Sustaining Culture Change: Experiences in the Green House Model

Abstract

Objective

To describe conditions that influence how Green House (GH) organizations are sustaining culture change principles and practices in a sample of GH skilled nursing homes.

Data Sources/Study Setting

Primary data were collected at 11 skilled nursing GH organizations from 2012 to 2014. These organizations have adopted the comprehensive and prescriptive GH model of culture change.

Study Design

To develop an understanding of sustainability from the perspective of staff who are immersed in GH daily work, grounded theory qualitative methods were used.

Data Collection Methods

Data were collected using semi‐structured interviews with 166 staff and observation of house meetings and daily operations. Data were analyzed using grounded dimensional analysis.

Principal Findings

Organizations varied in their ability to sustain GH principles and practices. An organization's approach to problem solving was central to sustaining the model. Key conditions influenced reinforcement or erosion of GH principles and practices.

Conclusions

Reinforcing the GH model requires a highly skilled team of staff with the ability to frequently and collaboratively solve both mundane and complex problems in ways that are consistent with the GH model. This raises questions about the type of human resources practices and policy supports that could assist organizations in sustaining culture change.

Rooted in the 1987 Omnibus Reconciliation Act, the U.S. nursing home culture change movement is an effort to deinstitutionalize long‐term care. The movement is based on six principles: resident‐directed care and activities; a homelike environment; close relationships among staff, family, residents, and community; staff empowerment; collaborative and decentralized management; and measurement‐based continuous quality improvement (Koren 2010). The movement continues to gain momentum nationally. A 2007 Commonwealth Fund survey found 56 percent of nursing homes self‐reported having adopted at least some aspects of culture change (Doty, Koren, and Sturla 2008). A 2009–2010 survey by Miller et al. (2013), using similar definitions, found that number had risen to 85 percent.

The rate of culture change adoption has been influenced by federal mandates (CMS 2009; U.S. Patient Protection and Affordable Care Act 2010) and state incentives. States have supported an estimated $61.65 million in financial investments and incentives for culture change over a 12‐year period (Bryant, Stone, and Barbarotta 2009). At a national level, the Green House (GH) model has gained considerable attention, being touted as one of the most comprehensive culture change models. Many states are granting waivers of bed moratoriums to support building GH homes (Leading Age 2013; Green House Project 2014), and several foundations offer low‐cost financing to promote growth of the model (Green House Project 2013). With these significant investments in culture change, it is important to look at the ability of organizations to sustain the model once implemented.

The Green House Model

Since the first GH opened in 2003, 33 organizations in 27 states have adopted the model (Cohen et al. 2016), including three U.S. Veteran's Affairs Centers. Each GH home in an organization has 8–12 elders (residents) and Shahbazim (self‐managed teams of certified nursing assistants, or CNAs, who serve as “universal workers”), who manage daily operation of the house. This is commonly referred to as a “household” model of culture change (Grant 2003), where all core functions of the unit (e.g., laundry, dining) are self‐contained. Nurses assigned to homes provide nursing care and supervise the clinical care provided by Shahbazim. A Guide, whose office is outside the home, supervises the non‐clinical aspects of Shahbazim work, acting primarily as a coach. GHs are commonly opened alongside a “legacy home,” the organization's original, more traditional nursing home. The legacy home provides support services to the GHs (e.g., rehabilitation services, maintenance, human resources, dietary oversight). The GH Guide is sometimes also the legacy home administrator or a legacy home department head.

To become a GH home, adopters must agree to specific architectural requirements (e.g., freestanding units with no connectors, inclusion of a large dining table to seat all residents) and workforce practices (e.g., universal workers who fulfill specific roles, a focus on resident preference). Other requirements include ensuring the GH becomes “home for life” for each elder as long as he or she wishes. The GH model promotes total transformation of the organizational culture and environment as the key to success (Thomas 2004). An organization must adopt all of the specified elements to be opened under the GH trademark (or negotiate with the national office). At start‐up, GH adopters receive abundant implementation guidance from the national GH program office, with opportunities for financial feasibility modeling, regulatory advocacy, architectural guidance, staff training, and ongoing technical support as requested. However, once a GH is open, it is unclear how the elements of the model are sustained. At the time of publication, there are currently no GH model fidelity systems and use of ongoing GH support mechanisms is optional.

