Abstract
The purpose of this study was to identify structural, market, and administrator factors of nursing homes that are related to the implementation of person-centered care. Administrators of Medicare/Medicaid-certified nursing homes in the Deep South were invited to complete a standardized survey about their facility and their perceptions and attitudes regarding person-centered care practices (PCCPs). Nursing home structural and market factors were obtained from public websites, and these data were matched with administrator data. Consistent with the resource-based theory of competitive advantage, nursing homes with greater resources and more competition were more likely to implement PCCPs. Implementation of person-centered care was also higher in nursing homes with administrators who perceived culture change implementation to be feasible in their facilities. Given that there is a link between resource availability and adoption of person-centered care, future research should investigate the cost of such innovations.
Introduction
The quality of nursing home care has been a major challenge in the past 30 years, and much remains to be done to achieve adequate levels of quality of care despite the efforts of the federal and state regulatory processes and quality improvement programs targeted at clinical quality indicators (Colon-Emeric et al., 2006; Comondore et al., 2009; Quadagno & Stahl, 2003; Rask et al., 2007; Scott-Cawiezell & Vogelsmeier, 2006). Similarly, consumers, both current and potential, typically respond quite negatively when asked their opinion of nursing homes, a reaction that is likely related to the fear of loss of independence as well as disdain for living in an institutional environment (Boisaubin, Chu, & Catalano, 2007; Kaiser Family Foundation, 2007; R. L. Kane & Kane, 2001; R. A. Kane, Lum, Cutler, Degenholtz, & Yu, 2007). These fears are somewhat realistic given that the nursing home environment and structure were initially modeled to resemble an acute care hospital setting (White-Chu, Graves, Godfrey, Bonner, & Sloane, 2009). Given the concerns of regulators and consumers, it is clear that a change in organizational culture is needed (Kaiser Family Foundation, 2007).
In an effort to address these problems, the culture change movement was launched, so named because it was conceived to literally change the entire organizational culture of the traditional nursing home. Culture change, also known as resident-centered care and person-centered care, is a movement directed toward deinstitutionalizing facilities by flattening the internal structural hierarchy in the nursing home, allowing more autonomy for residents and staff, enhancing the environment to make it more homelike, and improving the quality of care of residents (Pioneer Network, n.d.). The process of culture change implementation varies across long-term care settings, ranging from renovating the existing physical structure to improving residents’ quality of life (Hung, Chaudhury, & Rust, 2015) to changing daily care practices to promote resident autonomy (Simmons, Durkin, Rahman, Schnelle, & Beuscher, 2014).
Although culture change has received widespread publicity and strong support from the federal Centers for Medicare and Medicaid Services (Bowman, 2008), the majority of nursing homes have not yet committed to adopting this innovation (Doty, Koren, & Sturla, 2008). Moreover, according to findings from a national study of culture change, nursing homes in the South are less likely to implement person-centered care practices (PCCPs) compared with nursing homes in the western and northeastern regions of the United States (Doty et al., 2008).
Given emerging evidence that culture change adoption has positive effects on quality of care and operational success (Grabowski, O’Malley, et al., 2014; R. A. Kane et al., 2007), it is important to understand what factors may influence the implementation of PCCPs, especially within the Deep South. The purpose of this study was to identify structural, market, and administrator factors of nursing homes that are related to the implementation of person-centered care. Structural, market, and administrator factors were identified as possible moderators because a significant body of evidence has been built in the past 15 years finding that such factors are associated with the adoption of new innovations by long-term care facilities as well as higher levels of quality of care in general. Given the innovative nature of PCCPs, and the emerging evidence that PCCPs are associated with improved quality of care, we posited that these factors might also be moderators for PCCP adoption. In the next paragraphs, we review the evidence for each type of factor.
