structural descriptions and serves to describe the essence of the phenomenon” (Creswell, 2007, p. 161).
The successful reduction of the use of R/S in mental health settings requires an organizational culture change that begins with key executive organizational leadership staff acting as change agents; the ability of leaders and staff to change their beliefs and behaviors as new information is accumulated about what works; and the ability of leaders and staff to practice and model success, resolve challenges, and incorporate lessons learned along the way.
Hospital staff members’ beliefs at the beginning of the project were that (a) R/S use was a normal practice in the hospital, (b) R/S use kept the units safe, (c) R/S use was efficient, and (d) R/S interventions were ones that staff could control.
Hospital staff members’ beliefs about R/S significantly changed during the project time frame and came to be viewed as “practices that could be avoided by reducing opportunities for staff-to-client conflicts by minimizing hospital rules, learning new skills, adopting a prevention approach to conflict, including patients in the project, and finding champions among the direct-care staff to help,” as described by one study participant.
The work to reduce the use of R/S resulted in a number of challenges, as reported by hospital leaders and staff when they implemented change. These challenges included a reported “lack of resources, communication issues, resistance from some staff, and leadership behavior that was interpreted as ‘blaming staff’ for the use of seclusion or restraint that were important to manage in this study,” as detailed by one participant.
Approaching the R/S reduction project through a performance-improvement lens was helpful, as it avoided blame and focused on what worked. The use of data to direct practice changes was a key component of performance-improvement work in the current study.
Lessons learned by hospital leaders and staff included the following: if they were to repeat this project, they would immediately focus on staff development and training, manage staff concerns better and faster, involve clients immediately in the project, regularly show R/S data to direct care staff, and avoid anything that could be interpreted as blaming direct care staff.