Culture Change Sustainability

Green House adopter guidance has been primarily focused on initial implementation, similar to the focus of many culture change initiatives (Rahman and Schnelle 2008). Similarly, most research on culture change has focused on the initial implementation process (Stone et al. 2002; Scalzi et al. 2006; Shield et al. 2014; Shier et al. 2014; Tyler et al. 2014) or small sample studies of staff and resident outcomes (Coleman et al. 2002; Stone et al. 2002; Bergman‐Evans 2004; Kane et al. 2007; Miller et al. 2014). Relatively little is known regarding how organizations sustain principles and practices of culture change over time. In a study of three culture change nursing homes, Scalzi et al. (2006) described barriers and facilitators to sustaining culture change. Barriers included nurses being excluded from culture change activities, perceived regulatory compliance emphasis, an organizational focus on the “bottom line,” and high administrative and direct care worker turnover. Facilitators included a critical mass of change champions, participative leadership style, effective staff communication styles, mutual values and goals, resident and family involvement in the changes, and high levels of staff empowerment. However, the study provided little insight to exactly how these factors influenced sustainability of the culture change.

In a study of a coalition of 11 nursing homes that adopted the Wellspring model of culture change, Stone et al. (2002) found that sustainability of the model was “extremely challenging” and a return to former practice patterns was common. Central to sustaining Wellspring were a committed coordinator who could maintain momentum and problem solve with the team; continued engagement of the care team in model activities; widespread implementation of the changes; adequate staffing to carry out model activities; low staff turnover; effective staff relationships; and an engaged administrative team.

Organizational change theories have informed the guidance provided for culture change initiatives (Weiner and Ronch 2003; Bowers et al. 2007), but less attention has been given to these theories in terms of sustaining culture change. In a synthesis of organizational change literature, Buchanan et al. (2005) describe six categories of factors that influence the ability of an organization to sustain change: organizational (e.g., supportive policies), cultural (e.g., receptivity), political (e.g., powerful stakeholders), individual (e.g., appropriate skills), managerial (e.g., management style), and leadership (e.g., vision). Senge (1990) introduced the notion that sustainability is not a discrete stage in organizational change but is part of a cycle that is intertwined with other processes, noting that commitment to group problem solving and tackling the causes and outcomes of underlying problems are key to sustaining organizational change (Senge 1990; Senge and Roth 1999).

While limited prior research on sustaining culture change in nursing homes identifies barriers and facilitators, the research and practice community can benefit from a deeper understanding of how the complexity of daily practice influences the sustainability of culture change principles and practices. The purpose of this study was to describe how the principle and practices of the GH model of culture change are currently being sustained at a sample of GH skilled nursing homes, focusing on conditions that influence reinforcement or erosion of the model.

Methods

Qualitative methods were used for this study, as appropriate for describing complex real‐world phenomena in health services research (Bradley, Curry, and Devers 2007). Semi‐structured interviews and observations were completed during site visits (each 2–3 days in length) at 11 GH organizations. Through systematic examination of transcripts and field notes using grounded dimensional analysis, a conceptual framework was developed that describes staff experiences sustaining the GH model. The study was approved by the University of Wisconsin‐Madison IRB.

Sample

Staff (N = 166) at 43 individual homes within 11 GH organizations participated in this qualitative research study. Participants included Shahbazim (N = 54), nurses (N = 30), and other staff (N = 82), including therapy providers, administrators, Guides, and department heads working across the organization (both in the GHs and legacy homes). The organizations were selected to represent variation in key characteristics across the population of 19 skilled nursing GH organizations operating as of December 31, 2010 in the United States: length of time of operation (range: 2–7 years), number of GH beds (range: 20‐90), and urban (N = 3) or rural setting (N = 8).

Data Collection

Through semi‐structured interviews, staff were asked to describe what is important about the GH model, where it has been difficult to adhere to the model, and how and why they might have deviated from the model. While an initial structured question was used to start the conversation about an interview topic (e.g., “How often does the Guide come to the house? What does he/she do during these visits?”), researchers then followed up with unscripted prompts for elaboration (e.g., “You mentioned the Guide helped with whatever you need. What are some examples of things you've needed the Guide's help with?”). Interviews were recorded and transcribed for analysis.