Structural factors most often examined in relation to innovative changes and care quality include ownership status, chain affiliation, facility size, and Medicaid and Medicare census. Such factors might be important moderators for PCCP adoption because they tend to covary with facility priorities, available resources, and ability to nimbly implement change—three variables likely affecting PCCP adoption decisions. For example, nonprofit facilities have a value of reinvesting profits to support facility missions such as caring for vulnerable individuals. Chain facilities are more likely to belong to for-profit corporations. On the other hand, chain facilities are also more likely to have access to greater centralized resources, and those centralized resources are likely to be able to support large-scale systemic changes. Smaller facilities are less likely to have significant resources but, on the other hand, are more likely to be able to nimbly implement changes than larger facilities that are not part of chains. Finally, facilities with a larger proportion of Medicaid to Medicare census are likely to have less profit and thus fewer resources than those with the opposite proportion.
Regarding profit status, there is evidence that for-profit nursing homes tend to provide lower quality of care (Hillmer, Wodchis, Gill, Anderson, & Rochon, 2005), provide fewer staff hours per resident (Aaronson, Zinn, & Rosko, 1994; Harrington, Zimmerman, Karon, Robinson, & Beutel, 2000; McGregor et al., 2005; Schnelle et al., 2004), and have a higher number of deficiencies compared with nonprofit nursing homes (Harrington, Woolhandler, Mullan, Carrillo, & Himmelstein, 2001). Although early adoption of innovations, such as computerized medical records, is higher in chain affiliated compared with independent medical facilities, chain affiliation in nursing homes has been associated with lower overall quality of care (Zimmerman, Gruber-Baldini, Hebel, Sloane, & Magaziner, 2002) and a higher number of deficiencies (O’Neill, Harrington, Kitchener, & Saliba, 2003). Facility size may play a role in innovation adoption, but mixed results are found in the literature. For-profit homes tend to be larger in size and are associated with a lower quality of care (O’Neill et al., 2003), but the ownership status may contribute more to this finding rather than the size of the facility. A smaller facility may provide an environment that is more conducive to greater quality of care (Keays, 2007), though larger facilities may have greater access to resources and thus an increased likelihood of innovation (Banaszak-Holl, Zinn, & Mor, 1996). There is also evidence that the proportion of private pay versus Medicaid residents is an important factor in quality. Higher private pay and a lower Medicaid census are associated with early adoption of innovations (Castle, 2001), whereas higher Medicaid census is associated with lower quality of care in nursing homes (Gertler, 1989, 1992; Grabowski, Elliot, Leitzell, Cohen, & Zimmerman, 2014). In addition, higher Medicare census is associated with a decreased likelihood of implementing innovations such as an Alzheimer’s unit or subacute care unit (Banaszak-Holl et al., 1996).
There is a strong theoretical basis that is concordant with the prediction that market factors influence innovation. The resource-based theory of competitive advantage is a widely used framework in strategic management that posits that organizations that possess resources and capabilities are in a position to improve and offer better services to consumers (Barney & Clark, 2007; Lavie, 2006), which in turn makes those organizations more attractive to consumers. Castle (2001) found that nursing homes located in areas with higher median incomes and a greater number of nursing home beds were more likely to be early innovation adopters, which may be a result of greater resource availability. Also consistent with the resource-based theory regarding competition, Banaszak-Holl et al. (1996) found that innovation in nursing homes was less likely in areas that were highly concentrated with nursing homes and, thus, less conducive to high competition. They hypothesized that organizations in concentrated markets would be less likely to fund innovation because the market was not competitive.
Research on nursing home administrator (NHA) characteristics indicates that factors such as leadership style (Castle & Decker, 2011; Scott, Mannion, Davies, & Marshall, 2003), membership to a professional association (Castle & Banaszak-Holl, 1997; Castle & Fogel, 2002), education (Castle & Banaszak-Holl, 1997; Keays, 2007), and job tenure (Castle & Banaszak-Holl, 1997; Decker & Castle, 2011) affect overall quality of care and quality indicators in long-term care. However, there is little research on the potential influence that NHAs’ beliefs and attitudes related to culture change have on the adoption of PCCPs in long-term care. This is an important gap because administrators are a key component of the leadership team. They oversee daily operations and serve as the liaison for direct care staff, residents, family members, and upper level employees. Thus, they are ideal motivators for change (Allen, 2008). Two recent qualitative studies emphasize the value of NHAs’ knowledge, motivation, commitment, and vision of culture change practices (Corazzini et al., 2015; Shield, Looze, Tyler, Lepore, & Miller, 2014).
