An Empirical Typology of Nursing Home Culture Change Implementation

Abstract

Culture change in nursing homes (NHs) is a broad-based effort to transform NHs from impersonal institutions to genuine person-centered homes. Culture change practices have been implemented increasingly with varying levels of success. This study (a) generated an empirical typology of culture change implementation across Minnesota NHs using latent profile analysis based on the survey data from administrators in 102 NHs and (b) examined variations in NH characteristics and quality outcomes associated with the typology. Three types of culture change implementation were identified: high performers, average performers, and low performers. The distributions of culture change scores were distinct across the three types, with low performers lagging far behind others in family and community engagement, and end-of-life care. High performers were distinguished through demonstrating better resident quality of life and higher family satisfaction. The findings provide empirical support for policymakers, providers, and advocates to direct culture change expansion and resource allocation.

Keywords

nursing home culture change quality of care quality of life family satisfaction empirical typology latent profile analysis

Introduction

Culture change in nursing homes (NHs) is a broad-based and continuous effort to transform NHs from impersonal institutions to genuine person-centered homes, giving voices to the people living and working there (Koren, 2010). Culture change practices embrace changes in multiple domains of care including physical environment, resident-centered care, staff empowerment, staff leadership, and family and community engagement (Miller, Looze, et al., 2014). The implementation of culture change practices has been increasing in U.S. NHs with varying levels of success (Miller et al., 2018). A growing body of research has emerged to examine the effects of culture change practices on quality of care (Grabowski, Elliot, et al., 2014; Shier et al., 2014), and resident quality of life (QoL) and satisfaction with care (Hill et al., 2011; Kane et al., 2007; Poey et al., 2017). However, the findings were inconsistent, which may be in part due to difficulties in defining culture change and a wide variation in culture change implementation (Duan et al., 2020; Hill et al., 2011; Shier et al., 2014).

Culture change was initially proposed as a care paradigm, and therefore its operationalization varies substantially across NHs (Rahman & Schnelle, 2008). A heuristic typology that is solely based on opinions of NH administrators or other external evaluators was commonly used to identify the types of culture change implementation (e.g., full adopters, partial adopters, strivers, or nonadopters; Grabowski, Elliot, et al., 2014; Grant, 2008; Miller, Looze, et al., 2014; Poey et al., 2017; Zimmerman et al., 2013). However, the heuristic approach is limited because it may be subject to response bias or low replicability (Winch, 1947). An empirical typology that is derived from data of systematic measures of culture change has not been attempted. Contrary to a heuristic typology, an empirical typology applies statistical techniques based on empirical measures and has been widely used in social research to study complex and dynamic phenomena (Winch, 1947).

Accordingly, generating an empirical typology of culture change implementation in NHs requires a comprehensive and valid measure of culture change practices. A number of domain-specific measures of culture change practices in NHs have emerged (Sturdevant et al., 2018). However, existing studies using these standardized measures tended to examine culture change domains separately or sum all subscales to determine the overall adoption level (Chisholm et al., 2018; Miller et al., 2018; Sullivan et al., 2013). Identifying an empirical typology of culture change implementation by looking beyond a single aspect of practices will help care providers and policymakers make informed decisions toward allocating optimal resources and directing tailored interventions for NHs at a certain stage of culture change implementation.

A further examination of variations in NH characteristics and quality outcomes across the types of culture change implementation is crucial to fully inform the implementation of evidence-based culture change initiatives. A gap in the literature was the lack of voices from residents and family members speaking to the influence of culture change (Duan et al., 2020; Hill et al., 2011; Shier et al., 2014). Despite widely acknowledged face validity that culture change improves QoL and satisfaction, few studies have examined these resident- and family-reported outcomes (Hill et al., 2011; Shier et al., 2014). This may be likely due to a lack of QoL indicators in the current quality metrics for NH care (Castle & Ferguson, 2010). Minnesota is useful for this study because it is one of the few states that integrates resident-reported QoL and family-reported satisfaction in its quality measure system (Minnesota Department of Human Services, 2019), which enables a systematic examination of QoL and satisfaction as outcomes of culture change implementation.

To address the knowledge gaps, this study aimed to (a) generate an empirical typology of culture change implementation using latent profile analysis, (b) examine NH structural and organizational characteristics that are associated with the types of culture change implementation, and (c) examine variations in quality outcomes including clinical quality indicators (QIs), resident QoL, and family satisfaction across the types of culture change implementation.

Method

Sample

This cross-sectional study used data from an online culture change survey and administrative data of NHs in Minnesota. The survey was administered to NH administrators in all Medicare- and/or Medicaid-certified NHs in Minnesota (n = 363) through an online survey tool between August 2018 and January 2019. Administrative data were obtained from the Minnesota Department of Human Services and consisted of NH characteristics and quality measures. The University of Minnesota Institutional Review Board reviewed this study and determined it was exempted from human subject research because no personal questions of respondents were involved in the survey and only facility-level data were used.

Study Variables and Data Sources

Culture change practices

The survey instrument was adapted from a culture change assessment tool developed by researchers at Brown University (Miller, Looze, et al., 2014). The tool measures six domains of culture change practices including physical environment, staff empowerment, staff leadership, resident-centered care, family and community engagement, and end-of-life care. Table 1 lists items comprising each domain. Items in the domain of physical environment were measured using a two-level Likert-type scale (0-1), and items in other domains were measured using a three-level Likert-type scale (0-2). As suggested by the previous use of the instrument (Miller et al., 2018), a composite score was obtained for each domain by summing the raw item scores, which were then rescaled to the range of 0 to 100. The missing value of an item was imputed using the mean of completed items in a given domain if one or two items were missing for that domain (imputations were performed for 1–11 NHs per domain). Domain scores were reported as missing if more than two items had missing values (missing domain scores were reported for 5–9 NHs per domain). This instrument has been validated with satisfactory content validity and high internal consistency (Miller et al., 2018; Tyler et al., 2011). Cronbach’s alpha of each domain based on the current sample ranged from .43 for physical environment to .79 for family and community engagement (Table 1).

