Research in Gerontological Nursing

Empirical Research 

Effects of Physical Environment on Health and Behaviors of Residents With Dementia in Long-Term Care Facilities: A Longitudinal Study

Sook Young Lee, PhD; Habib Chaudhury, PhD; Lillian Hung, RN, MA

Abstract

The challenges in investigating the effects of the physical environment on residents with dementia include having a sample of comparable study groups and a lack of long-term follow-up evaluation. The current study attempted to address these two challenges by carefully matching residents and analyzing long-term measurement data. The aim of the study was to examine whether residents with dementia (N = 12) living in a traditional large-scale setting or a small-scale, home-like setting exhibit any difference in health and behaviors over time. Physical environmental assessment of the two care facilities was conducted using the Therapeutic Environment Screening Survey for Nursing Homes. Residents' behavioral assessments were performed using three tools at three assessments over a period of 1 year: (a) Multidimensional Observation Scale for Elderly Subjects, (b) Minimum Data Set, and (c) Dementia Care Mapping. The results suggest that older adults with dementia can have increased social interaction and engagement with the support of an optimal physical environment.

[Res Gerontol Nurs. 2016; 9(2):81–91.]

Abstract

The challenges in investigating the effects of the physical environment on residents with dementia include having a sample of comparable study groups and a lack of long-term follow-up evaluation. The current study attempted to address these two challenges by carefully matching residents and analyzing long-term measurement data. The aim of the study was to examine whether residents with dementia (N = 12) living in a traditional large-scale setting or a small-scale, home-like setting exhibit any difference in health and behaviors over time. Physical environmental assessment of the two care facilities was conducted using the Therapeutic Environment Screening Survey for Nursing Homes. Residents' behavioral assessments were performed using three tools at three assessments over a period of 1 year: (a) Multidimensional Observation Scale for Elderly Subjects, (b) Minimum Data Set, and (c) Dementia Care Mapping. The results suggest that older adults with dementia can have increased social interaction and engagement with the support of an optimal physical environment.

[Res Gerontol Nurs. 2016; 9(2):81–91.]

Currently, 747,000 Canadians are living with Alzheimer's disease and other dementias and this number will increase to 1.4 million by 2031 (Alzheimer Society, 2014). With the aging population and growing number of individuals living with dementia, the long-term care support for this population is a critical component in health care services for older adults. Dementia is one of the most common health conditions in Canadian nursing homes, with three in five residents having the diagnosis in their health records (Canadian Institute for Health Information [CIHI], 2013). To meet the population demand and improve quality of care, the culture of caregiving for older adults with dementia in western countries has been increasingly shifting away from the traditional large-scale model to the creation of smaller home-like environments for more familiar and non-institutional living (Kane, Lum, Cutler, Degenholtz, & Yu, 2007; Leibrock, 2000; Regnier, 2002; Van Amerongen-Heijer, 2013). Increasing evidence suggests that the design of the physical environment plays an important role in affecting health and behaviors of older adults with dementia residing in care facilities (Chaudhury & Cooke, 2014; Chaudhury, Hung, & Badger, 2013; Lee & Dilani, 2011; Schwarz, Chaudhury, & Tofle, 2004). The ecological environmental theory explains that as older individuals face physical, mental, and functional decline, they are more sensitive and vulnerable to environmental demands (Lawton & Nahemow, 1973). Therefore, modifying the environment to respond to the needs and capacity of older adults with dementia is a way to support everyday coping and adaptation. According to Lawton and Nahemow's (1973) ecological theory, either over- or under-stimulation can cause maladaptive behaviors as a function of the interaction of individual factors with physical and social environments.

Environmental gerontologists believe supportive environmental design can serve as an effective therapeutic resource to promote health and quality of life of residents in care facilities (Briller, Calkins, Marsden, Proffitt, & Perez, 2001; Calkins, 2013; Cutler, 2007). It is suggested that a care setting with a familiar home-like scale, as Marsden (2005) calls it human scale, which relates to the experience of home living, is likely to be more manageable for older adults with impaired physical and/or cognitive functioning. Further, a small-scale care setting may make it easier for staff to become familiar with a smaller number of residents in a more meaningful way, develop relationships, and provide more person-centered care.

There is growing support for creating smaller home-like environments for older adults with dementia. Recent Canadian studies reported that staff working in a smaller unit were able to provide care in a more person-centered manner, and thereby had positive effects on preserving personhood of residents with dementia (Hung & Chaudhury, 2011; Lee, Chaudhury, & Hung, 2014). In addition, residents living in small and home-like facilities in studies conducted in North America showed slower decline in mobility (Reimer, Slaughter, Donaldson, Currie, & Eliasziw, 2004; Saxton, Silverman, Ricci, Keane, & Deeley, 1998). Conversely, other studies found that small-scale units compared to traditional large-scale settings showed no effects in enhancing quality of life (de Rooij et al., 2012; Verbeek et al., 2010). Zeisel et al. (2003) found that residents demonstrated less social withdrawal behaviors in the larger units. Overall, the current evidence indicating the specific effects of small-scale environments on residents remains limited, unclear, and inconsistent (Verbeek et al., 2014).