The study aimed to identify structural, market, and administrator factors that are related to implementation of PCCPs in nursing homes throughout Alabama. This study is the first examination of culture change that is focused exclusively in the Deep South. The resource-based theory of competitive advantage served as our foundation for identifying structural and market-level characteristics that influence nursing home resources and competition (Barney & Clark, 2007). We hypothesized that nursing homes implementing fewer PCCPs would be for-profit, chain affiliated, smaller in size, and have a higher census of Medicaid/Medicare (i.e., lower proportion of private pay). In addition, we expected that market factors associated with fewer PCCPs would include rurality, lower median county income, and less market competition. Finally, we hypothesized that negative administrator attitudes toward the principles and feasibility of culture change, as well as poor knowledge about PCCPs, would be associated with fewer PCCPs.
Design and Method
Participants
All administrators from all Alabama skilled nursing facilities that were accessible to the general older adult population (i.e., excluding facilities exclusively for disabled children or veterans) were invited to participate. A comprehensive list of nursing homes was obtained from the Nursing Home Compare website (Centers for Medicare and Medicaid Services [CMS] & U.S. Department of Health and Human Services [USDHHS], n.d.) as well as the Alabama Nursing Home Association website (Alabama Nursing Home Association [ANHA], n.d.). Administrators were targeted as study informants because of their knowledge of the day-to-day operation of the facility and their influence over facility strategy, policies, and practices. A total of 234 nursing homes were contacted; 75 administrators participated, of whom two were responsible for two nursing homes each. As a result, data were collected representing 77 nursing homes, rendering a response rate of 33%.
Data Sources
An overview of the data used for this study and the associated data sources is presented in Table 1. Participants completed a modified version of the Commonwealth Fund 2007 National Survey of Nursing Homes (Doty et al., 2008). Each survey was assigned an identification number prior to distribution so that survey data could be linked with data from other sources following survey completion. The Commonwealth Fund survey was used because it was concise and included all of the major aspects of culture change principles and practices. Nineteen items that were not directly applicable to the study aims were omitted. Omitted items included topics such as number of employees, rate of staff turnover, future plans for PCCPs, and number of short-stay residents. The Commonwealth Fund 2007 National Survey of Nursing Homes was completed by Directors of Nursing and did not assess NHAs’ influence on PCCPs. Therefore, three items were added to the survey for the present study to assess the NHAs’ belief regarding the principles and feasibility of implementation and level of commitment toward change. In addition, demographic information about the NHA was added to the survey. The survey was made available to administrators in paper and web-based format. The web-based survey was created using Survey Monkey. Information about structural and market characteristics was gathered from public websites, with the exception of Medicaid and Medicare census which was derived from items on the survey. Nursing home structural characteristic information was gathered from the Nursing Home Compare website (CMS & USDHHS, n.d.). Market information was gathered from the U.S. Census Bureau (2009) website and the U.S. Department of Agriculture (2008) website. Market concentration was measured by the Herfindahl–Hirschman Index (i.e., combination of market shares, or number of beds in each nursing home, by county).
Data Variables and Sources.
Note. PCCP = person-centered care practice.
Participant Recruitment and Survey Administration Procedures
Approval for this study was granted by the institutional review board of The University of Alabama. Initial contact was attempted with each administrator via telephone. Up to five telephone call attempts were made in an effort to reach each administrator. The repeated telephone calls were made over a 2-week period after the initial call. When directly linked to the administrator via telephone, the principal investigator introduced herself and the study, encouraged participation, asked preference of survey method (web or paper format), and answered any questions.
Approximately 1 week after the initial phone calls, each participant was mailed an introductory letter and informed consent form. A paper survey was mailed along with the introductory letter for those administrators who requested a paper survey and/or could not be contacted by telephone. A web survey was emailed to the NHAs when requested, and the informed consent form was imbedded in first page of the web survey. Two follow-up reminders (by email or phone) were made after 1 month to increase response rate.
