Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: J Aging Environ. 2020 Nov 25;35(3):319–337. doi: 10.1080/26892618.2020.1848966

Profiles of Expectations Regarding Aging After Relocation to Assisted Living

Evan Plys 1,2,3, Ronald Smith 4
PMCID: PMC8412139  NIHMSID: NIHMS1667297  PMID: 34485981

Abstract

This study classified assisted living (AL) residents by expectations for future physical, cognitive, and socioemotional health in later life. The authors analyzed cross-sectional data from 202 AL residents. Study participants completed the 12-item expectations regarding aging survey. A K-means cluster analysis identified three subgroups: “lower expectations” (n = 55), “higher expectations” (n = 70), and “expecting adaptiveness” (n = 77). Subgroups varied by select demographic, health, and relocation-related factors. Results suggest that, despite similarities in physical and cognitive functioning, AL residents reported different expectations for health. Socioemotional functioning may help explain observed discrepancies between current health and future health-related expectations.

Keywords: expectations regarding aging, attitudes toward aging, health expectations, assisted living, long-term care

Assisted living communities (ALs) are a residential long-term care option for adults who require supportive services to manage everyday routines (Kane et al., 2007). ALs provide health-related support and access to social and quality of life enhancing services (e.g., recreational activities) within close proximity (Mollica et al., 2012). The AL model of care aims to prioritize holistic wellness and promote core values of autonomy, dignity, choice, and control (Kane et al., 2007). Increasing need for long-term services and supports, coupled with the desirable psychosocial model of care, contributed to the significant growth of the AL industry (Brown Wilson, 2007). The demand for, coupled with the growth and regulatory variability of, AL has also led many communities to accommodate residents with a broad range of physical, cognitive, and socioemotional functioning (Polivka & Salmon, 2011).

The majority of AL residents are aged 85 or older (Khatutsky et al., 2016); yet, vary in health profile and functional ability. For example, national estimates suggest that 77% of AL residents require assistance with three or more instrumental activities of daily living (e.g., medication assistance, money management, or shopping), 57% of residents ambulate with assistance (e.g., walker or wheelchair), and 37% require assistance with multiple activities of daily living (e.g., bathing, dressing, and toileting; Harris-Kojetin et al., 2019; Khatutsky et al., 2016). Mental health disorders are also prevalent in AL, with over 30% of residents reporting depression, and over 40% of residents scoring in the moderate or severe impairment range on cognitive screens (Harris-Kojetin et al., 2019; Khatusky et al., 2016; Zimmerman et al., 2014). These statistics represent significant variability in functioning; although, AL residents evidence greater physical and cognitive impairment compared to community-living older adults and lower levels of impairment compared to nursing home residents (Golant, 2004; Grabowski et al., 2015). Although, it should be noted that variability at the facility-level (e.g., admission and retention policies) complicates the ability for investigators to capture differences in resident functional profiles within the same community (Golant, 2004).

Relocating to a residential long-term care setting is not a normative life event in older adulthood. Relocation is often precipitated by a major decline in health or the threat of future decline (Chen et al., 2008). In previous studies, AL residents reported various reasons for moving to their community (e.g., previous home was too much to manage or to be closer to family and friends; Bäumker et al., 2012). Reasons and the level of perceived control in the relocation process can affect residents’ experiences and satisfaction with AL after the move (Brownie et al., 2014; Jungers, 2010). Thus, contextual factors related to relocation are important to consider in psychosocial research in AL.

Theoretically, the more supportive environment of AL should increase engagement with life after relocation (Mitty, 2010). However, findings are mixed, with some studies reporting that relocation to AL is met with new opportunities for social and recreational growth (Fraher & Coffey, 2011; Tracy & DeYoung, 2004), whereas other studies report that residents experience sedentary and isolated lifestyles (Pironhen et al., 2018; Pomeroy et al., 2011; Resnick et al., 2010). Taken together, ALs represent a diverse group of older adults who vary in health and ability, reasons for relocation, and psychosocial experiences after the move. These factors may relate to expectations for the future. However, little research investigates expectations for future health and functioning after relocation to AL.

Previous research suggests that the experiences and meanings associated with relocation to an AL may lead residents to feel disengaged from society and a loss of autonomy, which may also relate to negative self-views and internalized age stereotypes (Jungers, 2010; Moss & Moss, 2007). Conversely, it is possible that increased opportunities for social and recreational engagement may contribute to positive self-views and perceptions of aging (Andrews et al., 2017). Previous research suggests that older adults in residential care settings report both positive and negative changes in perceptions of aging and future health beliefs after relocation (Seifert & Schelling, 2018). However, the profiles of residents with positive and negative views of future health and aging are not clearly defined in the literature.

Expectations regarding aging (ERA) refer to the perceived ability to achieve and maintain or decline in physical, cognitive, and socioemotional functioning in older adulthood (Sarkisian et al., 2001; 2002). Previous research suggests that this construct, and associated measurement tools, captures perceptions and stereotypes of aging that manifest as expectations for health in later life (Levy et al., 2014; Sarkisian et al., 2005). Across multiple studies, higher ERA related with better physical health, functional ability, psychological well-being, health behaviors, social support, and quality of life in older adulthood (Dogra et al., 2015; Kim, 2009; Levy et al., 2014; Menkin et al., 2017; Sarkisian et al., 2002; 2005). To our knowledge, the ERA of AL residents has yet to be assessed. Although, previous studies found that greater attitudes toward own aging among AL residents related with higher subjective health, community satisfaction, and social support, as well as lower functional limitations and depression (Jang et al., 2006; Park et al., 2015). It should be noted, however, that ERA has similarities and differences with attitudes toward own aging. For example, both ERA and attitudes toward own aging capture individual and societal perceptions of older adulthood (Diehl et al., 2014). However, attitudes toward own aging measures individuals’ personal aging experience, whereas ERA assesses global perceptions of health and ability in later life (Lawton, 1975; Sarkisian et al., 2002).