One challenge in investigating the comparative effect of different physical environments on residents with dementia is creation of a sample of comparable resident groups in multiple study sites. Differences in characteristics and levels of dementia among residents can influence outcome measures, such as physical deteriorations and behavioral changes. These differences are likely reasons why previous research has been limited in producing evidence to identify beneficial effects of smaller units (Verbeek et al., 2014). A lack of long-term follow-up evaluation is another weakness of existing research (Day, Carreon, & Stump, 2000). The current study attempted to address these two challenges by carefully matching residents between groups in sample selection and including three time points of evaluation in the research design. The main aim of the current study was to examine whether residents with dementia living in different physical environments (i.e., a traditional large-scale institutional facility and a small-scale, home-like unit) exhibit any difference in health and behavioral outcomes. It was hypothesized that residents with dementia residing in the small-scale, home-like care unit would demonstrate a more positive mood, higher levels of social engagement, higher levels of physical functioning, and better health status compared with those residents with dementia living in the traditional large-scale facility.

Method

Settings

Two dementia long-term care facilities were selected as study sites using a mixed-sampling strategy (i.e., random and purposive sampling). In the first phase, 10 dementia care facilities were selected randomly from a total of 56 residential care facilities in Vancouver Coastal Health Authority, British Columbia, Canada. In the second phase, to ensure adequate representation of the variability in physical environmental characteristics (i.e., a small-scale care unit and an institutional large-scale care unit), three sampling criteria were used: (a) number of residents in a unit (>20 residents as a large unit), (b) length of corridor (>15 m as a long corridor), and (c) building layout (double-loaded floor plan [i.e., an architectural layout that has a corridor or hallway with resident rooms on both sides, contrasted with a single-loaded floor plan where rooms are on one side of the corridor] as an institutional unit). Through site visits, two dementia care units (Richmond Manor and Maple Manor) were selected from the 10 care facilities based on the highest variation in the environmental features from the three criteria. Richmond Manor is a purpose-built dementia care facility, with 12 residents with dementia on each unit, a relatively short corridor (approximately 14 m), all single bedrooms, and a single-loaded floor plan. The number of staff working in the daytime was 1.5 nurses and two care aides. Maple Manor is an institutional care setting, with 30 residents with dementia on each unit, a long corridor (approximately 40 m), mixed single/semi-private bedrooms, and a double-loaded floor plan. One nurse and four care aides worked with 30 residents during daytime hours. The two units were segregated from other sections of the facility with doors that could be unlocked only by using a keypad.

Ethical approval and permission to conduct the study was obtained from the Office of Research Ethics at Simon Fraser University, Vancouver Coastal Health Authority and administrative staff members in participating facilities. Written consent was obtained from all participants' family members or legal guardians.

Study Population

Residents eligible for inclusion in the current study were 60 and older, in the early or middle stage of Alzheimer's disease or a related dementia, and able to ambulate with or without an assistive device. To obtain rich data from observation of social behaviors of residents, the study excluded residents who were bed-bound or staying in their private rooms during the daytime. Seventeen residents from two facilities met the study criteria. During the study period, three residents passed away and two residents moved to another care facility or hospital. The final data analyses involved 12 residents: seven from Richmond Manor and five from Maple Manor. Mean age of the residents was 82.9 (SD = 8.9 years) in Richmond Manor and 77.6 (SD = 9.8 years) in Maple Manor. There were three women and four men in Richmond Manor and three women and two men in Maple Manor. Mean lengths of stay were 28.9 months (SD = 20.4 months, range = 12 to 72 months) and 29 months (SD = 10.2 months, range = 13 to 39 months) in Richmond Manor and Maple Manor, respectively. No statistically significant differences were noted between groups in age, gender, or length of stay.

Measurement and Data Collection

Physical Environmental Assessment. Physical environmental characteristics of the two care units were assessed using the Therapeutic Environment Screening Survey for Nursing Homes (TESS-NH; Sloane et al., 2002) by the principal investigator and a research assistant (S.Y.L., H.C.). The care units were assessed on February 21 and 23, 2012, respectively. Interrater reliability for 74% of items on the TESS-NH exceeds 0.60 (Sloane et al., 2002). The TESS-NH includes six domains with 13 sub-domains and one global rating: (a) privacy control/autonomy: unit autonomy, outdoor access, and privacy; (b) safety/security/health: exit control, maintenance, cleanliness, and safety; (c) stimulation: lighting, visual and tactile stimulation, and noise; (d) socialization: space and seating; (e) personalization: familiarity and home-likeness; and (f) orientation: orientation and cueing. Specific physical features were mostly rated on a scale from 0 = distinctly unpleasant to 3 = more favorable.

Residents' Behavioral Assessment. The Multidimensional Observation Scale for Elderly Subjects (MOSES; Helmes, Csapo, & Short, 1987), Minimum Data Set (MDS; CIHI, 2012), and Dementia Care Mapping (DCM; Brooker & Surr, 2005) were used to assess changes in residents' functioning and cognitive status and social behaviors over time. The three specific assessment data were collected in the two care units every 4 months over 1 year: data on the first assessment (T1) were collected from March to April 2012, the second assessment (T2) in October 2012, and the third assessment (T3) from March to April 2013.