Measures
Dependent variables
Four dependent variables measured constructs related to adoption of PCCPs (see Table 1). The first dependent variable was the administrator’s perception of degree of PCCP implementation, which was assessed by a question on the survey that asked, “How well does this definition of person-centered care provided above describe this nursing home?” The measure was scored on a 5-point Likert-type scale ranging from not at all (1) to completely (5).
The remaining three dependent variables were averages of survey items related to resident autonomy, resident decision making, and direct care staff decision making. Each survey item was scored on a 5-point Likert-type scale ranging from not at all (1) to completely (5). The following survey items were averaged as a value representative of resident autonomy in the nursing home: “Is it the practice in this nursing home that residents can . . . (a) access food from the refrigerator if they want to? (b) access appliances necessary to prepare their own meal? (c) eat when they want? (d) eat where they want? (e) request and receive favorite foods when they are not on the menu? (f) go to bed when they want? (g) get up when they want? (h) choose when they bath or shower, even if they need assistance? and (i) choose how they are bathed?”
The following survey items were averaged as a value representative of resident decision-making practices in the nursing home: “How involved are residents in . . . (a) creating the schedule for meals? (b) planning menus? (c) creating the calendar for social events, activities, and outings? (d) planning of social events, activities, and outings? (e) decorating of communal areas? (f) decisions about who provides their own hands-on care? (g) developing the resident’s care plan? and (h) [involvement in] creating policies?”
The following survey items were averaged as a value representative of decision-making practices among direct care staff in the nursing home: “How involved are direct care workers in . . . (a) scheduling of staff shifts? (b) staff assignments to residents? (c) performance evaluations? (d) hiring and staff selections? (e) planning social events? (f) budget and resource requests? and (g) policy and procedure development?”
Predictor variables
Table 1 outlines the variables used to represent resources and competition as well as administrator factors. Structural factors included facility size, chain affiliation, ownership status, and Medicaid census, and Medicare census. Market factors included median income in a county, market concentration, market density, and location. Finally, administrator factors included current knowledge of culture change, agreement with culture change principles, belief about feasibility of implementation in the nursing home, and level of commitment to implementation. All of these variables were measured on the survey using a 5-point Likert-type scale ranging from not at all (1) to completely (5).
Analysis
An analysis of variance (ANOVA) was conducted between respondents and nonrespondents on structural and market data obtained from public websites to test for significant differences between the two groups. Multivariate analyses of variance (MANOVAs) were conducted to determine whether each set of predictor variables had an overall effect on the set of dependent variables. If an overall effect was found significant, univariate between-subjects tests were examined.
Results
Sample Characteristics
More NHA respondents were male (59.7%) and completed the paper survey (54.5%). The mean age of NHAs was 47.7 (SD = 9.8). Nonrespondents did not significantly differ from respondents regarding structural or market factors (Table 2). Medicaid and Medicare census was not included in this analysis because this information was derived from the survey that NHA respondents completed, thus that information was not available for nonrespondents.
Facility and Market Characteristics of Nursing Home Respondents and Nonrespondents.
PCCPs and Structural Factors
A MANOVA was performed to examine the relationship between the five structural characteristics (i.e., facility size, chain affiliation, ownership status, Medicaid census, and Medicare census) and the four PCCP dependent variables (i.e., perceived degree of PCCP implementation, resident autonomy, resident decision making, and direct care staff decision making; Table 3). The overall omnibus test showed a significant main effect for Medicaid census, Wilks’s λ = .84, F(4, 62) = 3.03, p = .02, ηp2 = .16. Given the significance of the overall test, the univariate between-subjects tests were examined. A significant univariate main effect for Medicaid census was obtained for the degree of perceived PCCP implementation, F(1, 66) = 4.80, p = .032, ηp2 = .08; the direction of the effect was such that a Medicaid census level below 50% was related to an increase in NHAs identifying their nursing home as one that is consistent with PCCPs.
Effect of Structural Factors on Person-Centered Care Practices.
