Family Satisfaction: Differences Between Nursing Homes and Residential Care Facilities

Abstract

This study compared differences in overall family satisfaction, specific satisfaction domains, and correlates of satisfaction between nursing homes (NHs) and residential care facilities (RCFs), using data from the 2016 Ohio Long-Term Care Family Satisfaction Survey. Satisfaction was higher for RCFs overall and within nearly every domain, with the largest difference observed in the environment domain. In both facility types, higher satisfaction was associated with male respondents, older respondent age, White race, less-frequent visitation, longer anticipated length of stay, less help provided during visits, smaller facilities, lower Medicaid-reliant resident percentage, and nonprofit status. Resident age, visitation frequency, perceived assistance required, and kinship tie were differentially related to satisfaction between facility types. NH administrators should focus on the environment and the moving in process. All administrators should address how residents spend time and should be aware that residents’ and their family members’ characteristics may affect satisfaction levels.

Keywords

quality of life long-term care individual characteristics family satisfaction

Introduction

As state and government organizations put more emphasis on improving the quality of long-term care (LTC) facilities, data about quality markers in nursing homes (NHs) and residential care facilities (RCFs) are becoming increasingly important. Consumers (residents and their families) use data about LTC facilities to choose which facility will offer the best care and fit their needs. For example, Medicare.gov provides users an online Nursing Home Compare tool that combines information from Centers for Medicare and Medicaid Services health inspections, the Minimum Data Set, and Medicare claims data (Centers for Medicare and Medicaid Services, 2019). Others have noted the importance of considering factors like resident and family satisfaction when assessing LTC facilities (A. Williams et al., 2016). To partially fill that need, the Ohio Long-term Care Consumer Guide was created, which provides users with a searchable tool that reports both resident and family satisfaction scores for facilities (Ohio Department of Aging, 2018; Straker et al., 2007). Each of these approaches use data to provide a view of facility quality for families and residents to make informed choices.

Data related to facility quality ratings and resident satisfaction have been available for some time, but including family satisfaction is a newer quality marker (A. Williams et al., 2016). Resident and family satisfaction represents related but different aspects of facility quality (R. L. Kane et al., 2005; Shippee et al., 2017). The goal of this study was to compare differences in overall family satisfaction, specific domains of satisfaction, and correlates of satisfaction between two different types of LTCs: NHs and RCFs; also known as assisted living). Satisfaction levels of Ohio RCF residents’ family members were assessed for the first time in 2016 (Straker et al., 2017). The measure used for that study was nearly identical to the existing NH measure, thus providing the opportunity for a direct comparison of family satisfaction between these facility types.

NHs and RCFs have important differences, but key similarities. The primary difference between the two is the additional medical care provided in NHs (National Institute on Aging, 2017). Other important differences include payment source, size, and regulation (Lawriter, 2018; Spillman et al., 2002). In Ohio, an NH is licensed by the Ohio Department of Health (ODH, 2019a) to provide personal care and skilled nursing services. An RCF is licensed by ODH (2019b) to provide accommodations, supervision, and personal care services. RCFs may also provide limited skilled nursing care, up to 120 days in a 12-month period (ODH, 2019b). As these definitions illustrate, there are differences between NHs and RCFs, but also much overlap.

Background

In research about the quality of NHs and RCFs, satisfaction with domains of life in LTC (e.g., meaningful activities, privacy, and relationships) has frequently been used interchangeably with the term quality of life (QoL; R. A. Kane et al., 2003; Roberts & Ishler, 2018; Xu et al., 2013). Research has established a link between satisfaction/QoL and many institutional, individual, and family-related factors in different types of LTC facilities. This brief review will identify some of the factors that are related to either resident or family satisfaction in NHs and RCFs.

NHs

In NH-related research, the connection between facility characteristics and residents’ QoL has been well established, but with mixed results (see Xu et al., 2013, for a review). Studies have reported relationships between resident QoL and percentage of private rooms, facility size, location, and staffing issues (Shippee et al., 2020; Xu et al., 2013). Results related to NH resident characteristics and QoL are also mixed. Older resident age and female gender are associated with higher QoL on some domains but not others (Shippee et al., 2015). More consistent age-related results have been found in other research, however, with a positive association identified between older age and greater satisfaction (Chou et al., 2003). The effect of length of stay on QoL has revealed both positive and negative associations, depending on the domain (Shippee et al., 2015). Worse physical and mental health among NH residents is related to lower QoL (Shippee et al., 2015).

Family characteristics may also affect both resident and family satisfaction in the NH context. Research has found that residents’ gender is not related to family satisfaction, but male family members reported higher satisfaction (Roberts & Ishler, 2018). In contrast, others have found that a family member’s gender plays no significant role (Shippee et al., 2017). Family satisfaction is also related to frequency of visitation and the amount of help with activities the relative provides while visiting. More frequent visitation was related to lower family satisfaction, as was providing more care during visits (Roberts & Ishler, 2018).

In terms of race and NH family satisfaction, family members who identified as White were more satisfied with their relative’s care than those who identified as Black (Griffore et al., 2011). Among White respondents, proxy-reported resident satisfaction was negatively associated with health problems and the number of activities of daily living (ADLs) the resident could perform. These factors were not significantly related among Black respondents (Griffore et al., 2011). In another study, residents in NHs with higher concentrations of minority residents experienced lower QoL (Campbell et al., 2016; Shippee et al., 2018).