The MOSES (Helmes et al., 1987) is a compilation of five domains of behavior with 40 items: (a) self-care function: dressing, bathing, grooming, etc.; (b) disorientation: recognizing staff, awareness of place/time, memory for recent events, etc.; (c) depression: looking sad/depressed, reporting worry and anxiety, etc.; (d) irritability: cooperation with nursing care, reactions to frustration, verbal/physical abuse of others, etc.; and (e) social withdrawal: preferring solitude, responding to social contacts, interest in outside events, etc. Interrater reliability for 80% of items on the MOSES exceeds 0.70 (Helmes et al., 1987). Each item is rated on a 4- or 5-point severity scale, where 5 represents positive to great degree and 1 represents negative to great degree. Care aides familiar with the selected residents were asked to answer the MOSES items at each assessment.

The MDS 2.0 is a comprehensive instrument used to assess residents with dementia in long-term care facilities. Assessment with the standardized tool enables detection of residents' functions, needs, and latent risks to inform individualized care planning and monitoring. The MDS used in the current study contains 12 domains with 132 items: (a) cognitive patterns, (b) communication patterns, (c) mood/behavior patterns, (d) psychosocial well-being, (e) physical functioning and structural problems, (f) continence, (g) disease diagnoses, (h) health conditions, (i) oral status, (j) skin condition, (k) activity pursuit patterns, and (l) medications. These data were obtained from an RN at each care facility.

DCM (Brooker & Surr, 2005), a mapping method grounded in a person-centered approach to dementia care, is an observational tool devised to methodically document in detail the experience of individuals with dementia. DCM was conducted by a mapper, who had completed the course “Learning to use DCM” conducted by Jentle Harts Consulting (access http://netkraft.com/jh2012). The mapper conducted unobtrusive observations of the selected residents continuously at 5-minute intervals during the day for an average of 10 hours at each assessment. Observations were performed over 2 or 3 days at each assessment to observe each resident for at least 3 hours. The mapper tracked up to four residents simultaneously in the communal areas (Figure 1) where the selected residents were present most of the time. After each time period, Behavior Category Codes (BCCs) and Mood/Engagement (ME) values were recorded for each selected resident being observed. The BCCs describe one of 23 different categories of resident behavior that has occurred. Further details of the 23 categories are provided elsewhere (Lee, Chaudhury, & Lee, 2014). ME values are always recorded in the context of the BCC that they accompany (i.e., A/+1 or L/+1) and expressed on a 6-point scale with odd numbers (i.e., +5, +3, +1, −1, −3, −5), ranging from extreme positive state to extreme negative state. The ME values and BCCs offer an index of well-being for a specific time period for residents.


Primary communal spaces in Richmond Manor and Maple Manor.

Figure 1.

Primary communal spaces in Richmond Manor and Maple Manor.

Focus group interviews with staff members of the care units were conducted to collect qualitative data with open-ended questions. The findings from the focus group interviews were published in a separate article (Lee, Chaudhury, & Hung, 2014).

Data Analysis

Data were coded and analyzed using Microsoft® Excel® (2010 version) and SPSS version 20.0. Descriptive statistics were applied using Excel. Mean comparisons between the two units' environmental assessment were performed using t test in SPSS with a level of significance of <0.05. The general linear model (GLM) repeated measures analysis of variance (ANOVA) test was used to compare the study groups across the three assessments. The level of significance was set at <0.05. Functional and behavioral changes in resident measures were assessed across the three measurements in the units. For between-unit comparison, outcomes of the small group unit residents were compared with residents in the institutional care unit.

Results

Physical Environmental Assessment

Table 1 shows means and results of t test analysis on the physical environmental assessment of the two care units. Significant differences were found in stimulation (t = 2.19, p < 0.05) and personalization/familiarity/home-likeness (t = 4.43, p < 0.01). The domains of privacy/control/autonomy, safety/security/health, socialization, and orientation revealed no statistically significant differences between the two care units. Further details of sub-domain are provided elsewhere (Lee, Chaudhury, & Lee, 2014). Global rating based on TESS-NH was 7.0 in Richmond Manor and 3.5 in Maple Manor. The results indicate that Richmond Manor as a small-scale care unit had more positive features in design, including quality of lighting and visual, tactile, and acoustic stimuli and offered a more home-like atmosphere to residents with dementia as compared to the physical environment of Maple Manor.