Note. Univariate tests were analyzed only when multivariate tests were significant. However, all results are presented in this table. PCCP = person-centered care practice.
p < .05, and indicates a significant result for a univariate test that was not included in the final interpretation of results due to a nonsignificant multivariate test.
p < .05. **p < .001.
PCCPs and Market Factors
A MANOVA was performed to examine the effects of the four market factors (i.e., median household income in county, market concentration, market density, and rurality) on the four PCCP dependent variables (Table 4). There were significant multivariate effects for median household income and market density, Wilks’s λ = .83, F(4, 57) = 2.98, p = .028, ηp2 = .17, and Wilks’s λ = .81, F(4, 57) = 3.46, p = .027, ηp2 = .17, respectively. Univariate between-subjects tests were analyzed to determine effects. A significant main effect for median household income was achieved for two of the PCCP dependent variables, the NHAs’ perception of the degree of PCCP implementation and resident autonomy, F(1, 60) = 9.41, p = .003, ηp2 = .14, and F(1, 60) = 7.06, p = .011, ηp2 = .10, respectively. Administrators who worked in nursing homes that were located in wealthier counties were more likely to perceive their nursing home as being more patient centered. In addition, those nursing homes were more likely to allow greater resident autonomy among residents as indicated by the administrators.
Effect of Market Factors on Person-Centered Care Practices.
Note. Univariate tests were analyzed only when multivariate tests were significant. However, all results are presented in this table. PCCP = person-centered care practice.
p < .05, and indicates a significant result for a univariate test that was not included in the final interpretation of results due to a nonsignificant multivariate test.
p < .05. **p < .001.
Univariate analyses for market density showed a significant effect for NHAs’ perception of degree of PCCP implementation, resident autonomy, and staff decision making, F(1, 60) = 6.15, p = .016, ηp2 = .09; F(1, 60) = 11.05, p = .004, ηp2 = .13; and F(1, 60) = 4.40, p = .04, ηp2 = .07, respectively. As market density increased, NHAs perceived their nursing home to be more successful at implementing PCCPs, allowing greater resident autonomy, and promoting greater amount of staff decision-making practices.
PCCPs and Administrator Factors
A MANOVA was conducted to examine the effects of the four administrator factors (i.e., knowledge of culture change, agreement with PCCPs, belief about feasibility of PCCP implementation, and commitment to PCCP implementation) on the four PCCP dependent variables (Table 5). The results showed a significant multivariate effect for NHAs’ belief regarding feasibility of implementation in the nursing home, Wilks’s λ = .72, F(4, 63) = 6.07, p < .001, ηp2 = .28. The univariate analyses showed significance for three of the dependent variables. Specifically, NHAs who believed PCCP implementation was feasible perceived a higher degree of PCCP implementation, greater resident autonomy, and greater resident decision making in the nursing home, F(1, 66) = 12.27, p < .001, ηp2 = .16; F(1, 66) = 8.64, p = .005, ηp2 = .13; and F(1, 72) = 12.32, p = .001, ηp2 = .16, respectively.
Effect of Administrator Factors on Person-Centered Care Practices.
Note. Univariate tests were analyzed only when multivariate tests were significant. However, all results are presented in this table. PCCP = person-centered care practice.
p < .05, and indicates a significant result for a univariate test that was not included in the final interpretation of results due to a nonsignificant multivariate test.
p < .05. **p < .001.
Discussion
This study aimed to identify structural, market, and administrator characteristics that are associated with the implementation of PCCPs. Overall, there was partial support for the resource-based theory of competitive advantage. The findings were also partially supportive of the influence NHAs have on implementation of PCCPs. The results indicate that a complex set of intra and interfacility factors must come together if facilities are to undertake culture change efforts.
Two resource-related factors, Medicaid census and median household income in a county, were significant predictors of NHAs’ perception of PCCP implementation. As hypothesized, a higher Medicaid census and lower median household income in a county resulted in a less favorable perception of PCCP implementation among NHAs. We posit that these findings may be reflective of a lack of financial resources in facilities that rely more heavily on Medicaid payments. Medicaid reimbursement levels vary by state, but in all states, the federal matching rate for services is low (Kaiser Family Foundation, 2016). For example, the federal matching rate in Alabama was 77% in 2010 and 68% in 2011. Moreover, a lower median household income is representative of fewer market resources available to allocate to nursing home innovations. Administrators in facilities with a high Medicaid census located in counties with fewer resources may believe that they lack financial opportunities to implement PCCPs. Indeed, there is evidence that culture change implementation likely increases cost of care at the start of the intervention; however, these costs may fall below baseline after the first year (Coleman et al., 2002).