Table 1.

Study Variables and Data Sources.

Variables	Measures/coding	Data sources
Culture change practices
Physical environment transformation (α = .43)	12 items (e.g., private rooms, outdoor spaces, open kitchen and dining, eliminate nurse station and overhead page); 0–100	Online survey of MN nursing home administrators by the authors
Staff empowerment (α = .76)	13 items (e.g., self-managed work schedules, cross-training, consistent staff assignment, rewards for extra education); 0–100
Staff leadership (α = .63)	10 items (e.g., formal opportunities for CNAs to participate in management activities, in-service education); 0–100
Resident-centered care (α = .58)	Nine items (e.g., residents’ involvement in determining their schedules, activities, and care); 0–100
Family and community engagement (α = .79)	Nine items (e.g., formal opportunities for family and community members to participate in care and social activities); 0–100
End-of-life care (α = .70)	Six items (e.g., fulfill various needs of a terminally ill resident, provide emotional support for family members); 0–100
Administrator-reported CC	Three categories including traditional facility, striver, or adopter
Structural and organizational characteristics
For-profit	Yes/no	2017 Cost Reports submitted by facilities to the MN Department of Human Services
Affiliated with a chain	Yes/no
Geographic location	Three categories including metropolitan, micropolitan, and rural
Size	Number of active beds
Occupancy rate	Number of occupied beds divided by the total number of active beds
Payer mix	Percentage of resident days paid by a certain payer (Medicaid, Medicare, and private insurance or private pay)
Staffing	Hours of certain type of staff per resident day (RNs, LPNs, CNAs, mental health and social work staff, and activity staff)
Quality outcomes
Resident QoL summary score	Average of eight QoL domain scores; 0–100	2018 MN Nursing Home Report Card
Resident QoL domain scores
Meaningful activities	Five items (e.g., enough scheduled and enjoyable activities); 0–100
Food enjoyment	Four items (e.g., enjoy the food, menu changes enough); 0–100
Environment	Eight items (e.g., easy to get around, enough privacy, feel safe); 0–100
Dignity	Five items (e.g., staff listen to residents, treat residents politely); 0–100
Autonomy	Four items (e.g., choose time to get up, express preferences); 0–100
Relationships	Four items (e.g., staff stop by just to talk); 0–100
Caregiving	Nine items (e.g., get help in a timely way); 0–100
Mood	Nine items (e.g., often feel angry, bored, or happy, relaxed); 0–100
Family satisfaction summary score	Average of four satisfaction domain scores; 0–100	2018 MN Nursing Home Report Card
Family satisfaction domain scores
Care	12 items (e.g., include family’s opinions in care planning); 0–100
Staff	Eight items (e.g., staff know residents, staff’s attitude); 0–100
Environment	Six items (e.g., smell, cleanliness, safety); 0–100
Food	Three items (e.g., quality of food, atmosphere at mealtime); 0–100
Clinical QI summary score	Sum of 10 clinical QI domain scores; 0–100	2018 MN Nursing Home Report Card
Clinical QI domain scores^a
Psychosocial	Behavior or depressive symptoms; 0–10
Physical restraints	Use of physical restraints; 0–10
Continence	Bowel or bladder incontinence, indwelling catheters; 0–10
Infections	Urinary tract infections or other infections; 0–10
Accident	Falls with major injury; 0–10
Nutrition	Unexplained weight loss; 0–10
Pain	Moderate to severe pain; 0–10
Skin care	Pressure sores; 0–10
Psychotropic drugs	Use of antipsychotics without a diagnosis of psychosis; 0–10
Physical functioning	Functional or mobility dependence; 0–10

Note. CNAs = certified nursing assistants; CC = culture change; RNs = registered nurses; LPNs = licensed practical nurses; QoL = quality of life; QI = quality indicator.

The original measure of a clinical QI domain is the percentage of residents with certain conditions. Raw percentages were risk-adjusted and rescaled to 0 to 10 points with higher scores indicating better outcomes.

The survey also asked a single question about overall culture change implementation level as perceived by NH administrators (Miller, Looze, et al., 2014). The responses were categorized as traditional facility (there is no discussion around culture change or culture change is under discussion but no change in care delivery occurs), striver (culture change has partially changed care delivery in some, or all areas of the organization), and adopter (culture change has completely changed care delivery in some or all areas of the organization).

NH structural and organizational characteristics

Cost reports submitted by facilities to the Minnesota Department of Human Services were used to obtain data of several structural and organizational NH characteristics known to be associated with the implementation of culture change practices (Miller, Looze, et al., 2014; Miller et al., 2018). These variables included profit status, chain affiliation, geographic location, size, occupancy rate, payer mix, and staffing (Table 1).

Quality outcomes

Facility-level risk-adjusted quality measures on resident QoL, family satisfaction, and clinical QIs were from publicly available data published in the Minnesota Nursing Home Report Card (Minnesota Department of Human Services, 2019). Resident QoL and family satisfaction data are collected through face-to-face interviews or surveys with a random sample of residents or family members in every NH (Table 1). The Minnesota Department of Human Services contracts with an outside research vendor to complete these interviews and surveys annually (Minnesota Department of Human Services, 2019). The QoL and family satisfaction survey instruments have been validated with respect to validity and reliability with Cronbach’s alpha ranging from .53 to .77 for resident QoL domains (Kane et al., 2003) and .86 to .96 for family satisfaction domains (Shippee et al., 2017). Facility-level QoL scores and family satisfaction scores that used the average score of interviewed individuals were risk-adjusted to control for individual and facility characteristics that were generally not a result of provider performance. The QoL scores adjust for four resident-level factors including age, gender, cognitive, and activities of daily living (ADL) and one facility-level factor, that is, geographic location; the family satisfaction scores adjust for six risk factors, five for respondents (i.e., relationship to the resident, gender, frequency of visits and other communication with the resident, and survey format) and one for NHs (i.e., geographic location) (Minnesota Department of Human Services, 2019).