Given that there are multiple markers of health and wellness in older adulthood, ERA represents a multidimensional construct and measurement strategy (Sarkisian et al., 2002; 2005). Capturing these multiple dimensions may be scientifically and clinically meaningful. For example, individuals with some chronic conditions may expect to achieve high socioemotional well-being and, at the same time, expect physical functioning to decline. Further, previous studies found targeted relationships between specific ERA subscales (i.e., physical, cognitive, and socioemotional ERA) and outcomes, such as social support, exercise, social and recreational activities, and exercise (e.g., only physical ERA relates with physical activity; Meisner et al., 2013; Meisner & Baker, 2013; Menkin et al., 2017). Therefore, appreciating the multiple dimensions of health and wellness in older adulthood can provide additional insight into the relationships between ERA, health, and behaviors among older adults. The purposes of this study were to classify AL residents by their physical, cognitive, and socioemotional ERA, and to describe the differences in demographic, health, and relocation factors among identified subgroups.

Materials and Methods

Study Design

The current study analyzed secondary data from a cross-sectional, quantitative protocol that targeted activity engagement and psychosocial well-being in AL (see Plys & Qualls, 2019). The protocol was approved by the affiliated Institutional Review Board.

Participants

The current sample included 202 AL residents, ages 51 to 100 (M = 83.05, SD = 10.32). Consistent with national data (Khatusky et al., 2016), most participants were female (n = 143, 72%), white (n = 180, 90%), and widowed (n = 118, 59%). Sixty-seven percent (n = 133) had more than a high school education, which is greater than national estimates (i.e., 42%; Khatusky et al., 2016). Thirty-one percent (n = 61) used Medicaid as financing strategy, which is greater than the national average of 19% (Caffrey et al., 2012). Length of stay ranged from less than a month to over 13 years (M = 21.34 months, SD = 20.48).

Setting

Participants were recruited from 21 ALs ranging in size from 16 to 104 residents (M = 54.19, SD = 25.07). Most ALs were chain operated (n = 11) and cost more than $3,000 per month (n = 16). In 14 ALs, 75% or more of the resident population used private payments to finance services, and, in four ALs, 75% or more of the residents used Medicaid.

Procedures

The current study recruited residents at targeted ALs using multiple methods, such as staff nominations, public announcements, and snowball sampling. Inclusion criteria was: (a) age 50 or older; (b) able to speak and understand English; (c) own decision-maker; and (d) score of five or above on the Memory Impairment Screen (Buschke et al., 1999). Thirty-three residents were excluded based on these criteria. After providing written consent, participants completed questionnaires consisting of measures targeting demographics, health, relocation, and psychosocial factors. Additional study procedures are described elsewhere (see Plys & Qualls, 2019).

Measures

Clustering Variables

The 12-item ERA survey (ERA-12) contains three, four-item subscales assessing physical, cognitive, and socioemotional ERA (Sarkisian et al., 2005). This measure was developed for use with older adults and evidenced strong reliability and validity in community-living older adult samples (Sarkisian et al., 2002; 2005). Participants responded to items (e.g., “The human body is like a car: when it gets old, it gets worn out”) on a four-point scale, 1 = Definitely true, 2 = Somewhat true, 3 = Somewhat false, 4 = Definitely false. Consistent with procedures outlined by the measure’s developers (Sarkisian et al., 2005), we performed a linear transformation on each subscale to reflect a possible range of 0 to 100. Higher scores indicate expectations for achieving and maintaining health and functioning, whereas lower scores indicate expectations for decline in heatlh and functioning. To aid in interpretation, scores below 50 were considered “low ERA” and scores 50 and above were considered “high ERA”. Internal consistency with the current sample was acceptable for the physical (Cronbach’s α = .79), cognitive (Cronbach’s α = .78), and socioemotional (Cronbach’s α = .74) subscales.

Demographics

Staff at each AL reported community demographics, such as size and cost. Resident participants self-reported demographic information, including: gender, age, education, relationship status, race and ethnicity, and financing strategies (i.e., Medicaid or private payment). Length of stay, rounded to the nearest month, was calculated by subtracting month and year of testing from self-reported month and year of move-in.

Health Factors

The PROMIS Short-Form Global Health Scale v1.1 consists of two, four-item subscales: physical and socioemotional health (Hays et al., 2009). The physical health subscale contains items related to global physical health, functional ability, pain, and fatigue. The socioemotional health subscale consists of items related to global mental health, social satisfaction, and functional impact of mental health on social activities. Scores on each subscale range from 4 to 20, higher scores indicate greater health and functioning. Internal consistency with the current sample was acceptable for both physical (Cronbach’s α = .71) and socioemotional (Cronbach’s α = .73) health scales.

The Clock Drawing Task- Command (CDT) assessed cognitive ability. The CDT assesses a range of cognitive abilities and evidenced strong correlations with global cognitive screens (i.e., mini-mental status exam; Connor et al., 2005). Administration and three-factor quantitative scoring methods followed criteria outlined by Rouleau et al. (1992). Using a subsample of 28 participants, reliability between two independent raters was acceptable in the current study (ICC = .71, 95% CI = .48–.86, p < .001). Scores ranged from 0–10, higher scores indicate greater cognitive ability. Previous research suggests that scores of above seven are consistent with normal cognitive functioning (Connor et al., 2005). Based on this cutoff, 70 participants in the current sample evidenced scores consistent with, mostly, mild cognitive impairment.