Data were also collected through observation of staff meetings at six of the GH organizations. Due to scheduling difficulties, meetings could not be observed at all 11 organizations. Field notes were taken by the two‐person research team during GH unit meetings with a focus on who was present and involved, the type of issues discussed, how issues were contextualized by staff, and the process used to address issues. Additional field notes of general observations related to GH sustainability were also taken.

Data Analysis

Data were analyzed using grounded dimensional analysis, a theory‐generating approach derived from grounded theory (Glaser and Strauss 1967; Strauss 1987) and dimensional analysis (Schatzman 1991; Caron and Bowers 2000; Bowers and Schatzman 2009). The benefit of grounded dimensional analysis is that it provides a rigorous method to generate a theory of real‐world processes as described by research participants (Kools et al. 1996). In this method, data are coded and analyzed as they are collected in order to focus on and explore concepts and relationships as they are revealed. Later data collection is then transformed based on this early analysis (e.g., interview questions change, potential key variables are observed more closely) and more focused analysis occurs. The result is a model, typically a diagram, of relationships among processes and the conditions that influence them. The model is grounded in the views of those who are embedded in the organization and its processes.

Participant interviews were transcribed verbatim and, along with field notes, entered into NVivo10 (QSR International). One member of the research team (“primary analyst”) conducted analysis of the first few transcripts and field notes, developing an initial set of broad codes (“open coding”). These codes were presented to the four‐person research team for discussion, resulting in new, more highly focused, analytic questions. The primary analyst then returned to the data with these questions, conducting more open coding and some early coding of relationships among key concepts (“axial coding”). The researchers produced memos and diagrams that described evolving categories and categorical relationships, resulting in subsequent focused coding (“selective coding”) and revised diagrams. This cycle repeated itself as the data set grew with each site visit.

Results

At the time of our site visits, the GH homes showed varied levels of adherence to the principles and practices central to the GH model. Some were clearly working in ways that reinforced these principles and practices, while others appeared to have regressed to more traditional ways of operating, causing these principles and practices to erode. Where the model was being reinforced, staff were able to talk about how they used GH principles and practices in everyday thinking and actions. For example, staff in some homes described how despite challenges, they maintain a consistent commitment to empowered staff (e.g., Shahbazim self‐scheduling, ability to make spontaneous changes to menus or activities without complex approval). Erosion of GH was characterized by reversion back to traditional institutional ways of thinking or working. For example, staff at some homes described how they had once resisted the use of medication carts, but had reverted back to using the cart for staff convenience. This process was seen as antithetical to the GH principle of creating a homelike environment and putting the elder at the center of decision making.

An organization's continued adherence to GH principles and practices thus appeared to be dependent on how the organization responded to challenges as they arose. We came to conceptualize sustainability not as a steady state, but as an ongoing series of decisions and actions taken in the course of solving problems. Some patterns of problem solving clearly reinforced adherence to GH principles and practices, while others more commonly led to erosion. In certain instances, however, it was difficult to clearly discern when an action was reinforcing or eroding the model. GH adopters were sometimes willing to challenge the practices dictated by the GH model if they believed there was a strong case for a different way of practice that was driven by elder needs. This finding left the research team to view some problem solving as reinforcing a GH principle, while simultaneously eroding a GH practice.

Patterns of Problem Solving

Three distinct patterns of problem solving at GH organizations emerged from the data, each leading to reinforcement, erosion, or a mixture of reinforcement and erosion of the GH model. Each organization appeared to demonstrate a dominant pattern, which was applied across problems.

Coached Collaborative

In this pattern, problem solving takes a team approach, with Guides and other members of the clinical support team mentoring and supporting Shahbazim to identify and analyze problems and to devise and test solutions. The problem‐solving process is grounded in GH principles and the solutions devised are consistent with GH practices.

As your acuity changes, and it kind of rocks the whole place and sends everybody into almost panic mode. It's like, “we can't do this. We can't do all the things. Just bring in housekeepers, bring them in.” We just refuse to because that's not what the model says. … I mean, it's just been, “okay, we're not going to change it. We can do this. So how are we going to make it work with who we've got?” And sitting down together and looking at, okay, we need to change how we're staffing, what hours, because these hours are heavier, with the acuity, they're even heavier hours. Might be the middle of the afternoon, 2:00–3:00 p.m., but we look at a pattern, identify the weaker hours, and that is where we try to adjust. [GH Guide]

Management Led

In this pattern, problem solving is dominated by the Guide and other managers, who work together as a team to identify and analyze problems and to devise solutions. Their level of commitment to GH is often high, although most Shahbazim are not active problem solvers. While the process of problem solving diverges from GH principles and practices, the solutions may or may not be consistent with these principles and practices, although the process of problem solving itself is inconsistent with the commitment to inclusion.