Clinical QIs in 10 quality domains were derived from residents’ Minimum Data Set (MDS) assessments (Table 1). The MDS clinical QI scores were risk-adjusted to account for differences among the residents served in NHs. Examples of the adjustors included age, gender, cognitive performance, Alzheimer’s disease, stroke, and ADL (Minnesota Department of Human Services, 2019).

Analytic Plan

Latent profile analysis was used to generate a typology of culture change implementation. Latent profile analysis is a probability-based clustering technique that aims to identify hidden groups from observed data of continuous variables using maximum likelihood techniques (Oberski, 2016). It outperforms traditional clustering methods such as K-means by allowing unbiased estimation of profile means and providing various diagnostics for determining numbers of profiles and for comparing models (Magidson & Vermunt, 2002). In this study, variables used for generating latent profiles included scores of five culture change domains (in a 0–100 scale, i.e., physical environment, staff empowerment and staff leadership, resident-centered care, family and community engagement, and end-of-life care). Staff empowerment and staff leadership were strongly correlated (r = .63, p < .05). Therefore, to ensure the assumption of local independence, these two domains were combined by taking the average of the two domain scores.

The first analysis step was to establish the optimal number of profiles. To do this, several models with differing numbers of profiles were created and their model fit indices, including log likelihood and Bayesian information criterion (BIC), were compared. Each participating NH was assigned to a profile based on the highest predicted posterior profile probability. Predicted means of culture change scores were generated for each profile. Second, analysis of variance (ANOVA) or chi-square test was applied to examine variations in culture change scores and NH characteristics across profiles. Bonferroni tests were conducted to adjust the multiple comparisons across profiles. Finally, a set of regression models were fitted to examine across-profile variations in quality outcomes, controlling for NH characteristics that may be associated with the quality measures (e.g., profit status, geographic location, chain affiliation, size, occupancy, staffing; Shippee et al., 2015; Xu et al., 2013). Post-stratification was applied in ANOVA, chi-square tests, and regression analyses to adjust the sampling weights so that they sum to the population sizes within each post-stratum. Post-strata were determined based on profit status and geographic locations. All analyses were conducted in Stata 15.0 (StataCorp LLC, 2017).

Results

Administrators from 102 NHs participated in the survey with a response rate of 28.1%. No significant differences in NH characteristics (i.e., geographic location, profit status, size, occupancy, payer mix, and staffing in registered nurses, licensed practical nurses, mental health or social services staff, and activity staff) and quality outcomes (i.e., clinical QIs, resident QoL, and family satisfaction) were observed between participants and nonparticipants, except that participants were less likely to be affiliated with a chain and had slightly higher certified nurse assistant staffing (see Supplementary Material). Most of the participating NHs were nonprofit or government owned (75.73%) and located in metropolitan area (56.31%). About half of the participating NHs were affiliated with a chain (49.5%). On average, participating NHs had 76 active beds (range = 14–320), 84.40% occupancy rate (range = 35.2%–98.9%), and 53.17% Medicaid resident days (range = 0.2%–91.9%). The highest culture change domain score was observed for end-of-life care (77.74 ± 22.43), followed by resident-centered care (69.14 ± 17.02) and physical environment transformation (64.75 ± 14.66). The culture change domain scores were relatively lower in family and community engagement (27.90 ± 18.81), staff empowerment (38.10 ± 16.87), and staff leadership (39.43 ± 16.74).

An empirical typology with three types of culture change implementation was generated based on the latent profile analysis. The three-profile model had the best model fit (log likelihood = −1,998.20, BIC = 4,098.15) compared with models with two (log likelihood = −2,015.39, BIC = 4,104.78), four (log likelihood = −1,990.49, BIC = 4,110.48), or five (log likelihood = −1,988.58, BIC = 4,134.40) profiles. Figure 1 presents predicted means of five culture change domain scores for the three types of culture change implementation, labeled as low performers, average performers, and high performers. According to the predicted latent profile probabilities, about 14.19% NHs were classified as low performers, 54.92% were average performers, and 30.89% were high performers.

Figure 1.

Predicted means of culture change scores by latent profiles.

Table 2 presents culture change scores across the types of culture change implementation. All culture change scores varied significantly across the three types with F = 13.64–127.66 (p < .001). According to multiple comparisons with Bonferroni correction, each type was statistically different from one another in four culture change scores including physical environment, staff empowerment and staff leadership, resident-centered care, and end-of-life care. In general, higher levels of performance were associated with higher culture change scores. No statistical difference was found between low performers and average performers in family and community engagement. As shown in the first four rows of Table 3, administrator-reported culture change levels were significantly associated with the empirical typology. About 79% low performers and 88% average performers were self-identified as strivers. Although 44% high performers were self-identified as adopters, about half of them were self-identified as strivers.

Table 2.

Culture Change Scores by the Types of Culture Change Implementation.

Domains of culture change practices	Low performer (n = 13)		Average performer (n = 59)		High performer (n = 30)		F
Domains of culture change practices	M	SE	M	SE	M	SE	F
Environment transformation	50.63	2.66	61.41	1.56	75.40	1.48	52.54***
Staff empowerment and staff leadership	23.98	2.31	34.44	1.26	53.71	1.68	81.45***
Resident-centered care	59.85	4.07	66.79	1.68	77.39	2.04	13.64***
Family and community engagement^a	17.25	3.25	19.60	1.22	49.48	2.52	63.59***
End-of-life care	33.62	3.46	80.61	1.68	89.86	1.63	127.66***

Note. Weighted results are presented. SE is the linearized standard error. Bonferroni tests were conducted for multiple comparisons in ANOVA. Each profile was statistically different from one another in all culture change scores except for family and community engagement. ANOVA = analysis of variance.

No statistical difference was found in family and community engagement between low performers and average performers.

***

p < .001.

Table 3.

Nursing Home Characteristics by the Types of Culture Change Implementation.