Means and Results of t Test Analysis on the Therapeutic Environment Screening Survey for Nursing Homes

Table 1:

Means and Results of t Test Analysis on the Therapeutic Environment Screening Survey for Nursing Homes

Participating Residents' Behavioral Assessment

Differences of within- and between-unit comparisons for the residents' MOSES assessment were analyzed using the GLM repeated measures ANOVA test. Repeated measures ANOVA for the within-subjects factor of time revealed that the mean scores for self-care functioning (F2,10 = 5.93, p < 0.01), disoriented behavior (F2,10 = 4.51, p < 0.05), and withdrawn behavior (F2,10 = 6.72, p < 0.01) were statistically significantly different between the three assessments (Figure 2). The mean scores of self-care functioning at the three different assessments were 19.1 (SD = 6.1), 17.7 (SD = 6.4), and 15.9 (SD = 4.5) in Richmond Manor residents and 21.8 (SD = 6.0), 19 (SD = 7.4), and 17.4 (SD = 9.8) in Maple Manor residents. These scores indicate that the self-care functioning of residents in each group deteriorated significantly over the assessment period. Regarding disoriented behavior, the mean scores at the three assessments were 20 (SD = 6.7), 20.1 (SD = 8.8), and 18.3 (SD = 6.2) in Richmond Manor residents and 21.6 (SD = 7.4), 20.6 (SD = 7.3), and 15.6 (SD = 6.9) in Maple Manor residents. Residents' disoriented behavior also declined significantly over time for the two groups; residents' disoriented behavior in Maple Manor rapidly deteriorated at T3. For withdrawn behavior, the mean scores at the three assessments were 23 (SD = 8), 21.1 (SD = 7.5), and 21.4 (SD = 8.9) in Richmond Manor residents and 19.8 (SD = 3.2), 15.2 (SD = 5.1), and 14.2 (SD = 4.2) in Maple Manor residents. These scores demonstrated that the social interaction of the residents in Maple Manor had generally declined, especially from T1 to T2. However, the differences in mean scores for depressed/anxious mood and irritable behavior were not statistically significant according to the statistical analysis for the within-subjects factor of time.


Results of the repeated measures analysis of variance on the Multidimensional Observation Scale for Elderly Subjects. A. Self-care functioning: F2,10 = 5.93**; B. Disoriented behavior: F2,10 = 4.51*; C. Irritable behavior: F2,10 = 4.76†; D. Withdrawn behavior: F2,10 = 6.72**. *p < 0.05; **p < 0.01, within-unit comparisons; †p < 0.05, between-unit comparisons.

Figure 2.

Results of the repeated measures analysis of variance on the Multidimensional Observation Scale for Elderly Subjects. A. Self-care functioning: F2,10 = 5.93**; B. Disoriented behavior: F2,10 = 4.51*; C. Irritable behavior: F2,10 = 4.76; D. Withdrawn behavior: F2,10 = 6.72**. *p < 0.05; **p < 0.01, within-unit comparisons; p < 0.05, between-unit comparisons.

The results of the repeated measures ANOVA for the between-subjects factor of group revealed that the two groups differed significantly across three assessments regarding the dimension of the irritable behavior (F2,10 = 4.76, p < 0.05) (Figure 2). At the beginning of the assessment (T1), the means of irritable behavior of the two groups were similar: 28.9 (SD = 7.3) in Richmond Manor residents and 29.4 (SD = 3.8) in Maple Manor residents. However, the levels of irritable behavior in Maple Manor residents at T2 (mean [SD] = 22.8 [6.4]) and T3 (mean [SD] = 20.8 [6.8]) were significantly lower than ones in Richmond Manor residents (mean [SD] = 30 [5], 31 [7.8], respectively), indicating that residents in Maple Manor on average had shown more irritable behavior compared to residents in Richmond Manor at T2 and T3. The differences in mean scores for self-care functioning, disoriented behavior, depressed/anxious mood, and withdrawn behavior were not statistically significant according to the statistical analysis for the between-subjects factor of group.

Differences of within- and between-unit comparisons for the residents' MDS assessments were analyzed using the GLM repeated measures ANOVA test. Repeated measures ANOVA for the within-subjects factor of time demonstrated that the differences in mean scores for physical functioning II (F2,10 = 6.27, p < 0.01) were statistically significant among the three assessments (Figure 3). The mean scores of the physical functioning II were 50.1 (SD = 3.4), 49.1 (SD = 3.9), and 46.6 (SD = 3.8) in Richmond Manor residents and 50.4 (SD = 2.6), 50 (SD = 2.6), and 48.6 (SD = 4) in Maple Manor residents. These scores indicated the physical functioning of residents, such as balance while standing/sitting, modes of locomotion, and modes of transfer in each group, deteriorated significantly, especially at T3.


Results of the repeated measures analysis of variance on the Minimum Data Set. A. Physical functioning II (i.e., bathing, balance while standing and sitting, and modes of locomotion and transfer): F2,10 = 6.27**; B. Oral/nutrition status: F2,10 = 11.75‡; C. Skin condition: F2,10 = 16.66‡. **p < 0.01, within-unit comparisons; ‡p < 0.01, between-unit comparisons.

Figure 3.

Results of the repeated measures analysis of variance on the Minimum Data Set. A. Physical functioning II (i.e., bathing, balance while standing and sitting, and modes of locomotion and transfer): F2,10 = 6.27**; B. Oral/nutrition status: F2,10 = 11.75; C. Skin condition: F2,10 = 16.66. **p < 0.01, within-unit comparisons; p < 0.01, between-unit comparisons.