Relocation Factors

A single-item, adopted from Davidson and O’Connor (1990), assessed perceived decisional control in the move to AL. The item asked, “How much input would you say that you had in the decision to come live here?” on a five-point scale 1 = Little or none, 5 = A lot. Participants also self-reported previous residence.

Participants self-reported reasons for relocation using an index of 18 factors created for use in the current study, based on previous research in residential long-term care settings (Bäumker et al., 2012; Bekhet et al., 2009). Participants had the option to self-generate up to four additional reasons not included in the original index. Consistent with previous studies (Bäumker et al., 2012), reasons for relocation were categorized by push-related (i.e., push from a previous residence), pull (i.e., pull toward an AL community), and external factors, see Table 1. Responses in each category were summed and dichotomized to indicate: 1 = Reason(s) reported, 0 = Reason(s) not reported.

Table 1.

Reasons for Relocation

Push from Previous Residence n % Pull toward AL Residence n % External-Other n %
Health-Related Factors Security or Safety Offered 46 23% Healthcare Providers, Family, or Friends Chose/Recommended Relocation* 87 43%
 Health Problem 94 47% Nice Appearance 41 20%
 Spouse’s Health 25 12% Desirable Amenities 36 18%
 Major Health Event (e.g., fall or death of spouse)* 19 9% Good Reputation 21 10%
Housing-Related Factors Staff Seemed Nice 16 8%
 Needed More Help at Home 30 15% Cost of the Community 15 7%
 Downsize Responsibilities at Home 42 21% Close to Previous Home 14 7%
 Fear of Accident at Previous Home 53 26% Enjoyable Activities 12 6%
 Priced Out of Residence* 8 4% Friends Already Living in Community 11 5%
 Did Not Like Previous Residence 23 11% Opportunity to Age in Place/Continuing Care* 5 2%
Psychosocial-Related Factors Community Accepted Pets* 2 1%
 Wanted to be Closer to Family 43 21%
 Felt Lonely 17 8%
 Not Burden Others 34 17%
Open in a new tab

Note. Percentages based on entire sample (n = 202).

*

Not included in original index (i.e., participant-generated response).

Data Preparation & Statistical Analyses

Statistical analyses were conducted using SPSS v.26. The authors checked the dataset for missing data, outliers, and violations of univariate normality. Missing data was observed only for the CDT (n = 17; 8%), a mean imputation was used in analyses. No transformations to the data were required. Cronbach’s alpha with a cutoff of .7 assessed internal consistency for multi-item scales (Tavakol & Dennick, 2011).

To classify participants by physical, cognitive, and socioemotional ERA, the authors used cluster analysis. Cluster analysis organizes observed data into meaningful subgroups with homogeneous within-group and heterogenous between-group profiles (Hair Jr. et al., 2014). Prior to conducting the cluster analysis, statistical assumptions were checked. Based on visual inspection of boxplots and stem-and-leaf plots, no extreme cases were identified among the clustering variables. Skewness and kurtosis were acceptable among the clustering variables (ZSkewness = .16 to 2.51, ZKurtosis = −1.77 to −2.0; Kim, 2013). Subscales of the ERA were moderately to highly correlated (r = .43 to .52), without evidence of interdependence (i.e., r < .80; Sambandam, 2003). Thus, subscales of the ERA-12 were appropriate for cluster analysis.

The current study followed procedures outlined by Hair Jr. et al. (2014), which involves a combination of hierarchical and non-hierarchical clustering methods to enhance the accuracy of final cluster membership. First, to determine the number of clusters, we performed a hierarchical cluster analysis using a squared Euclidean distance measure and Ward’s linkage method. Scores were not standardized because subscales of the ERA-12 were already transformed consistent with previous recommendations (see Sarkisian et al., 2005). The authors compared various linkage methods (e.g., average linkage and Ward’s method) to optimize initial cluster seeds. The ideal number of clusters was identified by an inconsistent jump between coefficients in the agglomeration schedule and were verified using visual inspection of a dendrogram (Yim & Ramdeen, 2015), see Figure 1. These evaluations suggested a three-cluster solution. We then calculated silhouette coefficients, which measures both cohesion and separation, to evaluate the quality of the initial cluster solution; acceptable scores are greater than zero (Norusis, 2011). The three-cluster solution (silhouette coefficient = .5) evidenced a higher silhouette coefficient than a four- (silhouette coefficient = .4) or five-cluster (silhouette coefficient = .4) solution. Based on this comparison, the three-cluster solution was determined to be the best fit for the data.

Figure 1.

Figure 1

Open in a new tab

Dendogram from Hierarchical Cluster Analysis with Ward’s Linkage Method

Consistent with previous recommendations (Hair Jr. et al., 2014; Milligan, 1980), we identified final cluster membership using a K-means cluster analysis, which assigns all cases to the nearest cluster center (Zakharov, 2016). The K-means cluster analysis was conducted on subscales of the ERA-12 with a predefined three-cluster solution; maximum iterations were limited to 10 (Norusis, 2011). Sample size among final clusters was sufficient as indicated by each group containing at least 30 cases (i.e., 10*3 clustering variables; Sambandam, 2003). To validate cluster membership, we randomly generated a subsample of 104 participants, and conducted another K-means cluster analysis with just the subsample. Then, we compared cluster membership based on analyses run with the subsample and the complete sample using a kappa coefficient (Hair, 2014). Kappa coefficients were evaluated as: .4 to .6 = moderate; 6. to .8 = substantial; and .8 to 1.0 = almost perfect agreement (McHugh, 2012). This cross-validation strategy yielded almost perfect agreement in group membership (κ = .884). Final clusters are reported based on K-means groupings.