Nursing home characteristics	Low performers (n = 13)		Average performers (n = 59)		High performers (n = 30)		F/χ²
Nursing home characteristics	M/%	SE	M/%	SE	M/%	SE	F/χ²
Administrator-reported culture change level
Traditional	17.13%		3.77%		3.95%		20.89**
Striver	78.92%		87.67%		51.58%
Adopter	3.94%		8.58%		44.47%
Geographic location
Metropolitan	61.12%		58.75%		49.01%		2.72
Micropolitan	25.51%		14.41%		24.57%
Rural	13.29%		26.84%		26.42%
For-profit	38.44%		30.96%		31.56%		2.36
Affiliated with a chain	62.62%		47.00%		62.46%		2.63
Number of active beds	98.67	18.53	73.70	4.64	76.07	6.51	1.06
Occupancy	0.84	0.01	0.85	0.01	0.82	0.02	0.24
Staffing (HPRD)
Registered nurse	0.63	0.14	0.54	0.02	0.65	0.05	0.23
Licensed practical nurse	0.73	0.05	0.74	0.03	0.74	0.04	0.19
Certified nursing assistant	2.47	0.20	2.34	0.06	2.41	0.07	0.08
Mental health and social services staff	0.13	0.01	0.12	0.01	0.13	0.01	0.61
Activity staff	0.29	0.02	0.29	0.01	0.28	0.02	0.12
Proportion of Medicaid resident days^a	0.59	0.02	0.52	0.02	0.57	0.03	4.49*
Proportion of Medicare resident days	0.10	0.01	0.10	0.01	0.09	0.01	1.63
Proportion of private-pay resident days^b	0.31	0.02	0.38	0.01	0.34	0.02	3.68*

Note. Weighted results are presented. SE is the linearized standard error. Bonferroni tests were conducted for multiple comparisons in ANOVA. ANOVA = analysis of variance; HPRD = hours per resident day.

Low performers had a significantly higher proportion of Medicaid resident days compared with average performers (p = .004). ^b Low performers had a significantly lower proportion of resident days paid by private insurance or others compared with average performers (p = .008).

p < .05. **p < .01.

Table 3 shows NH characteristics by the three types of culture change implementation. Bonferroni-corrected multiple comparisons indicated that low performers had a higher proportion of Medicaid resident days but a lower proportion of private-pay resident days compared with average performers.

Table 4 presents variations in quality outcomes across the types of culture change implementation after controlling for NH characteristics. High performers had significantly better outcomes in QoL summary score and all QoL domain scores compared with average performers. High performers also reported higher scores in four QoL domains including meaningful activity, environment, dignity, and autonomy than low performers. In addition, high performers reported higher family satisfaction scores in domains of environment and food compared with both lower performers and average performers. With respect to clinical QIs, high performers demonstrated significantly better outcomes in the use of physical restraints and skin care, but reported a poorer outcome in accidental falls compared with low performers.

Table 4.

Regression Analyses: Quality Outcomes by the Types of Culture Change Implementation (Reference Group = High Performers).

	Low performers		Average performers
	B	95% CI	B	95% CI
Clinical QIs
Clinical QI summary score	−0.34	[−6.01, 5.34]	−3.81	[−7.67, 0.05]
Psychosocial	0.22	[−0.72, 1.16]	0.16	[−0.57, 0.89]
Physical restraints	−0.5	[−1.20, 0.21]	−1.49***	[−2.33, –0.65]
Continence	0.03	[−0.73, 0.80]	−0.48	[−1.11, 0.15]
Infections	0.00	[−1.21, 1.22]	−0.42	[−1.25, 0.41]
Accident	1.35*	[0.06, 2.64]	0.25	[−0.77, 1.28]
Nutrition	−0.75	[−2.10, 0.59]	0.05	[−0.97, 1.07]
Pain	0.47	[−0.90, 1.84]	−0.65	[−1.66, 0.37]
Skin care	−0.44	[−1.38, 0.50]	−0.88*	[−1.61, –0.15]
Psychotropic drugs	−0.81	[−2.23, 0.61]	−0.37	[−1.44, 0.70]
Physical functioning	−0.25	[−1.10, 0.61]	−0.58	[−1.26, 0.09]
Resident QoL
QoL summary score	−1.45**	[−2.46, –0.44]	−1.85***	[−2.54, –1.16]
Meaningful activities	−2.34*	[−4.39, –0.30]	−2.30***	[−3.58, –1.01]
Food enjoyment	−2.20	[−4.52, 0.13]	−3.03***	[−4.62, –1.44]
Environment	−1.62**	[−2.73, –0.50]	−1.04**	[−1.82, –0.26]
Dignity	−1.32***	[−1.96, –0.68]	−1.13***	[−1.60, –0.66]
Autonomy	−1.52*	[–2.76, –0.28]	−2.11***	[−2.96, –1.26]
Relationships	−0.45	[−2.09, 1.18]	−2.11***	[−3.24, –0.98]
Caregiving	−1.70	[−3.41, 0.01]	−2.37***	[−3.67, –1.07]
Mood	0.17	[−1.04, 1.39]	−0.99	[−2.13, 0.16]
Family satisfaction
Satisfaction summary score	−2.22	[−4.66, 0.23]	−1.47	[−3.17, 0.23]
Satisfaction with care	−0.98	[−3.36, 1.40]	−1.17	[−2.86, 0.51]
Satisfaction with staff	−1.00	[−3.50, 1.49]	−0.75	[−2.38, 0.89]
Satisfaction with environment	−3.98**	[−6.84, –1.13]	−2.05*	[−4.03, –0.06]
Satisfaction with food	−2.77*	[−5.17, –0.36]	−1.73*	[−3.34, –0.12]

Note. Weighted results are presented. The regression analyses controlled for covariates including geographic location, profit status, chain affiliation, number of active beds, occupancy, staffing (including licensed practical nurses, certified nursing assistants, mental health and social service staff, activity staff), and the percentage of Medicaid resident days. Complete presentation of the regression analyses can be found in Supplementary Material. CI = confidence interval; QI = quality indicator; QoL = quality of life.

p < .05. **p < .01. ***p < .001.