Furthermore, the analysis for the between-subjects factor of group determined that the two groups differed significantly regarding oral/nutrition status (F2,10 = 11.75, p < 0.01) and skin conditions (F2,10 = 16.66, p < 0.01) (Figure 3). No significant differences were noted between groups or across time in other variables related to health status (i.e., fluid, dizziness, fever, pain). For oral/nutrition status, such as chewing/swallowing and nutritional problems, the residents in Maple Manor (T1∼T3 mean [SD] = 17.0 [0.7], 17 [0.7], 16.6 [0.5]) showed significantly lower scores than residents in Richmond Manor (T1∼T3 mean [SD] = 18, 17.9 [0.4], 17.3 [0.8]). Regarding skin conditions (e.g., ulcers, lesions), the residents in Richmond Manor (T1∼T3 mean [SD] = 7.7 [1], 7.9 [0.7], 8.1 [0.4]) had significantly lower scores compared to residents in Maple Manor (T1∼T3 mean [SD] = 8.4 [0.5], 9, 9). These scores indicate that residents in Maple Manor had fewer skin problems compared to residents in Richmond Manor across the three time points.

The distribution of the behavior category profiles across the three assessments for the residents in the study groups are shown in Table 2. The most frequent behaviors observed throughout T1∼T3 in Richmond Manor were articulation (interacting with others) (18.7%, 23.4%, and 14.9%, respectively), food (eating or drinking) (17.4%, 15.6%, and 17.8%, respectively), nod (sleeping or dozing) (16.3%, 11.2%, and 15.7%, respectively), and borderline (passively watching) (11.9%, 14.3%, and 19.9%, respectively). In Maple Manor, cool behavior (withdrawn) (42.1%, 8.6%, and 6%, respectively), borderline (passively watching) (15.9%, 33.2%, and 30.3%, respectively), nod (sleeping or dozing) (13.8%, 16%, and 19.9%, respectively), and food (eating or drinking) (12.6%, 16.3%, and 14.5%, respectively) were noted most during observations. Distinct differences were noted in the cool behavior category at T1 and borderline behavior category at T2 and T3 between the two groups. According to these data, residents in Maple Manor showed poor social engagement compared with residents in Richmond Manor.


Distribution of Dementia Care Mapping Data
Distribution of Dementia Care Mapping Data

Table 2:

Distribution of Dementia Care Mapping Data

ME values across the three assessments are also shown in Table 2. Residents in Richmond Manor had higher scores on the mean throughout the measurements than the scores in Maple Manor. Positive mood or engagement values (≥+3) were 25.2%, 14.4%, and 5.7%, respectively, at each assessment in Richmond Manor and 4.2%, 2.2%, and 4.2%, respectively, in Maple Manor. These results show that the residents in Richmond Manor on average were in better moods and had higher engagement with their surroundings throughout observations across the three assessments compared with residents in Maple Manor.

Discussion

The current study investigated the effects of the physical environment on residents with early to moderate stage dementia. The longitudinal results show that residents in Maple Manor (i.e., the traditional large-scale unit) compared to Richmond Manor (i.e., the small-scale, home-like unit) had significant deterioration (between groups by time) in irritable behaviors and oral and nutrition status. For irritable behavior, residents in Maple Manor had severe deteriorations at T2 and T3, whereas residents in Richmond Manor showed continuous improvement at T2 and T3. This finding is congruent with those of other studies (Cioffi, Fleming, Wilkes, Sinfield, & Le Miere, 2007; Lee & Dilani, 2011) that showed a home-like environmental setting played a role in influencing positive behaviors and well-being of residents with dementia.

Related to health status, the residents in Richmond Manor showed better outcomes in oral and nutrition conditions at all three measurements of evaluation compared to those in Maple Manor. These outcomes may suggest that a smaller, home-like environment supports eating and drinking, makes dining more enjoyable, and thereby promotes favorable nutrition outcomes. For other factors related to health status, except skin problems, no significant differences were noted between and within groups. The change of physical functioning over time between groups was not significant, but the decline of physical functioning and self-care was significant within-group. The within-group change over time demonstrates the progression of disease process in dementia and general deterioration of health conditions in participants.

In the DCM observations, the residents in Maple Manor spent less time in social interaction compared to the residents in Richmond Manor across the three assessment points. In addition, residents in Richmond Manor were content or very happy during one quarter of the observation time in T1, whereas residents in Maple Manor were content in only 4% of the observation time. This finding is similar with the outcome of withdrawn behavior in the MOSES assessment, which demonstrated that the residents in Maple Manor were more withdrawn than residents in Richmond Manor. Taken together, these results suggest that older adults with early to moderate stage dementia can be supported to become socially active and engaged with others in an optimal environment. It is possible that a small-scale, home-like unit makes it easier to socially relate with others, as fewer people in the setting are not as overwhelming or over-stimulating. A smaller home-like environment may also offer a sense of comfort, security, and belonging. This finding is similar to the results of previously conducted studies (Cohen-Mansfield, Thein, Dakheel-Ali, & Marx, 2010; Smit, de Lange, Willemse, & Pot, 2012; Verbeek et al., 2014) that demonstrated positive effects of small-scale care facilities on social engagement/involvement.