After identifying cluster membership, the authors tested groups differences across clusters by conducting a series of one-way ANOVAs with Tukey post-hoc tests for continuous variables and Chi-Square tests of independence with z-tests of column proportions for categorical variables. All assumptions for ANOVA and Chi-Square tests were evaluated, including: independence of measures, sample size/cell size, normality, outliers, and homogeneity of variances. Select measures evidenced violation of the homogeneity of variance assumption, as evidenced by p < .05 on a Levene’s Test; thus, we used a Welch statistic for all ANOVA analyses (Pallant, 2013). Significance levels were set at .05 for all analyses.

Results

Descriptive Statistics

Total ERA ranged from 0 to 91.67 (Median = 36.11; M = 37.83, SD = 19.10). Physical ERA scores ranged from 0 to 100 (Median = 33.33; M = 32.55, SD = 23.99). Cognitive ERA scores ranged from 0 to 91.67 (Median = 29.17; M = 30.07, SD = 22.77). Socioemotional ERA scores ranged from 0 to 100 (Median = 50; M = 50.87, SD = 24.90). Physical ERA significantly correlated with socioemotional ERA (r = .52, p < .001) and cognitive ERA (r = .43, p < .001); cognitive ERA significantly correlated with socioemotional ERA (r = .43, p < .001). Descriptive statistics for all study variables and group differences between clusters are presented in Table 2.

Table 2.

Sample Characteristics and Group Differences Across Clusters

Total Sample
(n = 202) 		Lower Expectations
(n = 55) 		Higher Expectations
(n = 70) 		Expecting Adaptiveness
(n = 77)
M SD M SD M SD M SD
Clustering Variables
 Physical ERA 32.55 23.99 16.06b,c 17.56 55.60a,c 16.77 23.38a,b 16.61
 Cognitive ERA 30.07 22.77 15.00b,c 17.96 50.24a,c 16.42 22.51a,b 17.32
 Socioemotional ERA 50.87 24.90 20.15b,c 10.48 71.07a,c 18.37 54.44a,b 13.56
Health Factors
 Physical Health 12.83 2.92 12.18 2.98 13.17 2.96 12.99 2.79
 Socioemotional Health 13.41 3.12 11.89b,c 3.31 14.30a 3.18 13.68a 2.50
 Cognitive Ability 7.69 1.84 7.56 1.97 8.10c 1.55 7.40b 1.96
Demographics
 Age 83.03 10.27 83.33 10.41 80.79c 11.45 84.87b 8.64
 Women 144 71% 34b 62% 56a 80% 54 70%
 Widowed 119 59% 31 56% 38 54% 50 65%
 Some college+ 135 67% 41 75% 46 66% 48 62%
 Medicaid Status 61 30% 21c 38% 26c 37% 14a,b 18%
 Length of Stay (months) 21.17 20.44 20.25 18.01 23.01 25.17 20.16 17.17
Relocation Factors
 Decision in Move 3.12 1.53 2.69c 1.41 3.23 1.56 3.34a 1.54
 Previous Residence:
  Acute/Rehabilitation Setting 25 12% 5 9% 9 13% 11 14%
  Long-Distance Home 24 12% 11b 20% 5a 7% 8 10%
  Local Home 80 40% 18b 33% 37a,c 53% 25b 33%
  Independent Senior Living Community 36 18% 11 20% 7c 10% 18b 23%
  Other AL 20 10% 6 11% 7 10% 7 9%
 Reasons for Relocation:
  Health-Related Push Factors 122 60% 29 53% 42 60% 51 66%
  Housing-Related Push Factors 116 57% 30 55% 39 56% 47 61%
  Psychosocial-Related Push Factors 78 39% 17 31% 33 47% 28 36%
  Pull Factors 103 51% 25 45% 33 47% 45 58%
  External Factors 87 43% 27 49% 24 34% 36 47%
Open in a new tab

Note. Percentages for dichotomous variables are calculated with cluster size as the denominator.

a

Significant compared to Cluster 1 (lower expectations) at p < .05.

b

Significant compared to Cluster 2 (higher expectations) at p < .05.

c

Significant compared to Cluster 3 (expecting adaptiveness) at p < .05.

Cluster Analysis

The first cluster (n = 55, 27%) was characterized by low physical, cognitive, and socioemotional ERA; this cluster was named “lower expectations”. The second cluster (n = 70, 35%) was characterized by high physical, cognitive, and socioemotional ERA; this cluster was named “higher expectations”. The third cluster (n = 77, 38%) was characterized by low physical and cognitive ERA and high socioemotional ERA; this cluster was named “expecting adaptiveness.” See Figure 2 for ERA scores for each subgroup.

Figure 2.

Figure 2

Open in a new tab

Subscales of the ERA-12 Based on Final Cluster Membership

Clusters significantly differed by all subscales of the ERA-12: physical ERA (Welch’s F(2, 125.56) = 101.51, p < .001); cognitive ERA (Welch’s F(2, 125.91) = 78.69, p < .001); and socioemotional ERA (Welch’s F(2, 129.68) = 238.31, p < .001). Clusters also significantly differed by socioemotional health (Welch’s F(2, 119.91) = 8.85, p < .001); cognitive ability (Welch’s F(2, 123.82) = 3.21, p = .044); and perceived decisional control in relocation (Welch’s F(2, 128.63) = 3.45, p = .035). For categorical variables, clusters significantly differed by socioeconomic status (i.e., Medicaid status; χ2(2) = 8.54, p = .014) and relocating from a local community home (χ2(2) = 7.87, p = .020). There were no other significant differences across clusters observed. Post-hoc group differences are indicated in Table 2.