Discussion

The prevalence of culture change practices in Minnesota NHs was comparable to that of a national representative sample of NHs (Miller et al., 2018), wherein culture change practices associated with resident-centered care, physical environment transformation, and end-of-life care were more frequently implemented than practices associated with staff empowerment, staff leadership, and family and community engagement. This study, looking beyond individual culture change domains, integrated the multidimensional measures of culture change practices to identify an empirical typology of culture change implementation. Three types of culture change implementation were identified across Minnesota NHs. High performers appeared the most comprehensive in adopting all culture change domains, and they were particularly distinguished from others by their excellent performance in family and community engagement. Conversely, low performers reported the lowest scores in all culture change domains, and they particularly lagged behind in end-of-life care. Average performers were moderate in implementing culture change domains including environment transformation, staff empowerment and staff leadership, and resident-centered care. Yet, they had the end-of-life care score comparable to high performers and the family and community engagement score close to low performers.

The findings were consistent with previous studies that suggested NHs are at different stages in implementing culture change practices, reflecting the progressive nature of culture change in an NH (Miller, Looze, et al., 2014; Miller et al., 2018). However, previous studies tended to ask informants such as NH administrators or directors of nursing to gauge the overall level of culture change implementation. To this end, this study revealed that the empirical typology based on latent profile analysis was associated but not completely consistent with administrator-reported culture change levels, namely, the heuristic typology. A substantial portion of NHs identified as traditional facilities or strivers by their administrators were actually classified as high performers in our typology. In contrast, some administrators were less conservative as they identified their facilities as adopters which fell into the category of average performers in our typology. The inconsistency between the heuristic and empirical culture change typologies is likely due to variations in NH administrators’ personal knowledge, practical experiences, and expectations of culture change practices. In prior qualitative interviews, NH administrators have discussed various motivations, challenges faced, and strategic plans for the culture change implementation in their facilities (Shield et al., 2014). Administrators may use the relevant information in an implicit manner when gauging the level of culture change implementation and base the gauge on their own scales (comparing themselves to some ideal culture change models or comparing their current status to their past status). In contrast, the empirical typology, based on empirical measures of culture change practices and statistic techniques, maximizes both within-group homogeneity and between-group heterogeneity in terms of the implementation of multiple culture change domains. This may explain why the administrator-reported culture change levels were not as sensitive as the empirical typology in the tests of NH characteristics and quality outcomes associated with culture change implementation (the results are available upon request).

Although some culture change domains including physical environment transformation, staff empowerment and staff leadership, and resident-centered care demonstrated an even hierarchical distribution across the three types, family and community engagement and end-of-life care appeared extremely high or low in certain groups of NHs. This finding was not surprising because culture change practices associated with family and community engagement and end-of-life care have not been advocated as widely as other culture changes practices by professional organizations and regulation agencies (Rahman & Schnelle, 2008). For instance, resident-centered care practices that emphasize honoring residents’ preferences have been mandated by federal regulations (Centers for Medicare & Medicaid Services, 2016). Home-like environments and empowering direct staff have also been extensively promoted by those leading culture change models such as the Green House Project, the Household Model by Action Pact, and the Wellspring Model (Cohen et al., 2016; Hill et al., 2011; Kehoe & Van Heesch, 2003). Moreover, initiatives in family and community engagement and end-of-life care involve more stakeholders, more expertise, but less immediate outcomes (Puurveen et al., 2018; Schwartz et al., 2019; Zimmerman et al., 2013). The complexities involved in these initiatives may hinder their expansion in NHs that are at an early stage of culture change implementation (Sterns et al., 2010). Although more empirical evidence is needed (particularly for the effects of community engagement), existing studies have demonstrated the beneficial influence of family engagement and resident- and family-centered end-of-life care on resident, family, and staff outcomes (Hanson et al., 2005; Zimmerman et al., 2013). Given that such complex culture change initiatives were scarcely embraced by low and/or average performers, additional support from policies, resources, and expertise should be offered to help these NHs achieve comprehensive culture change.

The finding that culture change implementation was associated with payer mix was in line with previous studies (Chisholm et al., 2018; Grabowski, Elliot, et al., 2014; Miller et al., 2018). The findings from previous studies using the national data or the data of other states may reflect the fact that culture change requires considerable financial resources (Chisholm et al., 2018; Grabowski, Elliot, et al., 2014; Miller et al., 2018), and higher pay from Medicare and private payers versus Medicaid may facilitate culture change implementation (Chisholm et al., 2018; Lepore et al., 2015; Shield et al., 2014). However, in Minnesota, NHs cannot charge private-pay residents more than the Medicaid rate, with the exception of private rooms and special services (Minnesota House Research Department, 2016). As a result, factors contributing to the association between culture change adoption and payer mix for Minnesota NHs could be other facilitators such as higher occupancy rates, higher staffing, and better quality outcomes in NHs with less Medicaid residents and more private-pay residents (referred to as high-tier NHs; Mor et al., 2004).

Consistent with previous studies, this study indicated that a higher level of culture change implementation was associated with lower rates of restraint use and pressure ulcers but a higher rate of accidental falls (Coleman et al., 2002; Miller, Lepore, et al., 2014; Ransom, 2000). It was still worth noting that other clinical outcomes did not show significant variations across the types of culture change implementation, which was in accordance with previous studies reporting only few clinical outcomes associated with culture change implementation (Grabowski, O’Malley, et al., 2014; Miller, Lepore, et al., 2014). This may be related to the fact that culture change efforts do not focus on a given clinical outcome or a certain dimension of care but on the holistic well-being of residents and staff (Grabowski, O’Malley, et al., 2014). Nonetheless, person-centered value embedded in culture change practices may help shape staff’s norms which encourage giving priority to resident dignity, autonomy, and overall well-being in care delivery (Loe & Moore, 2012; Munroe et al., 2011). This may explain why high performers had a relatively lower rate of restraint use but a higher rate of accidental falls (as residents may be granted more autonomy for transferring and ADLs; Coleman et al., 2002; Gastmans & Milisen, 2006; White-Chu et al., 2009). In addition, culture change practices that promote staff empowerment and close resident–staff relationships may contribute to staff monitoring resident’s skin condition more closely and being more attentive to resident needs (Ransom, 2000). Qualitative research using data from field observations or in-depth interviews will be helpful to understand how care value and care delivery process is transformed in culture change NHs that may affect quality outcomes, particularly for outcomes demonstrating a direction of change opposite to the hypothesis.

