This article uses a retrospective approach to critique the research base underlying the nursing home culture-change movement—an effort to radically transform the nation's nursing homes by delivering resident-directed care and empowering staff. The article traces the development of the movement from its inception 10 years ago to 2005, when the Centers for Medicare and Medicaid Services implemented its own initiative to support the movement, thus giving it new momentum, to the present day. This historical overview provides context for a proposed research agenda aimed at strengthening the movement's empirical base, thereby facilitating culture-change interventions as well as helping the movement navigate the next step in its evolution.
Once considered by some to be long overdue, the nursing home culture-change movement is now making up for lost time. An effort to radically transform the nation's nursing homes (NHs) by delivering resident-directed care and empowering staff, the movement has evolved rapidly in the past 3 years, growing from a small grassroots movement 10 years ago to a rapidly expanding federally advocated initiative today. But was this movement mature enough to experience momentous growth without also experiencing growing pains? In the first half of this Forum article we probe this question by examining the movement's recent history. An account of seminal meetings, conferences, publications, and other developments that document that history brings to light the roots of a problem—in short, a weak empirical base—that, if left unaddressed, could undermine the movement's creditability and limit its effectiveness. In the second half, we propose a research agenda aimed at strengthening the movement's empirical base, thereby facilitating culture-change interventions as well as helping the movement navigate the next step in its evolution.
With this article, we hope to stimulate discussion, debate, and research about a social-change movement that in short order has sparked the imaginations and passions of NH providers across the country while skeptics stand to the side and, often off the record, question the movement's effectiveness. We believe that the culture-change movement has matured enough not only to withstand such a critical peer review but to benefit from it. Thus the ideas presented in this article are offered as a thoughtful starting place for a conversation whose time has come.
The Early Years
The NH culture-change movement was born, by most accounts, in 1997, following the first meeting of the NH Pioneers (now known as the Pioneer Network). In that meeting's resultant report, the term culture change was coined. What exactly is culture change in the NH environment? The concept has defied easy definition (even to today), in part because the movement encompasses diverse and sometimes contradictory components. Consequently, culture change is more often described in the literature than defined in it. In general, culture change is a process, and as such, the term connotes a transformation of NHs that goes beyond superficial changes to an inevitable reexamination of attitudes and behavior, and a slow and comprehensive set of fundamental reforms. Culture-change proponents aim to create caring communities where both empowered frontline staff and residents can flourish, and where residents experience enhanced quality of life. Within this framework, a culture change can vary in scope and scale, taking numerous forms, including, but not limited to, cross-training workers, organizing residential areas into small “households,” creating environments that appear more homelike, enhancing the dining experience, soliciting resident opinions on daily routines, and writing up care plans in the voice of the resident.
In its early years, the NH culture-change movement spread slowly. Pioneer Networks emerged within some states, and the state affiliates of the two major NH trade associations, the American Association of Homes and Services for Aging and the American Health Care Association, began conducting programs on culture change for their members. Additionally, a small chain of New England NHs, Apple Health Care, committed itself to culture change, and the largest for-profit NH chain, Beverly Corporation, initiated architectural and system changes in selected facilities.
There also emerged two distinct models of NH care that, although conceived prior to 1997, eventually became associated with the culture-change movement. The Eden Alternative, articulated in 1994 (Thomas, 1994), advocated, among other changes, that residents interact with children, pets, and plants to help combat feelings of boredom, loneliness, and helplessness. The Wellspring model, also initiated in 1994, focused on clinical quality improvement and environmental culture change within a consortium of NHs that shared ideas and some resources.
Between 1997 and 2005, a year whose significance is discussed later, few culture-change initiatives were rigorously evaluated. One reason is that some culture-change leaders were not measurement oriented, or were even opposed to measurement because the very act of measuring quality of life could seem to be a dehumanizing activity. Related to this, some culture changes possessed so much face validity that evaluating them seemed unnecessary (if the cost is reasonable, for instance, can there be any objection to replacing institutional white towels with richly colored ones?). In addition, some culture-change projects had a comprehensive scope and a long gestational process that rendered the attribution of effects difficult. Similarly, a lack of definition for culture change clouded the prospect of measuring cause and effect.