RCFs

Satisfaction with RCFs are also affected by facility characteristics. Staff satisfaction and relationships with care recipients are among the most important factors related to resident satisfaction (Chou et al., 2003; Kemp et al., 2010; Sikorska, 1999). Facility size, amenities, recreational activities, ownership status, and location also influence resident satisfaction (Chou et al., 2003; Sikorska, 1999). However, even when studies agree on which factors affect satisfaction, their direction of influence is not always the same. Chou et al. (2003), for example, found that larger facility size had a positive impact on social interaction and a negative impact on satisfaction with involvement, while Sikorska (1999) found higher overall resident satisfaction with smaller facilities.

Resident and family characteristics also play a role in RCF satisfaction. Among residents, better functional status, family involvement, fewer health conditions, and better health are related to higher resident satisfaction (Abrahamson et al., 2013; Curtis et al., 2005; Dobbs & Montgomery, 2005; Kemp et al., 2013; Mitchell & Kemp, 2000). Female residents, those with less education, and unmarried residents have reported higher satisfaction within some satisfaction domains (Curtis et al., 2005). RCF residents’ ability to manage instrumental ADLs is also related to higher satisfaction (Abrahamson et al., 2013), as is having at least some family contact (Mitchell & Kemp, 2000).

RCF family satisfaction research suggests that older age of the family member and reporting by a more distant relative is related to greater satisfaction (Dobbs & Montgomery, 2005). Care collaborations between the resident, staff, and resident’s family are also key factors in resident and family satisfaction (Kemp et al., 2013). Family member’s race is not related to overall differences in satisfaction; however, Black relatives reported higher satisfaction when the resident had more impairments, while White relatives reported the opposite (S. W. Williams et al., 2008). A positive social atmosphere was also related to higher satisfaction, at least among White family members (S. W. Williams et al., 2008).

NH and RCF comparisons

The few studies that have compared differences between facility types have focused on factors like family and staff relationships (Zimmerman et al., 2013), end-of-life care (Sloane et al., 2003), family involvement and caregiver health (Port et al., 2005), and outcomes among residents with dementia (Sloane et al., 2005). These studies have identified some differences, but have generally concluded that facility type played only a small role in their outcomes of interest. This study was intended to expand upon this previous research, and to provide some clarity about factors that influence satisfaction, given the mixed results described earlier.

Theory and Hypotheses

The person-in-environment (PIE) perspective (see e.g., Cvitkovich & Wister, 2001; Karls, 2002) provided a useful framework to consider whether and why satisfaction differences may exist between NHs and RCFs. The variables and domains of satisfaction in this study map on to PIE, as the framework considers the individual (relative/resident) within a larger context (NH or RCF) of factors within micro (e.g., moving in or spending time), mezzo (e.g., meals and dining, caregivers, care and services), and macro levels (e.g., environment and facility culture; Green & McDermott, 2010). Some of the key facility differences noted earlier suggest that family satisfaction within the various levels (and overall) may differ between the two settings. For example, on a mezzo level, families have expectations regarding the types of care and services the resident will receive. Due to greater medical needs in NHs compared to RCFs, however, families may report lower satisfaction. Using the PIE model, this study can speak to the multiple levels that make up NHs and RCFs while exploring patterns in family satisfaction (Karls & O’Keefe, 2009).

Our hypotheses were informed by previous research, the PIE theory, and common differences between NHs and RCFs. Specifically, we hypothesized that family satisfaction would be higher for RCFs than NHs overall, based on the PIE perspective and common differences between these facilities (e.g., payment source). In addition, the factors that predict satisfaction may work differently between facility types. For example, payment source differences (higher private pay percentage) could result in RCFs considering residents/families from a consumer perspective more than do NHs, which could increase family satisfaction more in that setting. Based on previous research, we hypothesized that larger facility size and higher percentage of Medicaid as primary payment sources would be related to lower satisfaction in both facility types, but more so among RCFs. We also used previous research to posit that female family members and higher visitation frequency would be related to lower satisfaction in NHs but not RCFs. Finally, we hypothesized that shorter anticipated length of stay would also be related to higher satisfaction in NHs only, given the larger role NHs play in postacute care.

Methods

Data

Data from the 2016 Ohio Long-Term Care Family Satisfaction Survey were analyzed. All procedures were reviewed and approved by the Miami University Institutional Review Board. The sampling and survey administration process is described in detail elsewhere (Straker et al., 2017), but the final sample included questionnaire results from family members of residents in 968 NHs and 668 RCFs in Ohio. The questionnaire, distributed by mail with an online option was sent to the most involved family member and was based on a random sample of residents in larger facilities and the population of smaller ones. Almost all NHs (98%) and 93% of RCFs in Ohio participated in the survey, with average family member response rates of 40% for NHs and 53% for RCFs (Straker et al., 2017). Several facility-level variables were also merged with the data set from the 2015 Biennial Survey of Long-Term Care Facilities (Scripps Gerontology Center, 2020).

Sample Size and Sampling Selection Process

The combined data set included 33,504 responses from 1,605 NHs and RCFs. Around 98.5% of individuals successfully merged with the facilities. Only respondents age 18+ with an NH/RCF relative who was age 65+ were included, for a sample size of 29,347. Missing values amounted to 10,870 cases. After excluding cases with missing values and outliers (1,001), our final analytical sample included 17,476, with 9,644 (55.2%) in the NH sample and 7,832 (44.8%) RCF respondents.

Statistical Approach

Independent samples t-tests were used to compare overall satisfaction and seven satisfaction domains between facility types. Two multiple regression analyses were used to investigate determinants of overall satisfaction scores separately for NHs and RCFs, using the same independent variables. T-tests were then used to assess differences in each of the regression coefficients calculated in the separate models. Analyses were performed using STATA 15 software (StataCorp, 2017). Facilities were defined as primary sampling units (PSUs) to adjust for the clustering effect on standard errors.

Handling missing values in the multiple regression analyses

To address missing data, we used multiple imputation techniques separately for the NH and RCF samples, using the same independent variables. Forty imputations were used to impute missing values for four variables that made up 75% of the total missing values. These variables were overall family satisfaction (missing = 12.3%), kinship tie between residents and respondents (13%), amount of family member help provided (32.3%) and payment source (17.4%). The regression model did not converge for the kinship tie variable, suggesting that we could not reasonably impute missing values for this variable. However, the regression models successfully converged for the remaining three variables. This increased the sample size for the regression analyses to 16,196 and 10,330 for NH and RCF samples, respectively (total = 26,526).

Measures

Dependent variable and its indicators

Family satisfaction was constructed using seven domains that assessed moving in, spending time, care & services, caregiving, environment, family culture, and meal and dining services. The NH questionnaire included 32 total satisfaction-related items, and the RCF questionnaire included 33 satisfaction items. The 30 items common to both questionnaires were included in the totals. For more information on items that comprise the domains see Straker et al. (2017). A 4-point scale was used for all satisfaction items, including the following labels: Definitely No (coded 1), Probably No (2), Probably Yes (3), Definitely Yes (4). Response scores were summed across items, and within domains for various aspects of satisfaction. Scores were re-scaled such that they ranged from 25 to 100.

Independent variables

Independent variables included resident gender and age, respondent gender, age, race, education level, and kinship tie between resident and respondent. Racial groups with few respondents were combined into a single category (Other). Anticipated length of resident stay (less than 1 month, from 1 to 3 months, greater than 3 months) and respondent visitation frequency (daily, several times a week, once a week, two or three times a month, once a month, few times a year) were also included. Amount of ADL help required included items about dressing, going to the bathroom, and transferring. Response options were needs no assistance or supervision from another person (coded as 1), needs some assistance . . . (2), needs a great deal of assistance . . . (3), and resident is totally dependent (4). These items were summed, where a larger number represented more help. Amount of family member help provided during visits included items about eating, dressing, toileting, grooming, and going to activities. Each used a 3-point scale (never, sometimes, and always) that was summed to create a measure ranging from 5 to 15 (more help). Number of ODH-licensed beds was used to classify facility size into small, medium, and large. Payment source was the percentage of residents using Medicaid as their primary source. Ownership was dichotomized as for-profit and not-for-profit/government. Location was rural or urban.

Results

Descriptive Statistics

Sample characteristics are shown in Table 1. Around three-quarters of the residents were female, as were around two thirds of family respondents. The largest age group of residents was 85 to 94, with 46% and 59% of NH and RCF residents, respectively. RCF respondents age 55 to 64 comprised the largest group (about 44%), and there was a nearly even split between the two largest NH age groups (55–64: 36%, 65–74: 35%). A majority of respondents identified their race as White (NH: 94%, RCF: 98%). Among NH respondents, around half (51%) held a high school diploma or less, compared to about 33% among RCF relatives. Visiting several times a week was the most common choice for both samples. Most respondents (NH: 93%, RCF: 99%) expected the resident to require a stay of longer than 3 months, and most were residents’ adult children (NH: 59%, RCF: 75%). Equal proportions of facilities were grouped as small, medium, and large. Approximately 72% and 70% of NHs and RCFs were for-profit, respectively. Most NHs (77%) and RCFs (83%) were located in rural areas.

Table 1.

Frequency and Percentage of Selected Characteristics of the Samples.

Characteristics	NHs frequency (%)	RCFs frequency (%)
Resident gender
Male	2,379 (24.67)	1,671 (21.34)
Female	7,265 (75.33)	6,161 (78.66)
Respondent gender
Male	3,193 (33.11)	2,733 (34.90)
Female	6,451 (66.89)	5,099 (65.10)
Resident age
<75	1,190 (12.34)	347 (4.43)
75–84	2,527 (26.20)	1,352 (17.26)
85–94	4,453 (46.17)	4,599 (58.72)
95+	1,474 (15.28)	1,534 (19.59)
Respondent age
<45	262 (2.72)	120 (1.53)
45–54	1,203 (12.47)	892 (11.39)
55–64	3,447 (35.74)	3,409 (43.53)
65–74	3,376 (35.01)	2,822 (36.03)
75+	1,356 (14.06)	589 (7.52)
Race
Black	459 (4.76)	99 (1.26)
White	9,051 (93.85)	7,683 (98.10)
Other	134 (1.39)	50 (0.64)
Educational attainment
<High school	161 (1.67)	24 (0.31)
High school	4,752 (49.27)	2,592 (33.09)
Completed college	3,257 (33.77)	3,242 (41.39)
Masters or higher	1,474 (15.28)	1,974 (25.20)
Visitation frequency
Daily	1,824 (18.91)	761 (9.72)
Several times a week	3,772 (39.11)	3,395 (43.35)
Once a week	2,069 (21.45)	2,166 (27.66)
2–3 times per month	1,035 (10.73)	761 (9.72)
Once a month	500 (5.18)	349 (4.46)
Few times a year	444 (4.60)	400 (5.11)
Anticipated length of stay
Less than 1 month	222 (2.30)	13 (0.17)
From 1 to 3 months	482 (5.00)	75 (0.96)
Greater than 3 months	8,940 (92.70)	7,744 (98.88)
Kinship tie
Spouse	1,092 (11.32)	250 (3.19)
Child	5,713 (59.24)	5,847 (74.66)
Grandchild	95 (0.99)	48 (0.61)
Niece/nephew	424 (4.40)	397 (5.07)
Son/daughter in Law	528 (5.47)	421 (5.38)
Brother/sister	697 (7.23)	303 (3.87)
Friend	155 (1.61)	78 (1.00)
Parent	320 (3.32)	188 (2.40)
Guardian	391 (4.05)	141 (1.80)
Other	229 (2.37)	159 (2.03)
Facility size
Small	3,201 (33.19)	2,608 (33.30)
Medium	3,355 (34.79)	2,576 (32.89)
Large	3,088 (32.02)	2,648 (33.81)
Ownership
For-profit	6,904 (71.59)	5,447 (69.55)
Not-for-profit	2,740 (28.41)	2,385 (30.45)
Location
Urban	2,257 (23.40)	1,363 (17.40)
Rural	7,387 (76.60)	6,469 (82.60)
Total	9,644 (100.00)	7,832 (100.00)

Note. All variables are respondent-related unless otherwise noted. NH = nursing home; RCF = residential care facility.

Reliability

To assess reliability of the satisfaction scales, Cronbach’s alphas were calculated for each domain, and for overall satisfaction in both facility types. The lowest Cronbach’s alpha value (Table 2) was 0.66, with most in the 0.8 to 0.9 range. Cronbach’s alpha for overall satisfaction was 0.95 to 0.96.

Table 2.

Cronbach’s Alpha of the Satisfaction Indicators and Overall Satisfaction by Facility Type.

Indicator	# items	NHs (n = 9,644)	RCFs (n = 7,832)
Indicator	# items	Cronbach’s alpha	Cronbach’s alpha
Moving in	3	0.82	0.77
Spending time	6	0.90	0.87
Care and services	2	0.86	0.86
Caregivers	5	0.85	0.83
Meals and dining	3	0.75	0.72
Environment	3	0.69	0.66
Facility culture	8	0.93	0.91
Overall	30	0.96	0.95

Note. NH = nursing home; RCF = residential care facility.

Validity

To assess the validity of the satisfaction measure, the relationship between the overall satisfaction score was assessed with an item about recommending the facility, often suggested as a good single-item measure (Ejaz & Straker, 2001). The overall satisfaction score was significantly correlated with that item in both samples (NH: r = .80, p < .001; RCF: r = 0.77, p < .001).

Comparing the Satisfaction Domains Between NHs and RCFs

Independent samples t-tests assessed differences in satisfaction level between NHs and RCFs (see Table 3). The mean satisfaction score was significantly higher for RCFs than NHs overall and within nearly all satisfaction domains (no significant difference for care and services). The magnitudes of the differences were assessed using Cohen’s d, with the largest differences in satisfaction between NHs and RCFs in the environment domain, followed by moving in.

Table 3.

Family Satisfaction Independent Samples t-Tests Results.

Indicator	Mean satisfaction		t value	Effect size (Cohen’s d)	Sig.
Indicator	NH (n = 9,644)	RCF (n = 7,832)	t value	Effect size (Cohen’s d)	Sig.
Moving in	88.76	93.21	−21.85	−0.33	.000***
Spending time	80.19	82.82	−11.22	−0.17	.000***
Care and services	89.36	89.50	−0.59	−0.009	.274
Caregivers	84.72	88.55	−18.28	−0.28	.000***
Meals and dining	83.72	86.31	−11.14	−0.17	.000***
Environment	83.41	89.20	−26.38	−0.40	.000***
Facility culture	86.23	88.27	−9.42	−0.14	.000***
Overall	85.20	88.27	−17.63	−0.27	.000***

Note. NH = nursing home; RCF = residential care facility; n = number; Sig = significance level.

***

p < .001.

Multiple Regression Results

Separate regression models for NHs and RCFs assessed the relationship between overall family satisfaction and resident/family/facility characteristics (see Table 4). Results indicated that resident and respondent gender played a role in satisfaction. Resident gender was only associated with RCF satisfaction (but was not significantly different between models), with satisfaction scores about 0.94 points lower for females compared to males. Female respondents were significantly less satisfied than males in both facility types, and this effect did not significantly differ between them. Resident age was related to a higher respondent satisfaction score in NHs only, and this finding was significantly different between the two settings. Older respondents were significantly more satisfied in both settings, as were White compared to Black respondents. Neither of these differences were significantly different between facility types. Respondent education was not related to their satisfaction in either setting.

Table 4.

Family Satisfaction Regression and t-Test Results Comparing NHs and RCFs.

Characteristics	NH: β (SE)	RCF: β (SE)	t value
Resident gender (ref = male)
Female	−0.27 (0.31)	−0.94 (0.31)**	1.50
Respondent gender (ref = male)
Female	−2.28 (0.25)***	−1.60 (0.25)***	−1.88
Resident age	0.05 (0.02)**	−0.01 (0.02)	2.08*
Respondent age	0.09 (0.02)***	0.07 (0.02)***	0.84
Race (ref = Black)
White	3.31 (0.64)***	2.82 (1.25)*	0.35
Other	1.57 (1.05)	−0.12 (1.76)	0.82
Education (ref ≤ HS)
HS completed	1.43 (0.83)	0.78 (1.71)	0.34
Completed college	1.43 (0.85)	0.55 (1.7)	0.46
Master’s or higher	1.58 (0.88)	1.46 (1.72)	0.06
Visitation frequency (ref = daily)
Several times a week	0.56 (0.37)	1.04 (0.48)*	−0.79
Once a week	1.06 (0.42)*	0.71 (0.51)	0.53
2–3 times per month	1.99 (0.52)***	1.17 (0.6)*	1.03
Once a month	2.76 (0.63)***	0.81 (0.69)	2.09*
Few times a year	3.07 (0.63)***	1.21 (0.67)	2.02*
Anticipated length of stay (ref ≤ 1 month)
From 1 to 3 months	0.56 (0.95)	1.97 (2.58)	−0.51
More than 3 months	2.75 (0.79)***	6.85 (2.33)**	−1.66
Perceived ADL assistance	−0.26 (0.04)***	−0.04 (0.05)	−3.48**
Kinship tie (ref = spouse)
Child	0.27 (0.59)	2.28 (0.9)*	−1.87
Grand child	−0.58 (1.43)	4.22 (1.73)*	−2.16*
Niece/nephew	0.9 (0.77)	1.9 (1.05)	−0.77
Son/daughter-in-law	0.91 (0.76)	2.44 (1.04)*	−1.19
Brother/sister	−0.09 (0.58)	1.71 (0.97)	−1.60
Friend	0.53 (1.02)	1.84 (1.3)	−0.80
Parent	−0.53 (0.81)	2.14 (1.17)	−1.88
Guardian	0.59 (0.78)	1.58 (1.28)	−0.66
Other	−1.17 (0.83)	2.16 (1.18)	−2.31*
Help	−0.78 (0.06)***	−0.68 (0.06)***	−1.10
Facility size (ref = small)
Medium	−1.58 (0.44)***	−0.74 (0.49)	−1.28
Large	−2.58 (0.45)***	−1.78 (0.52)***	−1.16
Payment source (Medicaid %)	−0.07 (0.01)***	−0.04 (0.01)***	−1.74
Ownership (ref = for profit)
Not for profit	2.34 (0.41)***	1.92 (0.43)***	0.72
Location (ref = rural)
Urban	−1.00 (0.41)*	−0.72 (0.51)	−0.44
Constant	78.24 (2.15)***	79.3 (3.86)***	−0.17
# of observations	16,196	10,330	−
R ²	.075	.051	−

Note. Individual variables are respondent-related unless otherwise noted; NH = nursing home; RCF = residential care facility; HS = high school; # denotes the sample size after multiple imputation.

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

There was a general trend of increasing satisfaction with less-frequent visitation in NHs, but less consistent RCF results. The t-tests were significant for those who visit once a month and a few times a year, indicating that the negative relationship between visitation and satisfaction is stronger in NHs than RCFs.

One of the largest satisfaction differences was observed for anticipated length of stay in RCFs. Satisfaction was almost seven points higher when respondents expected their relative to stay for more than 3 months, compared to less than 1 month. The satisfaction was only 2.8 points higher among NH respondents for the same, but the t-test did not confirm a facility difference.

Perceived level of ADL assistance required was significantly negatively associated with satisfaction in NHs only, and the coefficients were significantly different between facility types. Kinship tie was associated with satisfaction in RCFs only, with higher satisfaction for child, grandchild, and son/daughter in law, compared to spouse. The coefficient for grandchild was statistically different between the settings, with higher RCF satisfaction. The extent to which respondents help with activities during visits was negatively associated with satisfaction in both settings, with no difference between NHs/RCFs.

Respondents were less satisfied in medium and large, compared to small NHs, and large compared to small RCFs. A higher percentage of Medicaid-reliant residents was related to lower satisfaction in both settings. Compared to for-profit, respondents were more satisfied with nonprofit facilities in NHs and RCFs. Finally, lower satisfaction was observed for urban versus rural NHs only. No differences between NHs and RCFs were observed for facility size, payment source, ownership, or location.

Discussion

Our hypotheses suggested that family satisfaction would be higher for RCFs than for NHs. We also hypothesized that lower satisfaction would be related to female respondent gender, more visitation, and shorter length of stay in NHs only. Furthermore, we expected larger facility size and higher Medicaid proportion to be negatively associated with satisfaction in both facility types, but to play a larger role in RCFs. The first hypothesis was clearly supported, with significantly higher satisfaction in RCFs overall, and within nearly all satisfaction domains (except care and services). The findings provided mixed support for our other hypotheses. As predicted, more frequent visitation was significantly related to lower satisfaction, with a larger effect observed in the NH sample. Female respondent gender, longer length of stay, larger facility size, and higher Medicaid proportion were related to lower satisfaction in both settings, but no significant differences were observed between facilities.

Given one of our rationales for why satisfaction between facility types may be different (differences in the physical and social environments—drawn from the PIE perspective), it was particularly interesting to find the largest satisfaction difference within the environment-related domain. Those questions focused on perceptions about the physical environment, with items about facility cleanliness, safety of belongings, and whether residents can get outside often enough. The next biggest difference was observed within the moving in domain, which was the highest RCF domain, and second highest within NHs. Those items assessed whether information was received about moving in expectations, facility orientation, and welcoming of families. These factors are the family members’ initial interaction with facilities and could be considered aspects of a facility’s social environment, which also fits nicely with what PIE suggests.

For both facility types, the lowest level of satisfaction was observed within the spending time domain. Those items were related to whether residents could spend time doing enjoyable/meaningful things, connectedness to the community, and family involvement in the NH/RCF. Previous research supports this idea that family involvement is key to higher satisfaction (Kemp et al., 2013). The importance of this domain is underscored by the fact that over 93% of NH and 99% of RCF relatives expect their residents to stay at the facility for more than 3 months. Creating opportunities for care collaborations between family, residents, and NH/RCF staff is particularly important.

Female respondent gender was related to lower satisfaction in both settings. This is not surprising, given that previous research has suggested that women more frequently serve as a caregiver, and report more negative experiences (Port et al., 2005; Revenson et al., 2016). Although care may have transitioned from informal at-home care to a facility, the respondent was the person identified as the most involved family member in the resident’s care.

In both settings, older respondent age was associated with higher satisfaction. This is consistent with previous research about RCF family satisfaction (Dobbs & Montgomery, 2005). Visitation frequency was also significant in both models, with higher satisfaction as visitation frequency decreased for NHs, but a weaker, less clear pattern for RCFs. Likewise, providing more help during visits was related to lower satisfaction. It is possible that respondents visit more and provide more help as a response to lower satisfaction with the facility’s care level. Larger facility size, higher percentage of Medicaid payment, and for-profit ownership were significantly related to lower family satisfaction in both settings, with no differences between NHs and RCFs.

Several factors were associated with satisfaction in only one facility type, and several coefficients were significantly different between NHs/RCFs. Relatives of female RCF (but not NH) residents indicated lower satisfaction. This difference could be related to gender differences in the relationship between the respondent and resident, or communication differences between the respondent and RCF staff. Amount of help provided could also play a role if differences by resident gender exist. White NH respondents were also more satisfied than those who identified as Black, consistent with previous research (Griffore et al., 2011).

Frequency of visitation was significantly associated with satisfaction in both settings, but the pattern was clearer among NH respondents. If concerns about care underlie this difference (Roberts & Ishler, 2018), the increased likelihood that NH residents require more serious care could explain this finding. When an RCF respondent anticipated a resident stay longer than 3 months, their satisfaction score was nearly seven points higher than those who anticipated a short stay. This factor was also significant for NH respondents, but the difference was only about three points. This finding could reflect a more careful facility choice for a permanent home, rather than a quick choice for a postacute stay, particularly for RCFs.

More ADL assistance was associated with lower satisfaction among NH respondents only. This difference is somewhat counter-intuitive, given that care requirements of NH residents would likely be higher. It may be related to greater concern about declining health of respondents’ NH residents, which may be worse compared to RCFs.

Kinship ties only seemed to play a role in the RCF setting. Compared to spouses, higher satisfaction was observed for children, grandchildren, and son/daughter in laws. More closely connected kin may have higher expectations for care, particularly in RCFs.

Implications and Future Research

Understanding satisfaction differences between NHs and RCFs has theoretical implications as well as implications for potential residents, families, and facility administrators. This study investigated some physical contexts of PIE theory (e.g., facility size), but subsequent research must turn its attention to the role that psychological and social processes of PIE might play in family satisfaction across both NH and RCF settings. Such an approach would advance the PIE framework by permitting a direct assessment of a conceptual model that encompasses both environmental and psychological processes operating at different structural levels. Furthermore, behavioral (e.g., anxiety), cognitive (e.g., logic and reasoning), social (e.g., access to LTC), and perceptual factors of family satisfaction could be assessed through the investigation of psychological indicators. This approach might also include an exploration of resident’s and their family’s expectations (e.g., length of stay), preferences in terms of perceived fit, and changes in resident well-being over time. The utilization of such approaches could contribute to distinguishing which behavioral, cognitive, social, and overall psychological processes at various structural levels will inform family satisfaction for both the resident and their families.

In terms of implications for individuals, information about care quality and satisfaction can play a role in facility choice. Families may have different expectations for facilities, related to their gender, age, and roles as caregivers. In addition, administrators can use these results to understand satisfaction domains that could be improved. For example, NHs and RCFs should address how residents spend time, with an additional focus on the environment and dining in NHs. These results could also increase awareness among all stakeholders that characteristics of residents and family may affect their satisfaction.

Higher satisfaction in RCFs compared to NHs may also be linked to facility and individual differences not measured in this study. We did not have available data to account for other known factors, such as chain affiliation and residents’ physical, cognitive, or intellectual activity (Shippee et al., 2017). Future studies could also address differences in communication between family and staff. Potential modifying relationships could also be explored. For example, kinship ties could moderate the relationship between satisfaction and the help provided or visitation frequency. Likewise, gender and satisfaction could be moderated by the amount of help provided. Finally, future research could more fully address differences in NHs and RCFs by including family satisfaction, resident satisfaction, facility ratings, and facility characteristics in a single analysis.

Limitations and Strengths

This study is not without limitations. Only family satisfaction was assessed, not other measures of facility quality. In the regression models, the available independent variables only accounted for a modest amount of the variance in satisfaction. For example, information about the health or cognitive status of residents was not available, which could have an effect on family members’ satisfaction. This study was also cross-sectional, and only conducted within a single Midwestern state, which does not allow for any causal relationships to be determined and may limit generalizability. As is typical, our satisfaction measure was positively skewed suggesting that these results should be interpreted with caution.

This study also has several important strengths. To our knowledge, it represents the first to directly compare differences in family satisfaction between NHs and RCFs, including a comparison of different domains and satisfaction correlates. It includes a large sample and 30 items assessing many different satisfaction domains. The analyses allowed for independent assessments of how demographic, family, and health-related factors were related to each facility type, as well as a direct comparison of each coefficient across the models.

Footnotes

Acknowledgements

The authors would like to thank Jane Straker for her input on several aspects of the conceptualization and approach described in this paper. They would also like to thank the Ohio Department of Aging and the Scripps Gerontology Center for making the data for this project available.

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.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The analyses reported here were conducted without funding; original data collection was funded by a contract with the Ohio Department of Aging.

ORCID iDs

Nader Mehri

Nytasia Hicks

Jonathon M. Vivoda

References

Abrahamson

Bradley

D. B.

Morgan

K. H.

Fulton

B. R.

Ibrahimou

(2013). Influence of satisfaction with services on assisted living resident satisfaction. Journal of Housing for the Elderly, 27(1–2), 177–190. https://doi.org/10.1080/02763893.2012.754821

Campbell

L. J.

Cai

Gao

(2016). Racial/ethnic disparities in nursing home quality of life deficiencies, 2001 to 2011. Gerontology and Geriatric Medicine, 2, Article 653561. https://doi.org/10.1177/2333721416653561

Centers for Medicare and Medicaid Services. (2019). About Medicare Nursing Home compare data. https://www.medicare.gov/nursinghomecompare/Data/About.html

Chou

S.-C.

Boldy

D. P.

Lee

A. H.

(2003). Factors influencing residents’ satisfaction in residential aged care. The Gerontologist, 43(4), 459–472. https://doi.org/10.1093/geront/43.4.459

Curtis

M. P.

Sales

A. E.

Sullivan

J. H.

Gray

S. L.

Hedrick

S. C.

(2005). Satisfaction with care among community residential care residents. Journal of Aging and Health, 17(1), 3–27. https://doi.org/10.1177/0898264304268590

Cvitkovich

Wister

(2001). Chapter 1. A comparison of four person-environment fit models applied to older adults. Journal of Housing for the Elderly, 14(1–2), 1–25. https://doi.org/10.1300/J081v14n01_01

Dobbs

Montgomery

(2005). Family satisfaction with residential care provision: A multilevel analysis. Journal of Applied Gerontology, 24(5), 453–474. https://doi.org/10.1177/0733464805279374

Ejaz

Straker

(2001). Final report on development and testing for resident satisfaction survey (pp. 1–48). Miami University. http://sc.lib.miamioh.edu/handle/2374.MIA/4526

Green

McDermott

(2010). Social work from inside and between complex systems: Perspectives on person-in-environment for today’s social work. The British Journal of Social Work, 40(8), 2414–2430. https://doi.org/10.1093/bjsw/bcq056

10.

Griffore

R. J.

Phenice

L. A.

Post

L. A.

(2011). Racial differences in nursing home experiences. International Journal of Health, Wellness & Society, 1(1), 107–115. https://doi.org/10.18848/2156-8960/CGP/v01i01/41066

11.

Kane

R. A.

Kling

K. C.

Bershadsky

Kane

R. L.

Giles

Degenholtz

H. B.

Liu

Cutler

L. J.

(2003). Quality of life measures for nursing home residents. The Journals of Gerontology: Series A, 58(3), M240–M248. https://doi.org/10.1093/gerona/58.3.M240

12.

Kane

R. L.

Kane

R. A.

Bershadsky

Degenholtz

Kling

Totten

Jung

(2005). Proxy sources for information on nursing home residents’ quality of life. The Journals of Gerontology: Series B, 60(6), S318–S325. https://doi.org/10.1093/geronb/60.6.S318

13.

Karls

J. M.

(2002). Person-in-enviornment system: It’s essence and applications. In Roberts

A. R.

Greene

G. J.

(Eds.), Social workers’ desk reference (pp. 194–198). Oxford University Press.

14.

Karls

J. M.

O’Keefe

M. E.

(2009). Person-in-environment system. In Roberts

A. R.

(Ed.), Social workers’ desk reference (2nd ed., pp. 371–376). Oxford University Press.

15.

Kemp

C. L.

Ball

M. M.

Hollingsworth

Lepore

M. J.

(2010). Connections with residents: It’s all about the residents for me. In Ball

M. M.

Perkins

M. M.

Hollingsworth

Kemp

C. L.

(Eds.), Frontline workers in assisted living (pp. 147–170). Johns Hopkins University Press.

16.

Kemp

C. L.

Ball

M. M.

Perkins

M. M.

(2013). Convoys of care: Theorizing intersections of formal and informal care. Journal of Aging Studies, 27(1), 15-29. https://doi.org/10.1016/j.jaging.2012.10.002

17.

Lawriter. (2018). Ohio laws and rules: 3701 Department of Health—Administration and Director. codes.ohio.gov/oac/3701

18.

Mitchell

J. M.

Kemp

B. J.

(2000). Quality of life in assisted living homes: A multidimensional analysis. The Journals of Gerontology: Series B, 55(2), P117–P127. https://doi.org/10.1093/geronb/55.2.P117

19.

National Institute on Aging. (2017). Residential facilities, assisted living, and nursing homes. https://www.nia.nih.gov/health/residential-facilities-assisted-living-and-nursing-homes

20.

Ohio Department of Aging. (2018). Long-term care consumer guide. https://ohio.gov/wps/portal/gov/site/residents/resources/long-term-care-consumer-guide

21.

Ohio Department of Health. (2019a). Nursing homes/facilities. https://odh.ohio.gov/wps/portal/gov/odh/know-our-programs/nursing-homes-facilities/nursinghomesfacilities

22.

Ohio Department of Health. (2019b). Residential care facilities: Assisted living. https://odh.ohio.gov/wps/portal/gov/odh/know-our-programs/residential-care-facilities-assisted-living/residentialcarefacilitiesassistedliving

23.

Port

C. L.

Zimmerman

Williams

C. S.

Dobbs

Preisser

J. S.

Williams

S. W.

(2005). Families filling the gap: Comparing family involvement for assisted living and nursing home residents with dementia. The Gerontologist, 45(suppl_1), 87–95. https://doi.org/10.1093/geront/45.suppl_1.87

24.

Revenson

T. A.

Griva

Luszczynska

Morrison

Panagopoulou

Vilchinsky

Hagedoorn

(2016). Gender and caregiving: The costs of caregiving for women. In Revenson

T. A.

Griva

Luszczynska

Morrison

Panagopoulou

Vilchinsky

Hagedoorn

(Eds.), Caregiving in the illness context (pp. 48–63). Palgrave Macmillan. https://doi.org/10.1057/9781137558985_5

25.

Roberts

A. R.

Ishler

K. J.

(2018). Family involvement in the nursing home and perceived resident quality of life. The Gerontologist, 58(6), 1033–1043. https://doi.org/10.1093/geront/gnx108

26.

Scripps Gerontology Center. (2020). 2015 Biennial survey. https://miamioh.edu/cas/academics/centers/scripps/survey-pages/biennial-survey_2014/

27.

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

28.

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(4), 643–655. https://doi.org/10.1093/geront/gnt148

29.

Shippee

T. P.

Bowblis

Cooke

(2018). Racial disparities in quality of life for nursing home residents. Innovation in Aging, 2(Suppl. 1), 609–610. https://doi.org/10.1093/geroni/igy023.2267

30.

Shippee

T. P.

Roberts

A. R.

Bowblis

J. R.

(2020). Family satisfaction with nursing home care: Findings and implications from two state comparison. Journal of Applied Gerontology, 39(4), 385–392. https://doi.org/10.1177/0733464818790381

31.

Sikorska

(1999). Organizational determinants of resident satisfaction with assisted living. The Gerontologist, 39(4), 450–456. https://doi.org/10.1093/geront/39.4.450

32.

Sloane

P. D.

Zimmerman

Gruber-Baldini

A. L.

Hebel

J. R.

Magaziner

Konrad

T. R.

(2005). Health and functional outcomes and health care utilization of persons with dementia in residential care and assisted living facilities: Comparison with nursing homes. The Gerontologist, 45(suppl_1), 124–134. https://doi.org/10.1093/geront/45.suppl_1.124

33.

Sloane

P. D.

Zimmerman

Hanson

Mitchell

C. M.

Riedel-Leo

Custis-Buie

(2003). End-of-life care in assisted living and related residential care settings: Comparison with nursing homes. Journal of the American Geriatrics Society, 51(11), 1587–1594. https://doi.org/10.1046/j.1532-5415.2003.51511.x

34.

Spillman

B. C.

Liu

McGilliard

(2002). Trends in residential long-term care: Use of nursing homes and assisted living and characteristics of facilities and residents. The Urban Institute. https://aspe.hhs.gov/basic-report/trends-residential-long-term-care-use-nursing-homes-and-assisted-living-and-characteristics-facilities-and-residents

35.

StataCorp. (2017). Stata statistical software: Release 15.1 (Version 15.1) [Computer software].

36.

Straker

J. K.

Ejaz

F. K.

McCarthy

Jones

J. A.

(2007). Developing and testing a satisfaction survey for nursing home residents. Journal of Aging & Social Policy, 19(2), 83–105. https://doi.org/10.1300/J031v19n02_05

37.

Straker

J. K.

Parajuli

Eynon-Black

Nelson

Shanley

Chow

(2017). Implementation of the 2016 Ohio nursing home and residential care facility family satisfaction surveys. Scripps Gerontology Center, Miami University. //miamioh.edu/cas/academics/centers/scripps/research/publications/2017/08/Implementation-of-the-2016-Ohio-Nursing-Home-Family-Satisfaction-Survey.html

38.

Williams

Straker

J. K.

Applebaum

(2016). The nursing home five star rating: How does it compare to resident and family views of care? The Gerontologist, 56(2), 234–242. https://doi.org/10.1093/geront/gnu043

39.

Williams

S. W.

Desai

Rurka

J. T.

Mutran

E. J.

(2008). Predictors of satisfaction for African-American and White family caregivers of adult care home residents. Journal of Applied Gerontology, 27(5), 568–587. https://doi.org/10.1177/0733464808317169

40.

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(2), 127–142. https://doi.org/10.1016/j.archger.2013.03.015

41.

Zimmerman

Cohen

L. W.

Reed

Gwyther

L. P.

Washington

Cagle

J. G.

Beeber

A. S.

Sloane

P. D.

(2013). Comparing families and staff in nursing homes and assisted living: Implications for social work practice. Journal of Gerontological Social Work, 56(6), 535–553. https://doi.org/10.1080/01634372.2013.811145