One contribution to the evidence base regarding the benefits of culture change on residents’ well-being is the finding that high performers consistently reported better QoL of all domains relative to average performers and/or low performers. Although more rigorous research is needed, previous studies suggested that culture change practices improve residents’ overall satisfaction and foster a sense of being respected in terms of autonomy, privacy, dignity, and personal preferences (Burack et al., 2012; Grant, 2008; Kane et al., 2007; Poey et al., 2017). The benefits on QoL were primarily observed in comprehensive and sustained culture change implementation such as the Green House Model (Burack et al., 2012; Grant, 2008; Kane et al., 2007; Poey et al., 2017). In the Kansas pay-for-performance program targeting person-centered care, only participating facilities that have reached the highest stage of culture change (as measured with Stage 1 “ foundation level” to Stage 5 “full implementation”) reported significantly higher resident-reported QoL and higher satisfaction as compared with non-implementers (Poey et al., 2017). High performers identified by this study demonstrated the most comprehensive culture change implementation and they even gained some momentum in practices associated with family and community engagement and resident- and family-centered end-of-life care, two culture change domains that appeared challenging for average performers and low performers. In high-performing NHs, the culture change implementation may not be limited to structural and procedural changes but rather have achieved a level at which the person-centered value has been embedded into staff’s care philosophy and routinized into daily practices, which in turn improves resident QoL.

The finding that being high performers was associated with better family satisfaction was consistent with previous research. Two studies on the Eden Alternative indicated that family members’ overall satisfaction increased after 2 years of the model implementation (Ransom, 2000; Robinson & Rosher, 2006). Of particular interest about the finding of this study was that only two domains of satisfaction (i.e., environment, food) were found significantly higher in high performers. Although family members might witness those tangible changes in the environment and food services in high-performing NHs and thus experienced improved satisfaction, they might remain critical of other aspects of care (i.e., family involvement in resident care, communication with staff, staff attitude, and respect toward the resident). This highlights the significance of resident- and family-centered culture change practices that improve family satisfaction overall given the growing recognition of the family members’ role in NH care (Gaugler, 2005; Shippee et al., 2017). These practices may need to focus on promoting family involvement in resident care and establishing interactive and partnered relationships between staff and family members (Gaugler, 2005; Shippee et al., 2017).

Limitations, Directions for Future Research, and Implications for Practice

Some limitations of this study should be addressed in future research. The small sample size may reduce statistical power to examine subtle aspects of the typology and to test differences in facility characteristics and quality measures across types of culture change implementation. Despite a low response rate, negligible nonresponse bias in regard to NH characteristics and the application of weight adjustment may increase the generalizability of the findings to all NHs in Minnesota. Nonetheless, the findings cannot be generalized to all U.S. NHs. A national sample of NHs is needed to generate typologies of culture change implementation at a national level. In addition, potential social desirability bias may undermine the validity of data, although previous studies have found that NH administrators are credible when answering the survey items (Shield et al., 2018). Direct care staff and staff from multiple departments should be included in future studies to assess culture change practices in a more comprehensive manner. Finally, this study only examined a cross-sectional snapshot of culture change implementation. Future research should consider longitudinal designs to examine cause–effect relationships between culture change implementation and quality outcomes, and to test the sustainability of the effects.

The findings have several implications for practice. The existing three types of culture change implementation could guide resource allocations. For example, resources to support low and average performers need to particularly focus on family and community engagement, and resident- and family-centered end-of-life care. Changes of care philosophy, besides changes of physical and operational aspects of care processes, should be emphasized in actions of promoting and implementing culture change practices in low- and average-performing NHs. The positive relationships between culture change implementation and quality outcomes (particularly QoL and satisfaction outcomes) support actions of policymakers, care providers, and advocates to promote culture change extensively within or across NHs.

Conclusion

NHs have been committed to implementing culture change practices but with varying levels of success. This study generated a typology of culture change implementation based on empirical data. NHs with different types of culture change implementation demonstrated variations in organizational characteristics and quality outcomes. The findings highlight the value of the empirical typology approach to identifying types of culture change implementation across NHs based on comprehensive measures of culture change practices. The typology will provide empirical support for policymakers, care providers, and advocates to direct culture change expansion and resource allocation.

Supplemental Material

DuanYinfei_Supplementary_material_2_20_20 – Supplemental material for An Empirical Typology of Nursing Home Culture Change Implementation

Supplemental material, DuanYinfei_Supplementary_material_2_20_20 for An Empirical Typology of Nursing Home Culture Change Implementation by Yinfei Duan, Christine A. Mueller, Fang Yu, Kristine M. Talley and Tetyana P. Shippee in Journal of Applied Gerontology

Footnotes

Acknowledgements

The authors would like to thank all nursing home administrators for participating in the survey, Teresa M. Lewis from the Minnesota Department of Human Services for her assistance in data collection, and Margot Schwartz and Dr. Susan C. Miller from Brown University for their instructions on the use of the culture change assessment tool.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

This study was approved under IRB protocol no. STUDY00003659.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Yinfei Duan

Supplemental Material

Supplemental material for this article is available online.

References

Burack

O. R.

Reinhardt

J. P.

Weiner

A. S.

(2012). Person-centered care and elder choice: A look at implementation and sustainability. Clinical Gerontologist, 35, 390–403. https://doi.org/10.1080/07317115.2012.702649

Castle

N. G.

Ferguson

J. C.

(2010). What is nursing home quality and how is it measured? The Gerontologist, 50(4), 426–442. https://doi.org/10.1093/geront/gnq052

Centers for Medicare & Medicaid Services. (2016). 42 CFR part 483, subpart B. https://www.govinfo.gov/content/pkg/CFR-2018-title42-vol5/xml/CFR-2018-title42-vol5-sec483-24.xml

Chisholm

Zhang

N. J.

Hyer

Pradhan

Unruh

Lin

F.-C.