Well, there was a girl that was working a shift. It wasn't working out for her. She was working the night shift, and she was going to have to quit, because she had children at home. She didn't feel like she could leave them at home. So I wanted to keep her. She was a good worker. And I was trying to work out with HR how we could work it out for her to keep her here. And in the meantime, you know, I didn't include the shahbaz like I should have. And so, you know, that was not a good thing. [GH Guide]

Hierarchical

This pattern of problem solving is siloed and often adversarial, with the process rigidly divided by disciplinary and hierarchical authority. Shahbazim do not engage in problem solving due to lack of opportunity or interest, or need for extensive coaching to problem solve. Problem identification, analysis, and solution implementation are driven by the Guide, Director of Nursing, and other managers, who are often uncommitted to, or even hostile toward, the GH model. Problem solutions frequently run counter to GH principles and practices.

And in the interim, [an administrative staff person] kind of just took over and said, well, we're going to start doing this, and told everybody they can just go in when they want to, I don't have to knock [on the GH doors] anymore. [GH Guide]

Conditions Associated with Reinforcement

Whether or not problem‐solving processes, and the solutions devised, led to reinforcement of the GH model depended on several contextual factors (Figure 1, “Conditions”). These factors, in isolation, were not guaranteed to lead to reinforcement of the model. However, they set the stage for sustaining GH principles and practices.

Figure 1.

Supporting Sustainability of Green House Principles and Practices

Shahbazim Individual Capacity, and Opportunities, for Problem Solving

While not all problem solving falls to the Shahbazim, day‐to‐day management of the home and its activities are, according to the GH model, the responsibility of the Shahbazim. In the GH model, it is not simply that the solution to a problem needs to support the model, but it is believed that an empowered front line staff is pivotal to devising the best solution. Reinforcement of the model was supported when Shahbazim were encouraged by the Guide, nurses, and other staff to discuss problems and solutions, and felt supported by organization leaders to implement solutions. This process required the Guide, nurses, and administrative staff to actively step back and thwart their natural tendencies to “jump in” and solve the problem. Supporting Shahbazim to problem solve and implement solutions was particularly aided when Shahbazim possessed skills to solve problems that were complex or that required strong interpersonal skills. Guides and other leadership staff were pivotal in supporting Shahbazim to develop these skills when shortcomings were evident. Role modeling, practicing, and careful coaching were used to support development of skills.

Usually, when you coach somebody, they do get better. You explain, sometimes you take for granted that people know things, because we know them, and we think it's common sense. But sometimes when you'd explain and encourage and, you know, it gets a lot better, so that's the biggest thing for the Guide is coaching and encouraging and things like that. [GH Guide]

You want them to come up with the solving the problem themselves, but if they don't have problem‐solving skills as far as the steps on how to solve a problem, you can help walk them through that. And they can still solve the problem, but you're guiding them, basically, through the steps. [Social Worker]

Leadership Support of the Model

The orientation of the leadership toward the GH model dramatically affected the problem‐solving processes and resulting solutions. When administrators, directors of nursing, and other leaders understood and were committed to the GH model, reflecting this commitment in both the style and substance of their leadership, problem solving tended to function in a way that reinforced GH principles and practices. In organizations where the model was reinforced during problem solving, a survey citation—or someone bringing forth a perceived threat to elder safety (e.g., an elder with swallowing problems requesting a hamburger)—was met with leaders who became advocates for elder preferences, the homelike environment, and empowered frontline staff practices.