Limitations

The current study has a few limitations that need to be noted. Although residents were carefully selected to make comparable groups, it was not possible to match all health conditions and risk factors (e.g., fall risk, medications, wounds, social support, coping abilities), and these health conditions could have influenced the change in outcomes in health and behaviors. Another limitation was the inability to record baseline measurements for residents' health and behavior within a relatively short period of being admitted to the facility. In addition, differences in organizational philosophy, team culture, staff attitude, and staffing ratio may have contributed to changes in outcome measurements. Environmental assessment of the two selected care facilities was conducted with the TESS-NH tool, which generates an overall evaluation of public spaces in the setting, but does not provide focused and in-depth data of the social spaces. A detailed analysis of environmental variability of the social spaces could not be collected, as a validated environmental assessment tool specific to social spaces in care facilities does not exist. Thus, the current study does not claim that the results on resident outcomes can be linked with specific physical environmental features per se. Finally, there is no claim in the current study that the participating care facilities were representative of Canadian residential care facilities for the care of older adults. The small sample of residents also limits detection of clinical effects and generalizability of the findings.

Conclusion

The current study examined the influence of physical environmental characteristics on health and behavior of residents in long-term care facilities. Specifically, the study investigated the effects of a small-scale care facility compared to a traditional institutional care facility. It was hypothesized that residents with dementia residing in the small-scale, home-like care unit would demonstrate a higher level of positive mood, social engagement, physical functioning, and better health status compared with residents with dementia living in the traditional large-scale unit. Taken as a whole, findings of the current study indicate that small-scale facilities have positive effects on health and behavior of residents in long-term care facilities. The unique characteristics of the current study include examination of the relationship between environmental factors and resident outcomes by using a longitudinal research design with comparison groups and multiple validated measures. Future research should examine the effect of specific physical environmental features on residents' behavioral outcomes. In addition, qualitative research can be used to investigate the interaction of physical, social environment, and organizational factors and may offer insights into the role of environmental factors in facilitating staff to become more effective in providing good care.