(2018). Culture change in nursing homes: What is the role of nursing home resources? Inquiry: A Journal of Medical Care Organization, Provision and Financing, 55, 46958018787043. https://dx-doi-org-s.web.bisu.edu.cn/10.1177/0046958018787043

Cohen

L. W.

Zimmerman

Reed

Brown

Bowers

B. J.

Nolet

. . . Collaborative

T. R.

(2016). The Green House model of nursing home care in design and implementation. Health Services Research, 51(Suppl. 1), 352–377. https://doi.org/10.1111/1475-6773.12418

Coleman

M. T.

Looney

O’Brien

Ziegler

Pastorino

C. A.

Turner

(2002). The Eden alternative: Findings after 1 year of implementation. Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 57(7), M422–M427. https://doi.org/10.1093/gerona/57.7.M422

Duan

Mueller

C. A.

Talley

K. M

. (2020). The effects of nursing home culture change on resident quality of life in U.S. nursing homes: An integrative review. Research in Gerontological Nursing. Advance Online Publication. https://doi.org/10.3928/19404921-20200115-02

Gastmans

Milisen

(2006). Use of physical restraint in nursing homes: Clinical-ethical considerations. Journal of Medical Ethics, 32(3), 148–152. https://doi.org/10.1136/jme.2005.012708

Gaugler

J. E.

(2005). Family involvement in residential long-term care: A synthesis and critical review. Aging & Mental Health, 9(2), 105–118. https://doi.org/10.1080/13607860412331310245

10.

Grabowski

D. C.

Elliot

Leitzell

Cohen

L. W.

Zimmerman

(2014). Who are the innovators? Nursing homes implementing culture change. The Gerontologist, 54(Suppl. 1), S65–S75. https://dx-doi-org-s.web.bisu.edu.cn/10.1093/geront/gnt144

11.

Grabowski

D. C.

O’Malley

A. J.

Afendulis

C. C.

Caudry

D. J.

Elliot

Zimmerman

(2014). Culture change and nursing home quality of care. The Gerontologist, 54(Suppl. 1), S35–S45. https://doi.org/10.1093/geront/gnt143

12.

Grant

L. A.

(2008). Culture change in a for-profit nursing home chain: An evaluation. https://www.commonwealthfund.org/publications/fund-reports/2008/feb/culture-change-profit-nursing-home-chain-evaluation

13.

Hanson

L. C.

Reynolds

K. S.

Henderson

Pickard

C. G.

(2005). A quality improvement intervention to increase palliative care in nursing homes. Journal of Palliative Medicine, 8(3), 576–584. https://doi.org/10.1089/jpm.2005.8.576

14.

Hill

N. L.

Kolanowski

A. M.

Milone-Nuzzo

Yevchak

(2011). Culture change models and resident health outcomes in long-term care. Journal of Nursing Scholarship, 43, 30–40. https://doi.org/10.1111/j.1547-5069.2010.01379.x

15.

Kane

R. A.

Kling

K. C.

Bershadsky

Kane

R. L.

Giles

Degenholtz

H. B.

. . . Cutler

L. J.

(2003). Quality of life measures for nursing home residents. Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 58(3), 240–248. https://doi.org/10.1093/gerona/58.3.M240

16.

Kane

R. A.

Lum

T. Y.

Cutler

L. J.

Degenholtz

H. B.

(2007). Resident outcomes in small-house nursing homes: A longitudinal evaluation of the initial green house program. Journal of the American Geriatrics Society, 55, 832–839. https://doi.org/10.1111/j.1532-5415.2007.01169.x

17.

Kehoe

M. A.

Van Heesch

(2003). Culture change in long term care: The Wellspring model. Journal of Social Work in Long-term Care, 2(1–2), 159–173. https://doi.org/10.1300/J181v02n01_11

18.

Koren

M. J.

(2010). Person-centered care for nursing home residents: The culture-change movement. Health Affairs, 29, 312–317. https://doi.org/10.1377/hlthaff.2009.0966

19.

Lepore

M. J.

Shield

R. R.

Looze

Tyler

Mor

Miller

S. C.

(2015). Medicare and Medicaid reimbursement rates for nursing homes motivate select culture change practices but not comprehensive culture change. Journal of Aging & Social Policy, 27(3), 215–231. https://dx-doi-org-s.web.bisu.edu.cn/10.1080/08959420.2015.1022102

20.

Loe

Moore

C. D.

(2012). From nursing home to green house: Changing contexts of elder care in the United States. Journal of Applied Gerontology, 31(6), 755–763. https://doi.org/10.1177/0733464811401022

21.

Magidson

Vermunt

(2002). Latent class models for clustering: A comparison with K-means. Canadian Journal of Marketing Research, 20(1), 36–43. https://pdfs.semanticscholar.org/6add/265688cde63766bed6b920c4546e7c11ab99.pdf

22.

Miller

S. C.

Lepore

Lima

J. C.

Shield

Tyler

D. A.

(2014). Does the introduction of nursing home culture change practices improve quality? Journal of the American Geriatrics Society, 62(9), 1675–1682. https://doi.org/10.1111/jgs.12987

23.

Miller

S. C.

Looze

Shield

Clark

M. A.

Lepore

Tyler

. . . Mor

(2014). Culture change practice in U.S. nursing homes: Prevalence and variation by state Medicaid reimbursement policies. The Gerontologist, 54, 434–445. https://doi.org/10.1093/geront/gnt020

24.

Miller

S. C.

Schwartz

M. L.

Lima

J. C.

Shield

R. R.

Tyler

D. A.

Berridge

C. W.

. . . Clark

M. A.

(2018). The prevalence of culture change practice in US nursing homes. Medical Care, 56(12), 985–993. https://doi.org/10.1097/MLR.0000000000000993

25.

Minnesota Department of Human Services. (2019). Minnesota nursing home report card technical user guide. http://nhreportcard.dhs.mn.gov/technicaluserguide.pdf

26.

Minnesota House Research Department. (2016). Nursing facility reimbursement and regulation. https://www.house.leg.state.mn.us/hrd/pubs/nfreimb.pdf

27.