Well, the previous person [overseeing the GH] shot everything down, was against everything, told them she didn't want to hear stuff and on and on and on. And so there was no trust there. And sometime, I don't know, nursing [Director] kind of just took over and said, well, we're going to start doing this, and so they started issuing badges to everybody and said they can just go in when they want to, I don't have to knock anymore. Well, see, that's not right. [GH Guide]

Guide Roles

Green House Guides sometimes hold multiple roles in an organization, influencing the problem‐solving process through multiple paths. For example, some Guides are also the organization's nursing home administrator, and those with this particular dual role described challenges as suppressing the desire to jump in and solve problems versus coaching staff to problem solve (as the model requires); feeling they had inadequate time to coach; and the Shahbazim being nervous to bring a problem forward because they were unsure which “hat” the Guide/administrator would be wearing. Despite these challenges, benefits of this dual role were observed by researchers and appreciated by staff. Problems were often solved quickly when the administrator was also the Guide as he/she then had authority to reallocate resources when needed, was in a position to align GH needs with other organizational processes, and could approve trialing new ideas that deviated from usual policies and procedures. This was seen as supportive to the functioning of the self‐managed work team and reinforcing GH principles and practices

You know, in many ways, I mean, I wouldn't have traded that time for anything, being both the Guide and the administrator … I felt like I had to be the guide to understand what we were doing over there … it certainly made it a little bit easier when I didn't have to work through the chain of command, if you will. I reported right to the chief operating officer … So it made things a little bit easier. [GH and Legacy Administrator]

Nature of Problems

Problems that were not believed to pose existential threats to the organization were more commonly solved in a way that supported GH model principles and practices. For example, problems of daily routine were commonly described as amenable to a slower, collaborative process led by Shahbazim. When solutions could be devised slowly or problems were viewed as within Shahbazim capabilities to safely or reliably solve, the Shahbazim were more commonly supported to problem solve.

Budget

Budget was an important factor when problem solving about issues such as reducing overtime, eliminating staff positions, or purchasing certain foods or supplies. When there were fewer budgetary pressures, Shahbazim decision‐making tended to be supported to a greater extent. When budgetary pressures emerged, in some organizations with strong support of the model, leaders found it useful to involve the Shahbazim in budgetary decisions, providing them with monthly information on the house's food budget or seeking their input on options for resolving budgetary matters.

You know, at first, we had two nurses with each shift and two shahbazes each house to each shift, and we changed that [due to budget]. The shahbaz and nurses sat down with us and had to make a decision about the night shift. The shahbaz said, we would rather have an extra nurse. And, you know, that was hard because that meant that that was affecting their schedules. But they looked at the elders and thought of them first, and that says a lot about the people that they are. [Director of Nursing]

Competition for Workers

Some GH locations are rural, with limited workforce possibilities, while others are in locations with high competition for workers from hospitals and other long‐term care organizations. When there was less competition from other employers, GHs were able to enjoy the opportunity to select a workforce that was optimal to take on the complex duties required for self‐managing a GH home. Although employment competition presents challenges for any long‐term care organization, the significant additional skills required of GH Shahbazim make selective recruitment even more important.

We're in [rural city]. And it's harder to find shahbazim and nurses period, especially those who are really kind of on board, who are ready for this different role. It takes a while to figure this out. I mean, I know that. It doesn't happen overnight. And nurses are, have traditionally been challenged by this because it's so different than what they're used to. So I … yeah. It's a challenge. [GH Guide]

Regulation

When state surveyors were seen by the organization as supportive of culture change, this influenced problem solving. In these situations, administrative staff were more comfortable with Shahbazim solving problems because they felt surveyors would be supportive of efforts to bring this problem solving closer to elders. Some GHs went to great lengths to educate surveyors about the GH model and felt they could successfully advocate for almost any decision the surveyors might question.

The only thing that we had a challenge, and actually it's much better in our region now, but was the dietary and the food. Our, but the way we got around that one is, our dietician came in, who was very well educated on it. And everything they [surveyors] would bring up or say, she had the answer that did meet the criteria. But our surveyor does actually promote it to the point that, even in our state, they'll call me from [the state] regulatory base is and say, “okay, how do you do it … so that we'll learn how to survey it?” [Director of Nursing]

As noted, dominant problem‐solving patterns were observed at each facility during site visits. However, interviews with long‐time employees and the authors' experiences with previous visits to the same facilities (Bowers and Nolet 2014) suggested that the patterns observed within facilities were subject to variation over time, usually in response to changes in the conditions described above. A GH demonstrating a management‐led pattern of problem solving at the time of our site visit may have previously used a coached collaborative pattern. For example, when a new administrator arrived at one home, the GH practice of encouraging Shahbazim to make well‐reasoned staffing recommendations (hire or fire) was supplanted by traditional hierarchical decision making about staffing. This change was emblematic of a generalized shift to a management‐led problem‐solving style that tended to disempower the Shahbazim and restrict their ability to engage in a coached collaborative style of problem solving when addressing other issues.