Discussion

The current study used cluster analysis to identify three distinct profiles based on expectations for physical, cognitive, and socioemotional health and functioning within a sample of AL residents. Participants in the current study shared the non-normative life event of relocating to an AL, as well as the experience of congregate living with, primarily, older adults with varying health and abilities. Thus, results from the current study need to be interpreted within these unique individual and socio-environmental contexts. The two most prevalent clusters (i.e., 73% of the entire sample) were both characterized by expectations for maintaining or achieving high socioemotional health in later life, regardless of expectations for physical and/or cognitive decline. This key finding suggests that most residents in the current sample were confident that they could achieve high subjective well-being in the future. Although, we did not assess whether the current sample expected to achieve these outcomes while aging in place in their current residence.

Lower Expectations Cluster

The lower expectations cluster included participants who expected to decline in physical, cognitive, and socioemotional health and functioning. This group evidenced significantly lower socioemotional health compared to the rest of the sample. This finding is consistent with previous research with community-living older adults that suggests lower ERA is related with worse psychosocial well-being (Menkin et al., 2017; Sarkisian et al., 2005). Despite similar levels of physical health and cognitive ability, participants in the lower expectations cluster reported significantly lower physical and cognitive ERA compared to other clusters. This discrepancy may be attributed to the poorer socioemotional health observed among participants in this cluster. Consistent with previous findings (Jeste et al., 2013; Shenkin et al., 2014), poorer mental health may have helped shape health appraisals and expectations, attitudes toward aging, and perceived adaptability to life stressors.

Of note, almost 40% of the lower expectations cluster was made up of men, a higher proportion compared to other clusters. Previous research suggests that men may be particularly susceptible to internalized ageism and negative self-views after relocation to a residential long-term care setting (Moss & Moss, 2007). Even though our results do not provide insight into why men were over-represented in the lower expectations cluster, one possible explanation is that men may be more likely to report poor health expectations in the context of socioemotional distress. Future research is required to investigate possible gender differences in ERA in the AL setting.

Higher Expectations Cluster

The higher expectations cluster included residents who expected to maintain or achieve high physical, cognitive, and socioemotional health and functioning with advancing age. Therefore, participants in this group expected to achieve outcomes consistent with Rowe and Kahn’s (1997) definition of successful aging (i.e., absence of disease and disability, high cognitive and physical functioning, and retained activity engagement). Despite similar levels of physical health and cognitive ability, this cluster evidenced greater socioemotional health than the lower expectations group and significantly higher ERA in all three domains compared to other clusters. These findings further highlight the strong relationships between socioemotional health and ERA in the current sample of AL residents.

Expecting Adaptiveness Cluster

The expecting adaptiveness cluster was the largest subgroup in the current sample. Participants in this group anticipated limited socioemotional consequences from expected physical and cognitive declines, suggesting an adaptive view of aging. According to a large body of empirical and theoretical literature (Nygren et al., 2005; Ouwehand et al., 2007), older adults are likely to maintain socioemotional health in the context of declining physical and cognitive abilities. Therefore, it is possible that this cluster represent a pragmatic and accurate view of aging, especially among the current sample of AL residents whom were mostly in the oldest-old age group. The average age of the expecting adaptiveness subgroup was 85 (i.e., slightly older than the other clusters), making it likely that these participants will in fact experience some level of future physical and cognitive declines (Diehr et al., 2013). Additional work is needed to investigate the differential psychosocial, behavioral, and health outcomes of negative, positive, and realistic views of aging.

Group Differences in Relocation Factors

Group differences provide preliminary insight into the relationships between relocation-related factors and expectations for future health and functioning in AL. Participants in the expecting adaptiveness cluster evidenced greater perceived control in the decision to move to their AL compared to the lower expectations group. Although, there were no group differences in specific reasons for relocation across clusters. It is possible that, participants who perceived themselves as having control in the move believed that their AL community would be able to support their needs, allowing for high socioemotional health in the context of physical and cognitive decline. Further, consistent with Golant (2020), this group may have relocated to AL in response to disruptive life events or declining health, while, at the same time, chose to move to a community that would allow for the maintenance or promotion of socioemotional well-being. Understanding the contextual factors around relocation, and its longitudinal impacts on resident attitudes, goals, behaviors, and health is an important area of future research in the AL literature.

Targeted group differences in previous residence were observed across the clusters. Specifically, the higher expectations group consisted of more residents who move from a local home (i.e., in the same county) and fewer residents who moved from an independent senior living community, whereas the lower expectations group also had more residents who moved from a long-distance home. These findings suggest that there may be psychosocial benefits to relocating to a community that offers geographic familiarity and requires fewer demands associated with learning new environments and routines (see Golant, 2020). The relationship between relocation-related factors and psychosocial well-being in AL is often captured in qualitative research (Chen et al., 2008; Tracy & DeYoung, 2004), yet additional longitudinal quantitative studies are needed to empirically capture residents’ psychosocial trajectories based on contexts of relocation. Findings from the current study highlight the need for additional research targeting the impact of individual factors related to relocation and aging in place on psychosocial outcomes in AL.

Implications

In addition to the future research needs mentioned above, findings from the current study may inform future research targeting ERA in AL and among community-living older adults. First, future studies might benefit from considering, not just multiple dimension of ERA, but inter-relatedness across these dimensions. In addition, future studies may choose to investigate the profiles identified in the current sample across different environmental contexts, such as community residences and nursing homes. This area of study may elucidate the possible role of environment in expectations for future health and functioning in later life. Lastly, future longitudinal studies are necessary for identifying the health, behavioral, and psychological consequences of ERA profiles identified in the current study.

With additional research, assessing ERA profiles in AL may have clinical utility by identifying residents potentially at risk for negative psychosocial and behavioral outcomes, as well as for providing relevant educational resources. For example, a resident in the lower expectations group may benefit from additional screening for depression and anxiety, as well as educational materials that combat health-related age stereotypes. Findings from the current study also suggest that combating negative health-related age stereotypes in AL may benefit from highlighting the successful use of coping or adaptive compensatory strategies, rather than solely challenging stereotypes with, for example, images of remarkably healthy older adults (e.g., an older marathon runner).