References

  • Alzheimer Society. (2014). Dementia numbers in Canada. Retrieved from http://www.alzheimer.ca/en/About-dementia/What-is-dementia/Dementia-numbers
  • Briller, S.H., Calkins, M.P., Marsden, J.P., Proffitt, M.A. & Perez, K. (2001). Creating successful dementia care settings: Maximizing cognitive and functional abilities. Baltimore, MD: Health Professions Press.
  • Brooker, D. & Surr, C. (2005). Dementia care mapping: Principles and practice. Bradford, UK: University of Bradford.
  • Calkins, M.P. (2013). Senior environments: When we become them. Health Environments Research and Design, 6(2), 3–6. doi:10.1177/193758671300600201 [CrossRef]
  • Canadian Institute for Health Information. (2012). Resident Assessment Instrument (RAI) MDS 2.0 user's manual, Canadian version. Retrieved from https://www.cihi.ca/en/types-of-care/hospital-care/continuing-care/ccrs-resources-for-rai-assessors
  • Canadian Institute for Health Information. (2013). When a nursing home is home: How do Canadian nursing homes measure up on quality? Retrieved from https://secure.cihi.ca/free_products/CCRS_QualityinLongTermCare_EN.pdf
  • Chaudhury, H. & Cooke, H. (2014). Design matters in dementia care: The role of the physical environment in dementia care settings. In Downs, M. & Bowers, B (Eds.), Excellence in dementia care (2nd ed.) (pp. 144–158). London, UK: Open University Press.
  • Chaudhury, H., Hung, L. & Badger, M.L. (2013). The role of physical environment in supporting person-centered dining in long-term care: A review of the literature. American Journal of Alzheimer's Disease and Other Dementias, 28, 491–500. doi:10.1177/1533317513488923 [CrossRef]
  • Cioffi, J.M., Fleming, A., Wilkes, L., Sinfield, M. & Le Miere, J. (2007). The effect of environmental change on residents with dementia: The perceptions of relatives and staff. Dementia, 6, 215–231. doi:10.1177/1471301207080364 [CrossRef]
  • Cohen-Mansfield, J., Thein, K., Dakheel-Ali, M. & Marx, M.S. (2010). Engaging nursing home residents with dementia in activities: The effects of modeling, presentation order, time of day, and setting characteristics. Aging & Mental Health, 14, 471–480. doi:10.1080/13607860903586102 [CrossRef]
  • Cutler, L.J. (2007). Physical environments of assisted living: Research needs and challenges. The Gerontologist, 47, 68–82. doi:10.1093/geront/47.Supplement_1.68 [CrossRef]
  • Day, K., Carreon, D. & Stump, C. (2000). The therapeutic design of environments for people with dementia: A review of the empirical research. The Gerontologist, 40, 397–416. doi:10.1093/geront/40.4.397 [CrossRef]
  • de Rooij, A.H., Luijkx, K.G., Schaafsma, J., Declercq, A.G., Emmerink, P.M. & Schols, J.M. (2012). Quality of life of residents with dementia in traditional versus small-scale long-term care settings: A quasi-experimental study. International Journal of Nursing Studies, 49, 931–940. doi:10.1016/j.ijnurstu.2012.02.007 [CrossRef]
  • Helmes, E., Csapo, K.G. & Short, J.A. (1987). Standardization and validation of the Multidimensional Observation Scale for Elderly Subjects (MOSES). Journal of Gerontology, 42, 395–405. doi:10.1093/geronj/42.4.395 [CrossRef]
  • Hung, L. & Chaudhury, H. (2011). Exploring personhood in dining experiences of residents with dementia in long-term care facilities. Journal of Aging Studies, 25, 1–12. doi:10.1016/j.jaging.2010.08.007 [CrossRef]
  • Kane, R.A., Lum, T.Y., Cutler, L.J., Degenholtz, H.B. & Yu, T.-C. (2007). Resident outcomes in small-house nursing homes: A longitudinal evaluation of the initial green house program. Journal of the American Geriatrics Society, 55, 832–839. doi:10.1111/j.1532-5415.2007.01169.x [CrossRef]
  • Lawton, M.P. & Nahemow, L. (1973). Ecology and the aging process. In The psychology of adult development and aging (pp. 619–674). Washington, DC: American Psychological Association. doi:10.1037/10044-020 [CrossRef]
  • Lee, S.Y., Chaudhury, H. & Hung, L. (2014). Exploring staff perceptions on the role of physical environment in dementia care setting. Dementia. Advance online publication. doi:10.1177/1471301214536910 [CrossRef]
  • Lee, S.Y., Chaudhury, H. & Lee, S.J. (2014). Effect of physical environment on the behaviors of residents with dementia: A comparison between a small-group unit and a traditional care unit. Journal of Civil Engineering and Architecture, 8, 1353–1363.
  • Lee, S.Y. & Dilani, A. (2011). Environmental design features and behavioural health for the elderly with dementia. Journal of Asia Interior Design Institute Association, 8, 68–80.
  • Leibrock, C. (2000). Design details for health: Making the most of interior design's healing potential. New York, NY: John Wiley & Sons.
  • Marsden, J.P. (2005). Humanistic design of assisted living. Baltimore, MD: JHU Press.
  • Regnier, V. (2002). Design for assisted living: Guidelines for housing the physically and mentally frail. New York, NY: John Wiley & Sons.
  • Reimer, M.A., Slaughter, S., Donaldson, C., Currie, G. & Eliasziw, M. (2004). Special care facility compared with traditional environments for dementia care: A longitudinal study of quality of life. Journal of the American Geriatrics Society, 52, 1085–1092. doi:10.1111/j.1532-5415.2004.52304.x [CrossRef]
  • Saxton, J., Silverman, M., Ricci, E., Keane, C. & Deeley, B. (1998). Maintenance of mobility in residents of an Alzheimer special care facility. International Psychogeriatrics, 10, 213–224. doi:10.1017/S1041610298005328 [CrossRef]
  • Schwarz, B., Chaudhury, H. & Tofle, R.B. (2004). Effect of design interventions on a dementia care setting. American Journal of Alzheimer's Disease and Other Dementias, 19, 172–176. doi:10.1177/153331750401900304 [CrossRef]
  • Sloane, P.D., Mitchell, C.M., Weisman, G., Zimmerman, S., Foley, K.M., Lynn, M. & Montgomery, R. (2002). The Therapeutic Environment Screening Survey for Nursing Homes (TESS-NH): An observational instrument for assessing the physical environment of institutional settings for persons with dementia. Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 57, S69–S78. doi:10.1093/geronb/57.2.S69 [CrossRef]
  • Smit, D., de Lange, J., Willemse, B. & Pot, A.M. (2012). The relationship between small-scale care and activity involvement of residents with dementia. International Psychogeriatrics, 24, 722–732. doi:10.1017/S1041610211002377 [CrossRef]
  • Van Amerongen-Heijer, Y.E. ( 2013, June. ). De Hogeweyk: Normal living for people with severe dementia. Paper presented at the 20th IAGG World Congress. , Seoul, Korea. .
  • Verbeek, H., Zwakhalen, S.M., van Rossum, E., Ambergen, T., Kempen, G.I. & Hamers, J.P. (2010). Dementia care redesigned: Effects of small-scale living facilities on residents, their family caregivers, and staff. Journal of the American Medical Directors Association, 11, 662–670. doi:10.1016/j.jamda.2010.08.001 [CrossRef]
  • Verbeek, H., Zwakhalen, S.M., van Rossum, E., Ambergen, T., Kempen, G.I. & Hamers, J.P. (2014). Effects of small-scale, home-like facilities in dementia care on residents' behavior, and use of physical restraints and psychotropic drugs: A quasi-experimental study. International Psychogeriatrics, 26, 657–668. doi:10.1017/S1041610213002512 [CrossRef]
  • Zeisel, J., Silverstein, N.M., Hyde, J., Levkoff, S., Lawton, M.P. & Holmes, W. (2003). Environmental correlates to behavioral health outcomes in Alzheimer's special care units. The Gerontologist, 43, 697–711. doi:10.1093/geront/43.5.697 [CrossRef]

Means and Results of t Test Analysis on the Therapeutic Environment Screening Survey for Nursing Homes

VariableRichmond Manor, Mean (SD)Maple Manor, Mean (SD)t value (two-tailed) (df)
Privacy/control/autonomy2.15 (2.34)1.31 (1.11)NS
Safety/security/health1.67 (0.82)1.50 (1.14)NS
Stimulation1.50 (0.69)0.95 (0.89)2.19* (38)
Socialization0.90 (0.88)0.50 (0.97)NS
Personalization/familiarity/home-likeness1.40 (1.14)0.20 (0.45)4.43** (8)
Orientation0.54 (0.52)0.38 (0.51)NS
Global rating7.03.5

Distribution of Dementia Care Mapping Data

Richmond ManorMaple Manor
VariableT1, f (%)T2, f (%)T3, f (%)T1, f (%)T2, f (%)T3, f (%)
Behavior Category Code
A. Articulation (interacting with others)102 (18.7)75 (23.4)55 (14.9)43 (7.8)42 (9.5)48 (8.5)
B. Borderline (passively watching)65 (11.9)46 (14.3)75 (19.9)87 (15.9)147 (33.2)171 (30.3)
C. Cool (withdrawn)46 (8.4)26 (8.1)11 (2.9)231 (42.1)38 (8.6)34 (6)
D. Doing for self (self care)5 (0.9)6 (1.9)7 (1.9)0 (0)2 (0.5)8 (1.4)
E. Expressive (creative activities)17 (3.1)19 (5.9)3 (0.8)0 (0)9 (2)9 (1.6)
F. Food (eating/drinking)95 (17.4)50 (15.6)67 (17.8)69 (12.6)72 (16.3)82 (14.5)
I. Intellectual (using cognitive abilities)0 (0)4 (1.2)6 (1.6)0 (0)0 (0)0 (0)
J. Joints (engaging in exercise)0 (0)0 (0)0 (0)0 (0)15 (3.4)3 (0.5)
K. Come and go (walking, standing)41 (7.5)13 (4)14 (3.7)23 (4.2)34 (7.7)51 (9)
L. Leisure (recreational activities)38 (7)16 (5)37 (9.8)16 (2.9)3 (0.7)11 (1.9)
N. Nod (sleeping, dozing)89 (16.3)36 (11.2)59 (15.7)76 (13.8)71 (16)112 (19.9)
O. Objects (displaying attachment)8 (1.5)1 (0.3)14 (3.7)0 (0)2 (0.5)16 (2.8)
P. Physical (receiving personal care)18 (3.3)12 (3.7)12 (3.2)4 (0.7)5 (1.1)12 (2.1)
R. Religion (religious activity)0 (0)2 (0.6)0 (0)0 (0)0 (0)0 (0)
T. Timalation (engagement of senses)3 (0.6)0 (0)0 (0)0 (0)0 (0)0 (0)
U. Unresponded (without receiving a response)1 (0.2)2 (0.6)5 (1.3)0 (0)0 (0)6 (1.1)
V. Vocational (work-like activity)9 (1.7)2 (0.6)3 (0.8)0 (0)2 (0.5)1 (0.2)
W. Withstanding (repetitive self-stimulation)6 (1.1)7 (2.2)5 (1.3)0 (0)0 (0)0 (0)
X. Excretion (excretion)0 (0)4 (1.2)3 (0.8)0 (0)1 (0.2)0 (0)
Y. Yourself (interaction in absence of other)2 (0.4)0 (0)0 (0)0 (0)0 (0)0 (0)
Total545 (100)321 (100)376 (100)549 (100)443 (100)564 (100)
Mood and Engagement Value
+5 (very happy/very absorbed)7 (1.5)1 (0.4)0 (0)0 (0)0 (0)0 (0)
+3 (content/concentrating but distractible)108 (23.7)40 (14)18 (5.7)17 (4.2)8 (2.2)19 (4.2)
+1 (neutral/intermittent engagement)288 (63.2)212 (74.4)287 (90.5)215 (53.1)317 (85.2)376 (83.2)
−1 (small signs of negative mood/withdrawal)53 (11.6)32 (11.2)12 (3.8)172 (42.5)42 (11.3)41 (9.1)
−3 (considerable signs of negative mood)0 (0)0 (0)0 (0)1 (0.3)5 (1.3)8 (1.8)
−5 (very distressed)0 (0)0 (0)0 (0)0 (0)0 (0)8 (1.8)
Mean score1.31.110.20.80.7
Authors

Dr. Lee is Senior Researcher, Research Center Design & Health, Stockholm, Sweden; Dr. Chaudhury is Professor, Department of Gerontology, Simon Fraser University, Vancouver; and Ms. Hung is PhD Student, University of British Columbia, and Clinical Nurse Specialist, Vancouver General Hospital, Vancouver, British Columbia, Canada.

The authors have disclosed no potential conflicts of interest, financial or otherwise. This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MEST) (KRF-2008-220-G00002).

Address correspondence to Sook Young Lee, PhD, Senior Researcher, Research Center Design & Health, Hagalundsgatan 35 11tr Solna, 169 66, Stockholm, Sweden; e-mail: sookyoung23@gmail.com.

Received: February 24, 2015
Accepted: June 30, 2015
Posted Online: July 14, 2015

10.3928/19404921-20150709-01

Sign up to receive

Journal E-contents