Central Role of Problem Solving in Sustaining the Model: Examples from the Data

The following examples, drawn from our data, illustrate how problem‐solving processes influenced adherence to the GH model's principles and practices (see Figure 1).

Reinforcement

Shahbazim at this GH were growing frustrated because the supplies they used to complete care tasks, such as washcloths and briefs, often were not available where and when needed. Conflict began to develop among Shahbazim over who was responsible for keeping supplies stocked in the right balance—enough to meet elders' needs, but not so much that the elders' rooms would begin to look institutional. The problem was raised at a house meeting, and the Guide coached the Shahbazim through the process of analyzing the problem and brainstorming a solution. Together, they reached decisions about the amount of supplies that should be kept in the rooms and about an efficient, transparent, and acceptable division of labor among day, night, and weekend workers. Subsequently, supplies were adequately stocked and the conflict quelled.

The “coached collaborative” approach of solving this supplies problem, which emphasized Shahbazim cooperation in solving the problem and Guide coaching, was seen as reinforcing the self‐management principle of GH, and the solution supported the non‐institutional character of elder's rooms. Maintaining the GH model was assisted by the nature of the problem (being amenable to the slower, collaborative process), Shahbazim opportunities and capacity to problem solve (through regular house meetings and the Guide taking the time to coach Shahbazim in problem solving), and leadership support of the model (the Guide understanding the importance of empowering Shahbazim).

Erosion

A resident at this GH left the front porch of the house, where elders enjoyed sitting in nice weather, and made his way to an area toward the road, where he was soon recognized and returned to the house. The administrator reported the elopement to state regulators and the home was assessed a fine and other penalties. The state required the organization to provide a plan of correction within 24 hours. The administrator and the director of nursing devised a plan that mandated all residents should sign out when leaving the house and remain within view of the Shahbazim at all times. Elders were forbidden from going outside (including the porch) without supervision. Additionally, alarms were installed on the outer doors of the house. One person described the situation as being “on lockdown.” At one point, the Shahbazim suggested alternative approaches to preventing elopement, but these solutions were dismissed in favor of faster solutions devised by administration. This solution to the elopement problem was widely viewed as burdensome to the Shahbazim, and as a diminution of rights and freedom for the elders. One participant summarized the changes as “two, if not ten, steps backward to … a more traditional nursing home setting.”

This plan of correction was seen as institutionalizing elders' lives in a way that was contrary to the spirit of GH. The hierarchical way in which the solution was devised and imposed also damaged the autonomy of the self‐managed work team. In this instance, problem solving, and sustainability of the GH model, was influenced by the nature of the problem (requiring a quick solution); leadership (showing limited support for the model); and the regulatory environment (reinforcing traditional nursing home management techniques).

Reinforcement or Erosion?

One GH organization started out preparing the food in the GH units, as advocated by the GH model. However, this effort proved to be a source of continual problems: budget overruns due to inefficiency and waste were common; Shahbazim possessed variable cooking skills and elders complained about the food; and conflicts developed between shifts over who was responsible for food preparation. A tipping point was reached when elders' acuity began to increase, limiting the elders' capacity to participate in cooking. The Shahbazim, Guide, and department leaders attempted several solutions, including providing a list of “substitutes” for ingredients that often ran out, and educating young Shahbazim about the types of food elders were likely to enjoy. Several staff then heard that “some” other GH organizations had meals made in the legacy home and brought over to the GH for heating and serving. They decided to adopt this practice. Breakfast would still be “made to order” and some food items (e.g., sandwich fixings and favorite snacks) would be kept on hand at each house.

This change in practice caused controversy. Some staff perceived the change as a violation of both GH practices and principles. Elders would not see or hear the work of food preparation and would not have the opportunity to contribute to meal planning or preparation. Others felt that the elders were not interested in cooking, and that the smell of food heating, along with providing elders with opportunities to make some food choices, was enough to maintain the GH principle embedded in the practice. By shifting food preparation to the legacy home, the latter group argued, the Shahbazim became more available to the elders in ways that the elders needed and valued.

In this example, problem solving fluxed over months from “management led” to “coached collaborative,” and GH model sustainability was influenced by the following conditions: being non‐pressing in nature, budgetary implications (food waste), leadership support for the model (different interpretations of the value of preparing food in the home), opportunities and capacity of Shahbazim to problem solve (extent Shahbazim were empowered and coached), and Guide roles (Guide as administrator having the big picture and involving key players).

Discussion

This study focused on sustainability of one culture change model (The GH) and provides insights into some of the factors that influence sustainability, particularly strategies used successfully by GH organizations to sustain the model despite ongoing challenges. The study highlights the importance of attending to sustainability even in organizations that have “successfully” implemented the model.

The GH model differs from many of the current culture change initiatives in that instructions and support for implementing the model are highly prescriptive, providing detailed guidelines as well as significant training and support during preparation and adoption. For example, GH adopters must undergo a 1‐week intensive “Core Team Education” for at least all Shahbazim, Clinical Support Team members (social workers, nurses, therapists), and Guide in an new GH organization. Every GH organization must also appoint a GH educator tasked with providing ongoing GH training and support to the organization. Some optional, but encouraged, programs include “Coaching Supervision” (for Guides), “Coaching for Partnership” (for all Clinical Support Team members), and the Peer Network for continuing education and support. These programs focus on learning new team roles, and equipping staff to build a well‐functioning GH teams. While available data did not allow for a clear correlation, the researchers believe a link between sustainability and GH education may exist and merits further examination. For example, our finding that coaching Shahbazim to problem solve reinforces the GH model could mean the GH education about coaching is a factor in sustaining GH principles and practices.

While staff from GH homes invest a substantial amount of time learning about their new roles during implementation, there is considerably less guidance related to sustaining the model over time, either in general or in response to changing circumstances. Findings from this study highlight the significance of small, daily problems, how they are addressed, how upper management is involved, and the importance of development of Shahbzim skills in sustaining the model. The significance of how “small” problems are addressed is an important finding. Problems such as stocking linen would likely not be considered important enough to merit intervention from management or significant time investment. The findings suggest that these mundane problems are highly significant in both reinforcing the commitment to collaborative problem solving, and teaching direct care staff the skills they lack. This is at odds with how many culture change adopters have approached “empowering direct care staff,” often interpreted as “hands off” for upper management, leaving direct care staff to fend for themselves without the skills necessary.

The importance of management response to problems that are considered urgent, have budget implications, or are highly significant for the organization (such as regulatory or survey issues) also provides an important lesson. Management claiming total authority for “important” or urgent issues was perceived by direct care staff as both undermining the principle of collaborative decision making and, in some cases, as directly undermining a core practice. Whether a problem related to stocking linen or a serious survey deficiency, including staff at all levels in determining the response reinforced the commitment to collaborative decision making.

Direct care staff in this model, as well as in other culture change models, are afforded privileges, autonomy and self‐management, that require a different skill set than traditional CNA work. Thus, in sustaining the GH model, and culture change more broadly, questions about the long‐term care workforce and its preparation arise. Our study highlights the importance of fostering both collaborative problem‐solving skills in self‐managed work teams and coaching skills for leaders who support these teams. However, these practices raise questions about how to identify, recruit, compensate, and support staff who can successfully participate in self‐managed work teams and about appropriate levels of training and support. Bishop (2014) has also highlighted the need for more sophisticated human resource practices in order for self‐managed work teams in nursing home culture change to be successful.

Wages also remain low for CNAs with increasing job demands. The mean hourly wage for CNAs in nursing homes in the United States was recently reported at $12.01 per hour (U.S. Bureau of Labor Statistics 2013). Most GHs have recognized the need for salary differential and instituted a slightly higher hourly wage for Shahbazim. Bishop (2014) notes this higher wage has been a critical factor for the success of high performance work teams in other industries, and suggests that successful culture change or quality of life initiatives can lead to increased profits through attracting private‐pay consumers or Medicaid pay for performance incentives.

This study is limited in its ability to generalize findings to other culture change models outside of the GH model. Exploration of how this model of problem solving and sustainability applies to other approaches to culture change would advance our ability to develop guidance for culture change adopters more generally. Future work can also be undertaken to understand what makes culture change organizations more vulnerable to sustainability challenges, the elements of culture change that are most vulnerable to erosion, and specific actions key organizational leaders (e.g., administrators and department leaders) can take to influence sustainability.

Supporting information

Appendix SA1: Author Matrix.