Limitations

The current study has limitations. First, the sample may not be representative of an AL population, thus, limiting generalizability. Specifically, purposive recruitment and sampling, as well as inclusion based on cognitive ability, may have biased the sample toward residents with greater physical, cognitive, and socioemotional health. In addition, the ERA-12 was not previously validated with a sample of AL residents. Therefore, it is unclear if there are psychometric issues related to the use of this measure in the AL setting. Lastly, the current study did not test a complete model of ERA, which limits our ability to fully describe group differences across the identified clusters. In particular, measuring psychosocial, personality, behavioral, and health literacy factors would have allowed for a more detailed discussion of the scientific and clinical utility of clusters identified in the current study. Future research is needed to test a more complete model with a representative sample of AL residents.

Conclusion

Results from the current study add to the existing literature by offering insight into different profiles of ERA among a sample of AL residents. ERA is a multidimensional construct and this study demonstrates the potential inter-relatedness of ERA dimensions (i.e., physical, cognitive, and socioemotional expectations) in the unique context of AL. Findings also suggest that discrepancies in socioemotional health may be strongly related to AL residents’ expectations for future health and functioning, which may also serve as an intervention point. Although, additional research is needed to further investigate profiles of ERA that emerged in the current study, as well as their clinical implications.

Funding:

This work was supported in part [EP] by the National Institute on Aging [Grant No. T32AG044296].

References

  1. Andrews RM, Tan EJ, Varma VR, Rebok GW, Romani WA, Seeman TE, Gruenewald TL, Tanner EK, & Carlson MC (2017). Positive aging expectations are associated with physical activity among urban-dwelling older adults. The Gerontologist, 57(suppl_2), S178–S186. 10.1093/geront/gnx060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bäumker T, Callaghan L, Darton R, Holder J, Netten A, & Towers A-M (2012). Deciding to move into extra care housing: Residents’ views. Ageing and Society, 32(7), 1215–1245. 10.1017/S0144686X11000869 [DOI] [Google Scholar]
  3. Bekhet AK, Zauszniewski JA, & Nakhla WE (2009). Reasons for relocation to retirement communities: A qualitative study. Western Journal of Nursing Research, 31(4), 462–479. 10.1177/0193945909332009 [DOI] [PubMed] [Google Scholar]
  4. Brown Wilson K (2007). Historical evolution of assisted living in the United States, 1979 to the present. The Gerontologist, 47(suppl_1), 8–22. 10.1093/geront/47.Supplement_1.8 [DOI] [PubMed] [Google Scholar]
  5. Brownie S, Horstmanshof L, & Garbutt R (2014). Factors that impact residents’ transition and psychological adjustment to long-term aged care: A systematic literature review. International Journal of Nursing Studies, 51(12), 1654–1666. 10.1016/j.ijnurstu.2014.04.011 [DOI] [PubMed] [Google Scholar]
  6. Buschke H, Kuslansky G, Katz M, Stewart WF, Sliwinski MJ, Eckholdt HM, & Lipton RB (1999). Screening for dementia with the Memory Impairment Screen. Neurology, 52(2), 231–238. 10.1212/WNL.52.2.231 [DOI] [PubMed] [Google Scholar]
  7. Caffrey C, Sengupta M, Park-Lee E, Moss A, Rosenoff E, & Harris-Kojetin L (2012). Residents living in residential care facilities: United States, 2010. National Center for Health Statistics. https://www.cdc.gov/nchs/data/databriefs/db91.pdf [PubMed] [Google Scholar]
  8. Chen S, Brown JW, Mefford LC, Roche, A. de L, McLain AM, Haun MW, & Persell DJ. (2008). Elders’ decisions to enter assisted living facilities: A grounded theory study. Journal of Housing For the Elderly, 22(1–2), 86–103. 10.1080/02763890802097094 [DOI] [Google Scholar]
  9. Connor DJ, Seward JD, Bauer JA, Golden KS, & Salmon DP (2005). Performance of three clock scoring systems across different ranges of dementia severity. Alzheimer Disease & Associated Disorders, 19(3), 119–127. 10.1097/01.wad.0000174948.77113.a6 [DOI] [PubMed] [Google Scholar]
  10. Davidson HA, & O’Connor BP (1990). Perceived control and acceptance of the decision to enter a nursing home as predictors of adjustment. International Journal of Aging and Human Development, 31(4), 307–318. 10.2190/GBPM-H9CJ-LN22-02TT [DOI] [PubMed] [Google Scholar]
  11. Diehl M, Wahl HW, Barrett AE, Brothers AF, Miche M, Montepare JM, Westerhof GJ, & Wurm S (2014). Awareness of aging: Theoretical considerations on an emerging concept. Developmental Review, 34(2), 93–113. 10.1016/j.dr.2014.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dogra S, Al-Sahab B, Manson J, & Tamim H (2015). Aging expectations are associated with physical activity and health among older adults of low socioeconomic status. Journal of Aging and Physical Activity, 23(2), 180–186. 10.1123/japa.2012-0337 [DOI] [PubMed] [Google Scholar]
  13. Fraher A, & Coffey A (2011). Older people’s experiences of relocation to long-term care. Nursing Older People, 23(10), 23–27. 10.7748/nop2011.12.23.10.23.c8838 [DOI] [PubMed] [Google Scholar]
  14. Golant SM (2004). Do impaired older persons with health care needs occupy U.S. assisted living facilities? An analysis of six national studies. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 59(2), S68–S79. 10.1093/geronb/59.2.S68 [DOI] [PubMed] [Google Scholar]
  15. Golant SM (2020). The distance to death perceptions of older adults explain why they age in place: A theoretical examination. Journal of Aging Studies, 54, 100863. 10.1016/j.jaging.2020.100863 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Grabowski DC, Caudry DJ, Dean KM, & Stevenson DG (2015). Integrated payment and delivery models offer opportunities and challenges for residential care facilities. Health Affairs, 34(10), 1650–1656. 10.1377/hlthaff.2015.0330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hair JF Jr., Black WC, Babin BJ, & Anderson RE (2014). Cluster analysis. In Hair JF Jr., Black WC, Babin BJ, & Anderson RE (Eds.), Multivariate data analysis (7th ed., pp. 341–414). Pearson. [Google Scholar]
  18. Harris-Kojetin L, Sengupta M, Lendon JP, Rome V, Valverde R, & Caffrey C (2019). Long-term care providers and services users in the United States, 2015–2016. National Center for Health Statistics. https://www.cdc.gov/nchs/data/series/sr_03/sr03_43-508.pdf [PubMed] [Google Scholar]
  19. Hays RD, Bjorner JB, Revicki DA, Spritzer KL, & Cella D (2009). Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items. Quality of Life Research, 18(7), 873–880. doi: 10.1007/s11136-009-9496-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jang Y, Bergman E, Schonfeld L, & Molinari V (2006). Depressive symptoms among older residents in assisted living facilities. International Journal of Aging & Human Development, 63(4), 299–315. 10.2190/7TTA-DQWR-T429-C4N4 [DOI] [PubMed] [Google Scholar]
  21. Jeste DV, Savla GN, Thompson WK, Vahia IV, Glorioso DK, Martin AS, Palmer BW, Rock D, Golshan S, Kraemer HC, & Depp CA (2013). Association between older age and more successful aging: critical role of resilience and depression. American Journal of Psychiatry, 170(2), 188–196. 10.1176/appi.ajp.2012.12030386 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jungers CM (2010). Leaving home: An examination of late-life relocation among older adults. Journal of Counseling & Development, 88(4), 416–423. 10.1002/j.1556-6678.2010.tb00041.x [DOI] [Google Scholar]
  23. Kane RA, Chan J, & Kane RL (2007). Assisted living literature through May 2004: Taking stock. The Gerontologist, 47(suppl_1), 125–140. 10.1093/geront/47.Supplement_1.125 [DOI] [PubMed] [Google Scholar]
  24. Khatutsky G, Ormond C, Wiener JM, Greene AM, Johnson R, Jessup EA, Vreeland E, Sengupta M, Caffrey C, & Harris-Kojetin L (2016). Residential care communities and their residents in 2010: A national portrait. National Center for Health Statistics. https://www.cdc.gov/nchs/data/nsrcf/nsrcf_chartbook.pdf [Google Scholar]
  25. Kim HY (2013). Statistical notes for clinical researchers: Assessing normal distribution using skewness and kurtosis. Restorative Dentistry & Endodontics, 38(1), 52–54. 10.5395/rde.2013.38.1.52 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kim SH (2009). Older people’s expectations regarding ageing, health-promoting behaviour and health status. Journal of Advanced Nursing, 65(1), 84–91. 10.1111/j.1365-2648.2008.04841.x [DOI] [PubMed] [Google Scholar]
  27. Lawton MP (1975). The Philadelphia Geriatric Center Morale Scale: A revision. Journal of Gerontology, 30(1), 85–89. 10.1093/geronj/30.1.85 [DOI] [PubMed] [Google Scholar]
  28. Levy BR, Pilver CE, & Pietrzak RH (2014). Lower prevalence of psychiatric conditions when negative age stereotypes are resisted. Social Science & Medicine, 119, 170–174. 10.1016/j.socscimed.2014.06.046 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McHugh ML (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276–282. doi: 10.11613/BM.2012.031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Meisner BA, & Baker J (2013). An exploratory analysis of aging expectations and health care behavior among aging adults. Psychology and Aging, 28(1), 99–104. 10.1037/a0029295 [DOI] [PubMed] [Google Scholar]
  31. Meisner BA, Weir PL, & Baker J (2013). The relationship between aging expectations and various modes of physical activity among aging adults. Psychology of Sport and Exercise, 14(4), 569–576. 10.1016/j.psychsport.2013.02.007 [DOI] [Google Scholar]
  32. Menkin JA, Robles TF, Gruenewald TL, Tanner EK, & Seeman TE (2017). Positive expectations regarding aging linked to more new friends in later life. The Journals of Gerontology: Series B, 72(5), 771–781. 10.1093/geronb/gbv118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Milligan GW (1980). An examination of the effect of six types of error perturbation on fifteen clustering algorithms. Psychometrika, 45(3), 325–342. 10.1007/BF02293907 [DOI] [Google Scholar]
  34. Mitty E (2010). An assisted living community environment that optimizes function: housing enabler assessment. Geriatric Nursing, 31(6), 448–451. 10.1016/j.gerinurse.2010.10.004 [DOI] [PubMed] [Google Scholar]
  35. Moss SZ, & Moss MS (2007). Being a man in long term care. Journal of Aging Studies, 21(1), 43–54. 10.1016/j.jaging.2006.05.001 [DOI] [Google Scholar]
  36. Tracy JP & DeYoung S (2004). Moving to an assisted living facility: Exploring the transitional experience of elderly individuals. Journal of Gerontological Nursing, 30(10), 26–33. doi: 10.3928/0098-9134-20041001-09. [DOI] [PubMed] [Google Scholar]
  37. Norusis MJ (2011). Cluster analysis. In Norusis MJ (Ed.) IBM SPSS statistics 19 advanced statistical procedures companion (pp. 375–404). Prentice Hall. [Google Scholar]
  38. Nygren B, Aléx L, Jonsén E, Gustafson Y, Norberg A, & Lundman B (2005). Resilience, sense of coherence, purpose in life and self-transcendence in relation to perceived physical and mental health among the oldest old. Aging & Mental Health, 9(4), 354–362. 10.1080/1360500114415 [DOI] [PubMed] [Google Scholar]
  39. Ouwehand C, de Ridder DTD, & Bensing JM (2007). A review of successful aging models: Proposing proactive coping as an important additional strategy. Clinical Psychology Review, 27(8), 873–884. 10.1016/j.cpr.2006.11.003 [DOI] [PubMed] [Google Scholar]
  40. Pallant JF (2013). SPSS survival manual: A step by step guide to data analysis using SPSS for Windows (5th ed.). Open University Press. [Google Scholar]
  41. Park NS, Jang Y, Lee BS, Chiriboga DA, & Molinari V (2015). Correlates of attitudes toward personal aging in older assisted living residents. Journal of Gerontological Social Work, 58(3), 232–252. 10.1080/01634372.2014.978926 [DOI] [PubMed] [Google Scholar]
  42. Pirhonen J, Tiilikainen E, & Pietilä I (2018). Ruptures of affiliation: Social isolation in assisted living for older people. Ageing and Society, 38(9), 1868–1886. doi: 10.1017/S0144686X17000289 [DOI] [Google Scholar]
  43. Plys E, & Qualls SH (2019). Programmed activity attendance in assisted living: An application of the theory of planned behavior with additional health factors. Clinical Gerontologist. Advance online publication. 10.1080/07317115.2019.1645781 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Polivka L & Salmon JR (2011). The ideal assisted living alternative: What is should be and why. Florida Department of Elder Affairs. http://elderaffairs.state.fl.us/faal/documents/Chapter1_The_Ideal_Assisted_Living_Alternative_What_it_should_be_and_Why.pdf [Google Scholar]
  45. Pomeroy SH, Scherer Y, Runkawatt V, Iamsumang W, Lindemann J, & Resnick B (2011). Person–environment fit and functioning among older adults in a long-term care setting. Geriatric Nursing, 32(5), 368–378. 10.1016/j.gerinurse.2011.07.002 [DOI] [PubMed] [Google Scholar]
  46. Resnick B, Galik E, Gruber-Baldini AL, & Zimmerman S (2010). Perceptions and performance of function and physical activity in assisted living communities. Journal of the American Medical Directors Association, 11(6), 406–414. 10.1016/j.jamda.2010.02.003 [DOI] [PubMed] [Google Scholar]
  47. Rouleau I, Salmon DP, Butters N, Kennedy C, & McGuire K (1992). Quantitative and qualitative analyses of clock drawings in Alzheimer’s and Huntington’s disease. Brain and Cognition, 18(1), 70–87. 10.1016/0278-2626(92)90112-Y [DOI] [PubMed] [Google Scholar]
  48. Rowe JW, & Kahn RL (1997). Successful aging. The Gerontologist, 37(4), 433–440. 10.1093/geront/37.4.433 [DOI] [PubMed] [Google Scholar]
  49. Sambandam R (2003). Cluster analysis gets complicated. Marketing Research, 15(1), 16–21. [Google Scholar]
  50. Sarkisian CA, Steers WN, Hays RD, & Mangione CM (2005). Development of the 12-Item Expectations Regarding Aging Survey. The Gerontologist, 45(2), 240–248. 10.1093/geront/45.2.240 [DOI] [PubMed] [Google Scholar]
  51. Sarkisian CA, Hays RD, Berry SH, & Mangione CM (2001). Expectations regarding aging among older adults and physicians who care for older adults. Medical Care, 39(9), 1025–1036. 10.1097/00005650-200109000-00012 [DOI] [PubMed] [Google Scholar]
  52. Sarkisian CA, Hays RD, Berry S, & Mangione CM (2002). Development, reliability, and validity of the Expectations Regarding Aging (ERA-38) Survey. The Gerontologist, 42(4), 534–542. 10.1093/geront/42.4.534 [DOI] [PubMed] [Google Scholar]
  53. Seifert A, & Schelling HR (2018). Attitudes toward aging and retirement homes before and after entry into a retirement home. Journal of Housing For the Elderly, 32(1), 12–25. 10.1080/02763893.2017.1393484 [DOI] [Google Scholar]
  54. Shenkin SD, Laidlaw K, Allerhand M, Mead GE, Starr JM, & Deary IJ (2014). Life course influences of physical and cognitive function and personality on attitudes to aging in the Lothian Birth Cohort 1936. International Psychogeriatrics, 1–14. 10.1017/S1041610214000301 [DOI] [PubMed] [Google Scholar]
  55. Tavakol M, & Dennick R (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53–55. 10.5116/ijme.4dfb.8dfd [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Tracy JP, & DeYoung S (2004). Moving to an assisted living facility: Exploring the transitional experience of elderly individuals. Journal of Gerontological Nursing, 30(10), 26–33. 10.3928/0098-9134-20041001-09 [DOI] [PubMed] [Google Scholar]
  57. Yim O, & Ramdeen KT (2015). Hierarchical cluster analysis: comparison of three linkage measures and application to psychological data. Quantitative Methods for Psychology, 11(1), 8–21. 10.20982/tqmp.11.1.p008 [DOI] [Google Scholar]
  58. Zakharov K (2016). Application of k-means clustering in psychological studies. Quantitative Methods for Psychology, 12(2), 87–100. 10.20982/tqmp.12.2.p087 [DOI] [Google Scholar]
  59. Zimmerman S, Sloane PD, & Reed D (2014). Dementia prevalence and care in assisted living. Health Affairs, 33(4), 658–666. 10.1377/hlthaff.2013.1255 [DOI] [PubMed] [Google Scholar]

RESOURCES