Mor

Zinn

Angelelli

Teno

J. M.

Miller

S. C.

(2004). Driven to tiers: Socioeconomic and racial disparities in the quality of nursing home care. The Milbank Quarterly, 82(2), 227–256. https://doi.org/10.1111/j.0887-378X.2004.00309.x

28.

Munroe

D. J.

Kaza

P. L.

Howard

(2011). Culture-change training: Nursing facility staff perceptions of culture change. Geriatric Nursing, 32(6), 400–407. https://doi.org/10.1016/j.gerinurse.2011.07.001

29.

Oberski

(2016). Mixture models: Latent profile and latent class analysis. In Robertson

Kaptein

(Eds.), Modern statistical methods for HCI (pp. 275–287). https://doi.org/10.1007/978-3-319-26633-6_12

30.

Poey

J. L.

Hermer

Cornelison

Kaup

M. L.

Drake

Stone

R. I.

Doll

(2017). Does person-centered care improve residents’ satisfaction with nursing home quality? Journal of the American Medical Directors Association, 18(11), 974–979. https://doi.org/10.1016/j.jamda.2017.06.007

31.

Puurveen

Baumbusch

Gandhi

(2018). From family involvement to family inclusion in nursing home settings: A critical interpretive synthesis. Journal of Family Nursing, 24(1), 60–85. https://doi.org/10.1177/1074840718754314

32.

Rahman

A. N.

Schnelle

J. F.

(2008). The nursing home culture-change movement: Recent past, present, and future directions for research. The Gerontologist, 48, 142–148. https://doi.org/10.1093/geront/48.2.142

33.

Ransom

(2000). Eden alternative: The Texas project. https://digital.library.txstate.edu/handle/10877/4087

34.

Robinson

S. B.

Rosher

R. B.

(2006). Tangling with the barriers to culture change—Creating a resident-centered nursing home environment. Journal of Gerontological Nursing, 32(10), 19–25. https://doi.org/10.3928/00989134-20061001-04

35.

Schwartz

M. L.

Lima

J. C.

Clark

M. A.

Miller

S. C.

(2019). End-of-life culture change practices in U.S. nursing homes in 2016/2017. Journal of Pain and Symptom Management, 57(3), 525–534. https://doi.org/10.1016/j.jpainsymman.2018.12.330

36.

Shield

Looze

Tyler

Lepore

Miller

S. C.

(2014). Why and how do nursing homes implement culture change practices? Insights from qualitative interviews in a mixed methods study. Journal of Applied Gerontology, 33(6), 737–763. https://dx-doi-org-s.web.bisu.edu.cn/10.1177/0733464813491141

37.

Shield

Tyler

Berridge

Clark

Miller

S. C.

(2018). Innovative use of cognitive interviewing for nursing home research. Journal of the American Medical Directors Association, 19(12), 1133–1135. https://doi.org/10.1016/j.jamda.2018.05.004

38.

Shier

Khodyakov

Cohen

L. W.

Zimmerman

Saliba

(2014). What does the evidence really say about culture change in nursing homes? The Gerontologist, 54, S6–S16. https://doi.org/10.1093/geront/gnt147

39.

Shippee

T. P.

Henning-Smith

Gaugler

J. E.

Held

Kane

R. L.

(2017). Family satisfaction with nursing home care: The role of facility characteristics and resident quality-of-life scores. Research on Aging, 39(3), 418–442. https://doi.org/10.1177/0164027515615182

40.

Shippee

T. P.

Henning-Smith

Kane

R. L.

Lewis

(2015). Resident- and facility-level predictors of quality of life in long-term care. The Gerontologist, 55, 643–655. https://doi.org/10.1093/geront/gnt148

41.

StataCorp LLC. (2017). Stata statistical software: Release 15 (Version 15.0).

42.

Sterns

Miller

S. C.

Allen

(2010). The complexity of implementing culture change practices in nursing homes. Journal of the American Medical Directors Association, 11(7), 511–518. https://dx-doi-org-s.web.bisu.edu.cn/10.1016/j.jamda.2009.11.002

43.

Sturdevant

D. L.

Mueller

C. A.

Buckwalter

K. C.

(2018). Measurement of nursing home culture change: Systematic review. Research in Gerontological Nursing, 11(2), 103–112. https://doi.org/10.3928/19404921-20171205-01

44.

Sullivan

J. L.

Shwartz

Burgess

J. F.

Jr. Peköz

E. A.

Christiansen

C. L.

Gerena-Melia

Berlowitz

(2013). Person-centered care practices and quality in Department of Veterans Affairs nursing homes: Is there a relationship? Medical Care, 51, 165–171. https://doi.org/10.1097/MLR.0b013e3182763230

45.

Tyler

D. A.

Shield

R. R.

Rosenthal

Miller

S. C.

Wetle

Clark

M. A.

(2011). How valid are the responses to nursing home survey questions? Some issues and concerns. The Gerontologist, 51, 201–211. https://doi.org/10.1093/geront/gnq095

46.

White-Chu

E. F.

Graves

W. J.

Godfrey

S. M.

Bonner

Sloane

(2009). Beyond the medical model: The culture change revolution in long-term care. Journal of the American Medical Directors Association, 10(6), 370–378. https://doi.org/10.1016/j.jamda.2009.04.004

47.

Winch

R. F.

(1947). Heuristic and empirical typologies: A job for factor analysis. American Sociological Review, 12(1), 68–75. https://doi.org/10.2307/2086492

48.

Kane

R. L.

Shamliyan

T. A.

(2013). Effect of nursing home characteristics on residents’ quality of life: A systematic review. Archives of Gerontology and Geriatrics, 57, 127–142. https://doi.org/10.1016/j.archger.2013.03.015

49.

Zimmerman

Cohen

L. W.

Reed

Gwyther

L. P.

Washington

Cagle

J. G.

. . . Preisser

J. S.

(2013). Families matter in long-term care: Results of a group-randomized trial. Seniors Housing & Care Journal, 21(1), 3–20. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4167776/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB