| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a School of Public Health, Curtin University of Technology, Perth, Western Australia, Australia
b Freemasons Center for Research into Aged Care Services, Division of Health Sciences, Curtin University of Technology, Perth, Western Australia, Australia
Correspondence: Duncan P. Boldy, PhD, Department of Health Policy and Management, School of Public Health, Curtin University of Technology, GPO Box U 1987, Perth 6845, Western Australia, Australia. E-mail: dboldy{at}health.curtin.edu.au.
Decision Editor: Laurence G. Branch, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Long-term care • Quality • Staff care • Modeling
Quality improvement has gradually become recognized as an important approach to ensure service quality in residential aged care settings. Initially, quality improvement programs were implemented from the provider's point of view. However, factors assumed by service providers or policymakers to ensure quality of care may have little bearing on what residents perceive as important to their quality of life (Ryden et al. 2000
; Schmitt 2000; Sikorska 1999). In a broad sense, what we are dealing with are issues of person–environment fit. Readers wishing to place this article in a wider context that is beyond the scope of this article may find value in consulting key theoreticians in this area (e.g., Kahana 1974; Lawton 1982, Lawton 1983, Lawton 1989).
Recently, the importance of residents' views has gained recognition related to improving service quality (Bartlett 1993; Boldy and Grenade 2001). The process of incorporating the views of residents is important because only as a result of direct feedback can staff appropriately adjust their behaviors and actions (Schlesinger and Zornitsky 1991) to satisfy residents' needs. Undertaking a resident satisfaction survey has become widely accepted as an important approach, relevant to ensuring that residents receive the care they want and that such care is designed and provided as they would wish. Resident satisfaction surveys have also been recognized as one approach to empowering consumers (Bartlett 1993; Boldy and Grenade 2001; van Geen 1997).
Previous research has supported a positive relationship between service quality and satisfaction with health services (Duffy and Ketchand 1998; Taylor and Cronin 1994). It is assumed that higher quality of service (better service input, process, and output) leads to a higher level of resident satisfaction (better outcome) and contributes to improved life satisfaction. It is similarly assumed that if residents have a high level of satisfaction, then high quality of service is being provided. Hence, a thorough understanding of resident satisfaction can provide important insights for improving service quality and in terms of promoting residents' quality of life. This is particularly pertinent in residential aged care settings, where residents are vulnerable to poor quality of care. Through the process of assessing resident satisfaction and examining the relationships among satisfaction components, important aspects of service can be identified from the residents' perspective. Such information can be taken into account to improve service, targeting important aspects first.
In an earlier article, we assessed the factor structure, reliability, and validity of an existing Resident Satisfaction Questionnaire (RSQ) and developed a short-form RSQ for regular use in residential aged care settings (Chou, Boldy, and Lee 2001). Resident satisfaction was found to be a multidimensional construct composed of six factors—Room, Home, Social Interaction, Meals Service, Staff Care, and Resident Involvement. The purpose of this article is to assess the direction and magnitude of the relationships between various resident satisfaction components in residential aged care services in Australia.
In addition to nursing home care, Australia also provides hostel care. Most hostels broadly fit the definition of congregate care in the United States, that is, they are typically for frail yet semi-independent residents who are able to maintain a private apartment, with the flexibility of opting for different levels or amounts of group meals, housekeeping, and other support services provided through on-site management or service coordination. They normally stop short of assisted living, as defined by aides on-site or other more specialized support.
Given the different resident characteristics and care needs, staffing patterns, and staff mix, we anticipated that nursing home and hostel residents might have different patterns of relationships between satisfaction components. For example, the effect of social interaction on other aspects of satisfaction may vary between nursing home and hostel residents because nursing home residents are more physically dependent. In addition, nursing homes employ more qualified staff (e.g., registered nurses) who receive more formal training. In contrast, hostels employ more support staff who generally receive less formal training. In this study, as expected, more registered nurses (25% vs 2%) were employed to deliver care in nursing homes than hostels. Consequently, we addressed two research questions in this study.
This article focuses on the relationships among resident satisfaction components. As such, the effects on resident satisfaction of external facility factors, resident characteristics, staff factors, and so forth are the subject of a subsequent investigation, which will be reported elsewhere.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
, we used a cross-sectional survey design to collect the required data. The study sample of 1,146 residents was recruited from a variety of residential aged care facilities in Western Australia between April 1998 and April 1999.
Resident Satisfaction Questionnaire
Resident satisfaction was assessed through the self-completed RSQ (Chou et al. 2001). Twenty-four items covering six aspects (factors) of resident satisfaction (i.e., room, home, social interaction, meals service, staff care, and resident involvement) were used for this study; see Table 3 , later in this article. Each item consisted of either a 3-point (no = 1, depends = 2, yes = 3) or 4-point (poor = 1, fair = 2, good = 3, excellent = 4) response. Satisfaction scores for each factor were obtained by proportionally weighted factor score regressions to combine individual items. A higher score would indicate a greater satisfaction with respect to that factor.
|
; Joreskog and Sorbom 1996). Because the six satisfaction scores were slightly skewed with nonzero kurtosis, a weighted least-squares procedure was considered appropriate for parameter estimation.
Resident satisfaction components are latent variables that cannot be measured precisely. The SEM approach was deemed appropriate for this study because it permits the assessment of measurement properties of the satisfaction constructs. SEM allows for the specification and simultaneous estimation of relationships among multiple dependent and independent latent variables and accounts for measurement errors in the latent variables (Bollen and Long 1993). On the basis of SEM, the researcher can hypothesize complex models of satisfaction and can test or confirm models statistically.
Measurement Model
A one-factor congeneric measurement model was first fitted to each of the six factors to assess construct validity and reliability, using the LISREL package (Joreskog and Sorbom 1999). A fitted congeneric model allows large numbers of observed variables to be reduced to a single composite scale and subsequently reduces the number of variables to be included in the SEM (Holmes-Smith and Rowe 1994). To keep the measurement model simple, we aimed to keep 3–5 observed variables (items) per factor. The composite scale reliability, composite factor loadings, and measurement error variances were then estimated for each latent factor and fixed in the measurement part of the structural model (Holmes-Smith and Rowe 1994; Rowe and Rowe 1999).
Structural Model
We next investigated the relationships among the six key resident satisfaction components, using the structural part of the SEM. The parameters to be estimated were the regression coefficients. A conceptual model was formulated, representing the expected relationships among the six identified satisfaction components: room (
1), home (2), social interaction (3), meals service (4), staff care (5), and resident involvement (6). The hypothesized relationships, graphically presented in Fig. 1, have been identified partially from the literature but mainly from discussions with residents, relevant researchers, and stakeholders.
|
are only partially appropriate for certain relationships in our settings, and there is a lack of literature investigating the relationships among satisfaction components. Hence, we developed a logic/stakeholder views–based model for testing purposes. The (alternative) hypotheses follow.
Some of these relationships may need to be modified in view of the model-fitting results. Apart from theoretical and practical considerations, it is recommended to assess model adequacy on the basis of the following goodness-of-fit indices and criteria: normed chi-square (
2/df) less than 3, root-mean-square error of approximation (RMSEA) less than 0.05, nonnormed fit index (NNFI) greater than 0.90, comparative fit index (CFI) greater than 0.90, goodness-of-fit index (GFI) greater than 0.90, and adjusted goodness-of-fit index (AGFI) greater than 0.90. More details can be found in Byrne 1998, Joreskog and Sorbom 1996, Kline 1998, and Maruyama 1998.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
.
|
). As expected, the vast majority of nursing home residents (95%) were classified under RCS Categories 1–4 (high care), and 96% of hostel residents were classified as low care (RCS 5–8). This indicates a clear distinction between nursing home and hostel residents in terms of dependency.
Measurement Properties
Standardized factor loadings of RSQ items are given in Table 3 . Each of the six one-factor congeneric measurement models fitted satisfactorily, based on the goodness-of-fit statistics of each factor (see Table 3 ). The corresponding maximized composite reliabilities are also given in Table 3 . The high composite reliabilities for the resident satisfaction constructs suggested that all observed variables (items) were reliable measures of the underlying latent construct. Moreover, all observed variables had good item reliability (above 0.5) and were significant (p < .01), implying a high level of convergent validity.
Resident Satisfaction Levels
The sample of residents was split into hostel (n = 752) and nursing home (n = 394) groups. Table 4 presents the means and standard deviations of the six resident satisfaction composite variables by facility type and shows that nursing home residents had lower scores on all aspects of satisfaction than hostel residents. A series of independent sample t tests confirmed such differences as significant, with p < .01 in each case. Consequently, examining whether the structural regression paths vary according to high-care facilities (nursing homes) and low-care facilities (hostels) is justified.
|
Relationships Among Satisfaction Components for All Residents (Combined Model)
Tests of the hypothesized structural model for all residents revealed poor model fit (see Table 5 ). The results showed that two paths, ß16 (resident involvement 
room) and ß26 (resident involvement home), were not statistically significant (p > .05), indicating that H2 was only partially supported by the data. All other paths were significant (p < .001), thus confirming alternative hypotheses H1, H3, H4, and H5. A post hoc analysis suggested the addition of three structural paths to the model (room social interaction, ß31; home social interaction, ß32; social interaction meals service, ß43). The model was respecified and reestimated. Table 5 shows that the fit of the revised combined model is satisfactory.
|
).
The result indicates a significant difference between the constrained model and the unconstrained model fits, 
2(12, N = 1146) = 22.198, p = .035; see Table 5 . One may therefore conclude that the structural regression paths did vary between the two groups (H6). This provided support for developing separate models for hostel and nursing home groups.
Hostel Model
Using the combined model to fit hostel data, the results suggested that the hostel model was slightly overfitted (see Table 5 ). All paths were significant (p < .01) except ß65 (ß = 0.103, t = 2.259). Thus, no modification is required for the hostel sample. The combined model appears to be hostel dominated. This may be because a large proportion of residents were from a hostel (66%).
The fitted hostel resident satisfaction model, presented in Fig. 2, illustrates the fundamental relationships among the six satisfaction domains. To keep the diagram simple, only the structure part of the SEM is shown. A summary of the direct, indirect, and total effect sizes for the final hostel model is given in Table 6 . The total effect size of staff care on all other aspects of resident satisfaction ranged from 0.327 to 0.548. This suggests that satisfaction with staff care plays an important role in determining overall resident satisfaction.
|
|
social interaction) and ß43 (social interaction meals service), were not significant. A post hoc analysis suggested an additional structural path ß14 (meals service room). The model was subsequently respecified and reestimated. Table 5 shows that the fit of the revised nursing home model is satisfactory. The nursing home model is schematically presented in Fig. 3. Table 7 provides the direct, indirect, and total effect sizes derived from the SEM. Similar to hostels, satisfaction with staff care has a moderate and positive effect on all other components of resident satisfaction (total effect sizes ranging from 0.360 to 0.484).
|
|
social interaction, social interaction meals service) and the new path (meals service room) in the nursing home model. It appears that in nursing homes (high-care residents), aspects of the meals service are important and can exert a direct positive impact on satisfaction with room (ß = 0.254) and an indirect positive effect on satisfaction with home (ß = 0.130; see Fig. 3 and Table 7 ). In terms of physical environment, both room (ß = 0.291) and home (ß = 0.293) had direct influences on social interaction in the hostel model. Unlike the hostel model, only home (ß = 0.429) had a direct impact on social interaction in the nursing home model. Resident involvement and social interaction were linked in both nursing home and hostel models, but the direct relationship between social interaction and meals service no longer existed in nursing homes.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Relationships Among Satisfaction Components
Staff Care
Residents' satisfaction with staff care plays a central role in determining other aspects of resident satisfaction (see Table 6 and Table 7 ). In general, the total effects on other aspects of resident satisfaction were between 0.327 and 0.548, which represent moderate to large effect sizes. The staff care component itself, however, was not influenced by any other aspect of satisfaction. The importance of staff care in residential aged care is a common theme that has frequently emerged from studies using a qualitative approach. For example, Rantz and colleagues 1999(p. 31) stressed that "the bottom line from consumers is that without good staff nothing else is possible," and residents are clear that their primary concerns are staff and care.
Schneider 1991 stressed that attempting to maximize service quality on all dimensions is extraordinarily difficult and that relevant excellence in service represents a more meaningful and appropriate goal than generically excellent service. Organizational resources and staff efforts should therefore be directed toward those dimensions that are most important to residents. Staff should be recruited, trained, appraised, and rewarded for performing critical tasks in ways that correspond to customer values (resident needs; Ulrich, Halbrook, Meder, Stuchlik, and Thorpe 1991).
Room and Home
Residents' satisfaction with the home is mainly influenced by their satisfaction with their room and with staff care, and their satisfaction with their room is mainly influenced by their satisfaction with staff care, for both nursing home and hostel residents. It is easy to understand the importance of residents' rooms as their personal space, whereas the home contains public areas that they need to share with other residents. If residents are satisfied within their own boundaries and live comfortably, they are more likely to be satisfied with the overall home environment.
These results also suggest that if the physical environment is less than satisfactory, perhaps staff care could compensate in some way. It also reinforces the importance of staff care. If staff care is insufficient, then no matter how well designed and equipped a facility is, residents are likely to live under the staff's shadow. Staff represent the "soft" environment that makes up the atmosphere and spirit (soul) of the facility, whereas the physical environment, such as room and dining room, represent the relatively less important "hard" environment. This parallels the work of Goffman 1961 related to mental patients.
Social Interaction
Residents' satisfaction with social interaction was mainly influenced by their satisfaction with the physical environment and with staff care. The results suggest that satisfaction with room and home also encourage social interaction and that staff might act as a catalyst in enlivening the social process. The findings are consistent with the view of Hugman 2000 concerning the importance ascribed to personal space for managing relationships. In this regard, residents' rooms are their personal space, and within this boundary they have control and can use such space to meet friends and to enjoy activities with other residents. Moos and Igra 1980 also noted that physical and architectural features, organizational policies and programs, and resident and staff characteristics can influence the social environment that emerges in a particular setting.
Meals Service
Satisfaction with the meals service was influenced by satisfaction with the physical environment, social interaction, staff care, and resident involvement, either directly or indirectly. It is not difficult to explain why physical environment and staff care have a positive impact on residents' satisfaction with meals service. Similar to restaurant service, pleasantness of the physical environment and staff create an atmosphere that increases residents' appetites and produces an emotional reaction receptive to enjoying a meal. Regarding social interaction, the results indicate that if residents eat with their close friends they are likely to enjoy their meals. Moreover, if residents are involved in the decision-making process, such as what, how, and when they prefer to eat, they are likely to be more satisfied with their meals service.
Resident Involvement
Satisfaction with resident involvement was directly affected by satisfaction with social interaction and staff care. This suggests that residents who are unhappy about their social interaction are more likely to suffer from social isolation and be less comfortable with expressing any concerns to the staff.
Moos and Lemke 1980 found that residents who were in friendly, cohesive, and well-organized facilities oriented toward independence and resident autonomy tended to show higher morale and to participate in self-initiated activities. Their findings assumed that greater resident involvement leads to more satisfied social interaction. The relationship between the two variables appears to be quite complicated and reciprocal.
Variation by Facility Type
As a group, hostel residents generally expressed higher levels of satisfaction on all aspects than did nursing home residents (Table 4 ). In addition, the relationships among resident satisfaction components varied according to facility type, especially between room, home, social interaction, and meals service.
Physical Environment and Social Interaction
The influence of physical structure and amenities is quite different between the two types of facility. Physical comfort and social interaction appear to be more important for residents who are relatively independent. The findings also suggest that physical comfort can foster social interaction and increase satisfaction with meals service in a hostel. Because hostel residents are more independent, they can move around and do things in different areas. In contrast, high-dependent residents in nursing homes are more likely to be constrained to or isolated in a certain place (such as their room or in a lounge) and spend most of their time within a fixed area.
Furthermore, residents in different types of facility may have access to different environmental resources. For example, the physical environment, such as a resident's room, lounge area, and dining room, provides space and places (environmental diversity) and the opportunity for residents to socialize with each other, to do things, and to enjoy meals together. Taking residents' room arrangement as an indication of personal space, only 1% of hostel residents in this study were required to share both room and bathroom with others, whereas this was true for more than 50% of nursing home residents. A high proportion of nursing home residents in multiple bedrooms is almost certainly an indication of lack of privacy and personal space. Such an arrangement also predefines the boundary and richness of their social world and might limit their opportunities to socialize with others in the way they want. In contrast, hostel residents can freely use their rooms (a symbol of personal space) for their social activities.
These findings partially support the comment made by Kruzich, Clinton, and Kelber 1992(p. 348) that "varying levels of functional limitations may influence how residents interact with the environments." Moos and Lemke 1996 also noted that some environmental factors seem to act as a stimulus for residents who are relatively intact or only moderately impaired to engage in self-initiated activities in the facility.
Meals Service and Social Interaction
The results suggest that meals become the main part of dependent residents' lives, because food is one of the most important basic needs. Hence, satisfaction with the meals service, directly or indirectly, has a positive effect on all other aspects of satisfaction except staff care.
The absence of an effect of social interaction on meals service may be because high-care residents are more difficult to engage in social activities and maintain friendships, and thus they are less likely to enjoy meals with friends. The small indirect effect of meals service on social interaction in nursing homes suggests that more dependent residents are likely to be located in a fixed area and use meal times to maintain a certain degree of interaction with others, such as staff assistance with their meals. Hence, they might consider the meals service as a way of satisfying their basic need and of getting limited contact rather than as a social activity whereby they enjoy meals with others. The results also support the view that food can be seen as a symbol of security and a medium for socialization (Chien, Stotsky, and Cole 1973; Linn, Gurel, and Linn 1977).
Conclusion
Importantly, the results indicate that by enhancing residents' satisfaction with staff care, all aspects of resident satisfaction can be increased, for both high- and low-care residents. Such results support the findings in the literature that good staff care is important to residents and has a profound effect on them (Bliesmer 1993; Rantz et al. 1999). It is therefore essential that a facility not only be designed with a good physical environment but also provide excellent staff care.
To improve resident satisfaction, policymakers, aged care providers, management, and staff need to understand the driving forces of resident satisfaction and focus on those aspects that are most likely to effect improvement. As indicated in this study, it is important that residents be "comforted and cared for by staff who value the elderly" (Bliesmer 1993, p. 33). Efforts should be directed toward promoting resident satisfaction by enhancing their satisfaction with staff care, for example, by recruiting the right person to do the right job.
For management or service providers in residential aged care settings, evaluating the staff recruitment process, assessing staff members' attitudes toward residents, evaluating staff practice, and monitoring both staff and resident satisfaction may be useful to identify areas of concern and to ensure that appropriate people are employed to care for vulnerable residents. Guidelines for staff selection and expected professional behavior might also be developed to enable greater congruence between staff and service requirements. Moreover, staff should be informed and educated about the important role that they play and the significant impact that they have on residents' quality of life.
The Forum
Book Reviews
Practice Concepts
|
Acknowledgments |
|---|
Received for publication May 30, 2001. Accepted for publication October 31, 2001.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Street, S. Burge, J. Quadagno, and A. Barrett The Salience of Social Relationships for Resident Well-Being in Assisted Living J. Gerontol. B. Psychol. Sci. Soc. Sci., March 1, 2007; 62(2): S129 - S134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-C. Chou, D. P. Boldy, and A. H. Lee Factors Influencing Residents' Satisfaction in Residential Aged Care Gerontologist, August 1, 2003; 43(4): 459 - 472. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