Market density, a measure of competition, significantly predicted NHAs’ perception of their facilities’ implementation and adoption of PCCPs. Administrators in nursing homes situated in counties with a higher population of older adults were more likely to perceive their facility as being supportive of PCCPs. Higher market density also predicted more resident autonomy and greater decision making among direct care staff as perceived by the administrators. We posit that, consistent with the resource-based theory of competitive advantage, having a higher concentration of potential customers in a county may be a motivating factor for nursing homes to implement PCCPs.
Although resource and competitive-based characteristics are found to play a role in PCCP implementation, nursing home leaders are ultimately the guiding factor in any innovation. This study found that NHAs’ belief regarding feasibility of implementation was a significant predictor, though NHAs’ familiarity with, attitudes toward, and commitment to culture change did not affect the implementation of PCCPs. Increased feasibility of PCCP implementation predicted an increase in perceptions of nursing home alignment with PCCPs, greater allowance of resident autonomy, and greater allowance of resident decision making. Many of the NHAs in Alabama were familiar with the concept of culture change, but this familiarity did not necessitate adoption of such practices. Because NHAs play an integral part in the daily operations of the nursing home by serving as liaison between direct care staff and administration, it is logical to assume they would also serve as a driving force for PCCP implementation. Implementation of facility-wide innovations can require extensive change from the practice norms, and commitment from all employees is necessary for success. Therefore, a NHA may believe in culture change principles, but unless the NHA believes it is feasible to implement such practices, he or she will have difficulty promoting PCCPs. Essentially, NHAs who believe that PCCP implementation is an unattainable goal will not attempt to implement resident-centered care practices or policies. With this understanding of how belief of feasibility can play a crucial role in NHA influence, one can make the argument that it is important to prepare the nursing home for change by warming the soil and educating staff on the importance of such principles and the positive outcomes that could ensue.
Limitations
The resource-based theory of competitive advantage was used as the framework for identifying structural and market variables. A limitation of this theory is that it cannot account for specific managerial actions (Barney, 2001; Priem & Butler, 2001). The theory posits that organizations will move toward innovations that improve competitiveness or profit, but not how that happens. It is possible a manager could choose another type of innovation or investment over implementation of PCCPs.
A second limitation of the study is related to the method that was used for obtaining information about the nursing homes’ practices and level of implementation of PCCPs. Administrators provided information related to PCCPs in the nursing home, and this information may not be accurate if the administrator did not routinely communicate with direct care staff or the Director of Nursing. In addition, NHAs could have responded to survey items in a biased manner, though it is assumed that they answered questions accurately considering they were aware their anonymity would be honored. For example, many NHAs responded to survey items that portrayed their nursing home as being more traditional rather than patient centered. This would suggest that these NHAs were not falsifying information to portray the nursing home in a favorable light.
Another limitation of the present study concerns the evidence for validity of the chosen subset of survey items. The survey items derived from the Commonwealth Fund 2007 National Survey of Nursing Homes (Doty et al., 2008) were developed by experts in the field, have been used in other studies (Miller et al., 2014; Sterns, Miller, & Allen, 2010), and have content validity. However, we used only a subset of these items in the current research, in a configuration that has not previously been used. In addition, we developed three additional survey items that measured NHAs’ agreement with and opinions about the feasibility of PCCP implementation and level of commitment to change. Thus, the interpretation of these items must be undertaken with caution because there is no existing psychometric evidence for these three items.
A fourth limitation, which is an extension of the aforementioned topic, is lack of another opinion or perception of the current state of each participating nursing home. The national Commonwealth Fund study that investigated nationwide implementation of PCCPs assessed opinions and perceptions from the Directors of Nursing (Doty et al., 2008). It is reasonable to assume that administrators and Directors of Nursing may have differing perceptions or opinions of the practices in the nursing home, especially because Directors of Nursing generally have more contact with direct care staff compared with administrators.
Finally, generalizability may be limited due to the sample population, sample size and response rate. This study included only nursing homes in Alabama, which may affect the generalizability of the findings. However, the results of the Commonwealth Fund national study suggest that there is little difference in culture change implementation among regions, though there were more nonrespondents from the southern region (Doty et al., 2008). The response rate for this study was 33%, which is comparable with the 37% response rate of the Commonwealth Fund national study (Doty et al., 2008). According to the survey literature, survey response rates have progressively declined over the years (Cook, Heath, & Thompson, 2000; Dey, 1997; Sheehan, 2001). It is important to note that there were no significant differences between respondents and nonrespondents on facility and market characteristics, suggesting that the results of the study may be an accurate representation of the state of PCCPs in the state of Alabama. Due to the inability to gather information from NHA nonrespondents, NHA characteristics were not compared; thus, it is possible that there may be differences among NHA respondents and nonrespondents. However, a low response rate is not necessarily indicative of response bias, and a sample obtained with a low response rate may still allow for an accurate representation of the population (Asch, Jedrziewski, & Christakis, 1997).
Implications for Future Studies
This study tested the resource-based theory of competitive advantage and its predictive ability in implementation of PCCPs. Two resource-driven factors (Medicaid census and median income in county) contributed to the adoption of culture change. Results from this study are consistent with findings from a recent study by Grabowski, Elliot, et al. (2014) in that long-term care facilities with greater resources are more likely to implement PCCPs. Although there is little research on the true cost of such an innovation, the small literature that is available is slowly growing and is suggestive of cost expenditures in the short term but cost savings in the long term (Coleman et al., 2002; Jenkens, Sult, Lessell, Hammer, & Ortigara, 2011). More research is needed to understand the necessary resources for successful implementation and maintenance of PCCPs.
This study also investigated NHA perceptions, but administrators are only one piece of the nursing home puzzle. More research is needed to understand the similarities and/or discrepancies between NHA perceptions and other employees such as the director of nursing, chief executive officer, and direct care staff. In addition, it would be helpful to know whether there are patterns of consistency in perceptions among staff and if these perceptions provide an accurate representation of nursing home culture as measured by validated culture change measures.
In sum, the results of this study indicated that Medicaid census, market density, median household in a county, and NHA perception of implementation feasibility contributed to the adoption of PCCPs in nursing homes. The conditions required for adoption of PCCPs are thus complex and are located both within and outside of the facility. We posited that such variables would likely serve as moderators of PCCP adoption because they covary with the types of resources and types of priorities/values that are concordant with PCCPs. However, our findings are only correlational. More research is needed to understand causality and specifically how these factors play a role in PCCP implementation, especially considering the existence of mixed findings in the literature regarding the effect of these factors on innovation implementation more broadly. In the introduction, we speculated about possible ways in which structural factors could affect PCCP adoption, and noted that some of these influences might work in opposite directions (e.g., chain status is typically associated with for-profit status which might be expected to negatively influence PCCP adoption and yet is also typically associated with a strong centralized governance that might also provide key resources for a large-scale systemic change like PCCP). It will be important for future research to examine whether such possibly opposing forces could serve as confounds, or as suppressor variables, essentially canceling out certain effects. Furthermore, most research has focused on macro-level variables such as market and structural factors. However, there is little research on micro level, individual factors as they relate to adoption of innovations. Such factors that may potentially play a role include, but are not limited to, wages of direct care staff, education level of direct care staff, education level of leadership members, job satisfaction of all employees, communication among staff, and motivation for change among all employees. Future research should focus on individual levels because the driving force of change is the people within the nursing home.
Footnotes
Authors’ Note
M. Lindsey Jacobs is now at the Salem Veterans Affairs Medical Center, Salem, Virginia. Jullet A. Davis is now at the School of Allied Health at Florida A&M University, Tallahassee, Florida.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.