Given this, it is not surprising that evaluations were most likely when the culture change was clearly defined. Thus, in 2002, Stone and colleagues published their well-regarded evaluation of the Wellspring model. This study found improved quality outcomes as measured by observations and interview results, better staff retention rates, and reduced turnover rates. Eden Alternative facilities were also the subject of a few small-scale studies, which found mixed results (Bergman-Evans, 2004; Coleman et al., 2002; Rosher & Robinson, 2005). The most ambitious collection of articles about culture change at this time appeared in a double issue of Social Work in Long-Term Care, and this collection was later released as a book in 2004 (Weiner & Ronch, 2004). This anthology combined opinion, argument, description, case studies, and some quantitative and qualitative results of tests of various aspects of culture change.
During this same period, broadly based efforts to improve NH life were sometimes regarded with skepticism as untested. An Institute of Medicine report about quality in long-term care, released in 2001, recommended increased staffing and regulatory enforcement, but it endorsed only studies of consumer autonomy (Wunderlich & Kohler, 2001). This somewhat chilly stance toward direct efforts to improve quality of life for NH residents probably also helps explain the dearth of culture-change studies at this time.
In any event, 8 years after its birth, the movement was struggling to gain a solid foothold in the NH industry, according to a report for the National Commission for Quality Long-Term Care. The report's authors noted that culture-change initiatives were “swimming against the tide of regulation, limited resources, and established practices” (Capitman, Leutz, Bishop, & Casler, 2005, p. 33). They also concluded that, based on their examination of the available literature, “[culture change] models have not been empirically evaluated yet, including focus on impacts on quality of life/quality of care, costs, and variations in results due to case mixes.”
That report was published in March of 2005. Its rather discouraging but fair assessment of the NH culture-change movement up until that time, however, seemed to have no bearing on an event that occurred soon afterward. In June 2005, the Pioneer Network and Quality Partners of Rhode Island hosted the St. Louis Accord, a “landmark meeting” that brought together 400 long-term-care leaders to work toward hastening culture change in the nation's NHs (Quality Partners of Rhode Island, 2005). In attendance and in support were representatives of the Centers for Medicare and Medicaid Services (CMS) and State Survey and Certification Agencies (Quality Partners of Rhode Island)—regulatory agencies that heretofore had been considered potential impeders of the movement.
The CMS further endorsed culture change in August 2005, when it directed state Quality Improvement Organizations (QIOs), in their 8th Scope of Work, to work with NHs to “improve organizational culture” (CMS, 2005). With this official blessing, and with 50 government-contracted agencies to support it, a movement that had seemed to be stalled suddenly gained momentum.
Recent Past to Present
In the nearly 3 years since, the movement has enjoyed increased support from funders (the Commonwealth Fund in particular) and increased attention from the media: Consider, for example, that 18 of the 19 news articles and television reports listed under “Movement in the Media” on the Pioneer Network's Web site (Pioneer Network, 2007) were published or aired after April 2005—the lone exception appeared in 2002. The CMS also is now more involved in the movement. In April 2006, it released a 79-item questionnaire—the Artifacts of Culture Change Tool—as a self-study tool for facilities to assess their own progress toward culture change (CMS, 2006). This protocol is now being used in the CMS's Staff Time and Resource Intensity Verification Project. More recently, in September of 2006 the CMS helped launch “Advancing Excellence in America's Nursing Homes,” a 2-year quality-improvement campaign whose eight goals include creating a culture of person-directed care (Nursing Home Quality Campaign, 2007).
This expansion constitutes progress, but it also exposes a problem: The research base underlying the movement has not grown at a similar rate. To be sure, a handful or so of culture-change efforts have been well described in recent articles (Cohen-Mansfield & Bester, 2006; Nijs et al., 2006; Rabig, Thomas, Kane, Cutler, & McAlilly, 2006), particularly in nursing journals where transformed bathing techniques are described (e.g., Rader et al., 2006). But for these exceptions, today we know little more about the efficacy of culture-change strategies than we knew 3 years ago—and we knew little then. In the most recent report to the National Commission for Quality Long Term Care (April, 2007), Weiner, Freiman, and Brown echo the conclusion by Capitman and colleagues in 2005: “Although there has been a lot of media coverage, these [culture change] innovations have not been rigorously evaluated or replicated under varying leadership, ownership, and case mix circumstances” (p. 31).
The movement's message, however, is being widely disseminated, though largely outside the purview of peer-reviewed journals, through books, conference reports, and the Internet. Culture-change advocacy groups such as the Pioneer Network play a role, as do professional culture-change agents, such as ActionPact. The CMS and the QIOs also are disseminating the movement's message.
Perhaps because this message has escaped critical peer review, culture-change interventions are often advocated with little mention of their mostly untested premises. Instead, case studies and anecdotal reports are often presented as evidence of success, typically with no mention of the caution needed when one is attempting to generalize from this information (see, e.g., “Featured Stories” on the ActionPact Web site, www.culturechangenow.com). Unfortunately, this failure to rigorously inform providers could have serious consequences. NHs that unwittingly implement innovations whose outcomes are largely unproven may wind up wasting time and money on strategies that fail to produce positive results. In the worst cases, vulnerable residents in these facilities may suffer as a result.
As the movement has gained momentum, its leaders have produced increasingly explicit implementation materials. Once content to exhort each NH to “choose for itself what works best in its own unique environment (Norton, 2006),” some culture-change reformers are now issuing what amount to instructions for achieving culture change. Among these materials are the previously referenced Artifacts of Culture Change Tool (CMS, 2006); Household Matters, a toolkit by Meadowlark Hills and ActionPact that includes “hundreds of policies on putting the person first (Shields, 2006);” and a stage model of culture change toward households developed by ActionPact. Additionally, the organizational improvement goals of the Advancing Excellence campaign, which include measuring resident satisfaction, improving staff retention, and implementing consistent assignment, represent a concerted effort to guide culture change in NHs.
This step toward a more targeted message is understandable: For culture change to be readily accessible to the nation's 17,000 NHs, detailed guidelines for achieving it are needed. At the same time, this trend could undermine the movement's credibility, for it may seem to confer “best practice” status on interventions that have yet to pass muster in the research literature. Consider, for example, the Advancing Excellence campaign's goal to encourage NHs to adopt consistent assignment—the practice of assigning nurse aides to the same residents on a daily or nearly daily basis. With the CMS among its founders, the campaign coalition advocates with some authority, and NHs are responding. As of June 1, 2007, there were 5,246 nursing homes—33% of the nation's total—that had registered as “official participating providers” in the 2-year campaign. Of these, 31% have committed to implementing consistent assignment (Nursing Home Quality Campaign, 2007).
Nevertheless, a review of the literature shows that relatively few studies have examined staff assignment, and these are of mixed quality, yielding mixed results (Burgio, Fisher, Fairchild, Scilley, & Hardin, 2004; Campbell, 1985; Patchner, 1989; Patchner and Patchner, 1993; Zimmerman et al., 2005). Indeed, the most recently reported studies conclude that the question of preferred staff assignment—rotating versus consistent—merits further attention (Burgio et al.; Zimmerman et al.). “Surprisingly,” write Burgio and colleagues, “[rotating assignment] and [permanent or consistent assignment] resulted in few differences in quality-of-care outcome indicators, precluding a clear conclusion about which type of staffing assignment should be recommended to nursing home management” (p. 376).
We believe that the culture-change movement, particularly participating NHs, will benefit from a forthright acknowledgment of what to us seems apparent: The movement is being built largely around innovations, a term that implies by definition that the changes are mostly untested and their outcomes somewhat uncertain. There is no shame in this admission—nor should it be construed as a reason to abandon the movement. On the contrary, this is common course, or at least not an uncommon course, for the evolution and maturation of a popular social-change movement. We contend that the more compelling fact is that, notwithstanding a relatively weak research base, the movement has grown rapidly in the past 3 years and appears to be expanding. Such a groundswell of enthusiasm demands respect as a prerequisite for innovation. Moreover, many culture-change innovations are promising and worth trying, especially when compared with the status quo, which has proved disappointing—or worse—to many.
Far from an admission of weakness, acknowledging the untested nature of many culture changes recognizes that the movement is now mature enough to withstand a more critical analysis. Indeed, we view this acknowledgment as a step that advances the movement, for NHs that embark on culture change are likely to achieve better results if they understand that in essence they are undertaking an experiment. Consider again the example of adopting consistent assignment. It seems likely that NHs that implement consistent assignment with the understanding that, though promising, the practice still merits study will stay alert to potential implementation pitfalls, which they can then work to resolve. This understanding thus empowers providers, making success more likely.
Moving Forward: A Research Agenda
Although NHs should bear some responsibility for evaluating the innovations they implement, researchers should work toward lightening this burden by building an empirical base that can inform and guide the culture-change movement in the years ahead. In this section, we present a research agenda aimed at strengthening the movement's scientific foundation. This agenda has two goals. First, it serves the general goal of shedding light on important but unanswered questions about culture-change interventions, especially questions about staffing requirements and the residents most likely to benefit from these interventions. Second, it will generate evidence-based protocols that NHs can use to strengthen their culture-change practices. Quality-improvement efforts of all types, including those that fall within the culture-change movement, require that intervention processes and their related outcomes be defined and measured and that factors contributing to the staff's ability to implement the processes be identified. As we discuss in the paragraphs that follow, much of this work remains to be done for culture-change innovations. Until the research catches up with culture-change practice, NHs are on their own to meet these quality-improvement requirements. Over time, however, the proposed research agenda will generate evidence-based protocols that can strengthen the culture-change infrastructure.
This section—our proposed research agenda—is organized around five research questions pertaining to culture-change interventions: What are the potential outcomes? What care processes are related to these outcomes? What factors limit staff ability to implement the intervention? What are the staffing costs of implementing the intervention? Do all residents, including those with cognitive impairments, benefit from the intervention?
These questions are pertinent for three reasons. First, we presently lack data relevant to all five questions for most culture-change innovations. Second, such data could be quickly generated in preliminary intervention studies. Third, the resultant data could be used to identify the more promising interventions, which could then be evaluated in controlled trials, as well as to generate evidence-based protocols that could strengthen culture-change practices in NHs. We discuss each research question in turn.
Care processes are the day-to-day focus of quality-improvement efforts, but these should be related to and over time should influence selected outcomes. In culture-change interventions, expected outcomes can vary widely—from staff turnover rates to depression rates among residents. In this article, we focus on just one—resident satisfaction or, more aptly, resident perception of quality of life—because this outcome is of central importance in the culture-change movement.
As part of the culture-change movement, NHs are increasingly being urged to assess resident satisfaction. The QIO's 8th Scope of Work, for instance, instructs these agencies to work with NHs to assess resident satisfaction. The Advancing Excellence campaign also includes resident satisfaction assessment as one of its goals. The assessment tools recommended by both the campaign and the QIOs, however, are broad-based survey instruments that work as global needs assessments but are too blunt to be useful for quality-improvement purposes. What are needed instead are outcome measures that are specific to the intervention being tested. Thus, for example, outcome measures for examining the success of an intervention designed to improve mealtime enjoyment might include residents' satisfaction with the food, dining ambience, and staff assistance.
Complicating matters is that measuring satisfaction and the preferences of NH residents, most of whom are cognitively impaired, is challenging, and, if done thoughtlessly, can lead to erroneous conclusions (Simmons et al., 1997; Simmons & Schnelle, 2001). An acquiescence bias—as documented by evidence that residents often express satisfaction with low levels of care—can distort preference reports (Levy-Storms, Schnelle, & Simmons, 2002; Simmons & Schnelle, 1999). So too can a resident's history of care (Simmons & Levy-Storms, 2006). The use of family and staff proxies for cognitively impaired residents introduces a serious error, for there is ample evidence that these residents can reliably indicate their preferences and report on their experiences (R. A. Kane et al., 2003; R. L. Kane et al., 2005).
A recent study by Simmons and Ouslander (2005) shows promise for developing satisfaction measures that are useful for evaluating culture-change innovations. The researchers found that discrepancy questions, which reflect the difference between perceived and preferred care (e.g., “What time did you get out of bed today?” vs. “What time do you like to get out of bed?”) were sensitive to improvements in care. By contrast, resident responses to commonly used direct-satisfaction questions (e.g., “Are you satisfied with when you get up in the morning?”) did not change even when the services in question were enhanced. More work is needed, however, to further develop reliable and accurate measures of resident satisfaction and preferences.
Another research challenge is to identify the care processes related to desirable culture-change outcomes and assess these in a framework that allows for ongoing staff training, expansion of the approach, and systematic improvement. Ideally, care processes should be measured routinely in NHs and staff should help analyze this data so as to identify ways to improve process implementation. For example, a primary culture-change outcome is to encourage residents to make choices about daily activities. Existing information systems for monitoring daily care processes, however, primarily measure the frequency of care delivery in clinical domains (e.g., repositioning) as opposed to the more subtle aspects of care that may be related to choice and quality of life, such as how much time the staff spends encouraging residents to make choices and how well that encouragement is offered. Does the staff member maintain eye contact? Smile and appear interested in the resident? Or is he or she interested only in the resident's quick consent to care? Consider a typical (in our experience) staff communication with a resident: “Hi, Ms. Smith, it's time to get up now. Is that OK?” Some would argue that this question encourages resident choice, whereas others would contend that, to the contrary, it merely prompts quick consent. To improve staff behavior in this situation, specific information about how the staff encourages residents to make choices is needed.
Staff Implementation Challenges
Related to the aforementioned information, staff opinions about barriers to honoring resident choices should be solicited. Let us return to our just-mentioned example: The staff member reported that the resident was depressed and would never get up if offered an open-ended choice (e.g., “Hi, Ms. Smith. When would you like to get up today?”). In this and similar cases, true culture change is possible only if caregiver opinions also are considered so that informed decisions can be made about whether to respect resident choices with potentially adverse health consequences and so that staff training can occur if alternative care patterns are desired.
Researchers must also resolve challenges in measuring how well the staff interacts with residents if the quality-of-life goals so fundamental to the culture-change movement are to be realized. Direct observations of staff–resident interactions may be necessary to develop a quality-of-life information base that is useful for improving these interactions. Researchers are best qualified to develop observation-based information systems, because the methods for collecting accurate data through observation protocols, though well described in the literature, are poorly understood by health care personnel in all settings.
Researchers and staff advocates have argued that, as a result of the twin problems of understaffing and staff turnover in NHs, most interventions can be sustained only if targeted to the residents who need them most. This concern raises a question for future research: Must culture-change interventions be similarly targeted or are these interventions—all or some of them—either not labor intensive or successful in reorganizing and retaining staff such that staffing levels and turnover are not limiting factors?
Related to this is the need to examine staff concerns about workload. Culture-change proponents contend that their reforms redistribute and redefine staff workloads, not increase them. If so, then it is important to prove the point with defensible measures of what all staff members do for residents on a daily basis and how residents respond. Such proof is needed to refute evidence that staffing is inadequate in most NHs (Harrington et al., 2000) and to counter concerns that “[culture change] approaches may end up requiring more staff and higher costs” (Weiner et al., 2007, p. 32).
Many NH providers, researchers, and resident advocates question whether cognitively impaired residents can make free choices about their daily activities, given their high dependence on staff for assistance. Monitoring which residents make choices and how they are encouraged to do so would provide important information about whether and how cognitive impairment limits the efficacy of culture-change interventions.
A Starting Point
Among the first interventions to consider for study can be those that increase resident choice over daily activities. Resident choice is a core value within the culture-change movement. Moreover, federal regulations require that NHs offer residents choice. Specifically, survey regulations advise surveyors to observe whether residents are provided choice; deficiency tags (F tags) have been designated if surveyors find unspecified evidence that choice is not offered. This quality-of-life domain possesses face validity and is widely accepted as important; surprisingly, however, we lack information about how to offer meaningful choices to all NH residents and information about the costs and outcomes of doing so.
If data collected during preliminary research succeeds in defining specific care processes that encourage choice, and if such care appears feasible to implement, then a follow-up controlled trial could document its importance for such outcomes as depression, resident sense of control, and perception of life quality. Based on our experience, it is plausible to hypothesize that all of these outcomes would show improvement. Such findings would provide the credibility needed to advance the culture-change movement to the next level.
Eden Alternative founder Bill Thomas once identified “storytelling ability” as the number one skill needed to drive change in organizations serving elders. “The stories I'm talking about,” he told Nursing Homes Magazine (Keane, 2004), “encapsulate lessons into digestible, memorable chunks. … The history of the Eden Alternative is paved with such anecdotes.” On the strength of these and similar anecdotal reports and case studies as well as support from the CMS, the nursing home culture-change movement is expanding rapidly. Consider that when the Pioneer Network celebrated the movement's 10th anniversary at its annual conference in August 2007, up to 1,000 professionals registered—more than twice the number who attended just 2 years earlier.
Nevertheless, a review of the scientific literature shows that the culture-change movement is spreading in advance of a solid research base to support its quality-of-life improvement claims. Left unaddressed, this gap could erode the movement's credibility and threaten its successes. We contend in this article that the movement is now mature enough that its methods can withstand rigorous research analysis. Our proposed research agenda is designed to strengthen the movement's empirical base and thereby facilitate culture change and further advance the movement. We propose research questions that will bring to light the costs and benefits of the myriad culture-change innovations currently underway in NHs across the country, as well as the relative advantages and disadvantages of these new interventions. The knowledge gained from this research can help guide and inform the movement as it continues to evolve.
Scripps Gerontology Center, Miami University, Oxford, OH.
Center for Quality Aging, Vanderbilt University, Nashville, TN.
Decision Editor: William J. McAuley, PhD