jnhr journal

AND option

OR option



L. Yong1, P. Upendranatha Reddy2, J. Tan3, D. Avers4, P.-H. Ong5, W.-P. Wong5


1. MOH Holdings, Singapore; 2. Woodlands Health Campus, Singapore; 3. NTUC Health, Singapore; 4. College of Health Professions, Upstate Medical University, New York; 5. Singapore Institute of Technology, Singapore. Corresponding author: Limin Yong, MOH Holdings, Singapore, limin.yong@mohh.com.sg, +65 91004283

Jour Nursing Home Res 2020;6:6-13
Published online April 10, 2020, http://dx.doi.org/10.14283/jnhrs.2020.2



Background: Decline in physical function is associated with increased frailty and hospitalization, resulting in decline of mental status and loss of functional independence. Nursing homes engage residents with variable intensity and frequency of physical activities to reduce functional decline. However the extent to which the physical function of nursing home residents change (that is, improve, maintain or decline) remains unknown. Objectives: The aims of this study were to document the change in physical function and functional category among 222 nursing home residents over a period of 9 months; and to identify the factors associated with changes in their physical function. Design: This is a longitudinal and multi-site study, with outcomes measured at point of recruitment and measurements repeated every three months for nine months. Setting: Three nursing homes in Singapore. Participants: A total of 222 residents (59% male and 41% female) with a mean age of 77.4 years, and an average length of stay of 64 days, participated in the study. Participants did not differ significantly from the 698 total residents in terms of age, number of medical comorbidities and physical function. The key exclusion criteria preventing participation in the study were cognitive status (MMSE < 11) and being placed on cardiac and pulmonary precautions. Measurements: Physical function was measured using the following outcome measures: sit and reach for flexibility, single leg stance for balance, 5 times chair-stand for overall lower extremity muscle strength, 10-meter walk tests for gait speed, and Modified Barthel Index (MBI) for activities of daily living (ADL). Participants were categorised into functional categories (“Fun”, “Functional”, “Frail”, “Failure”) for each test. Factors that are potentially associated with changes in physical function such as number and type of physiotherapy sessions received, number of hospitalisations, age, and length of stay, Charlson Comorbidity Index, and baseline Geriatric Depression Scale were also recorded. Results: There were significant improvements between baseline to 3rd month in flexibility (sit and reach distance from toe in centimeters) [2.69; 95% confidence interval [CI] = 1.05, 4.34; p = .001], balance (single leg stance in seconds) [8.46; 95% CI = 6.81, 10.12; p < .001), MBI [5.06; 95% CI = 2.61, 7.51; p < .001] but not strength and gait speed. Improvements were maintained until 9th month except on flexibility test. Gait speed significantly improved between baseline and 9th month [0.13, 95% CI = .009, .242; p = .035]. The group that improved the most were those who were in the “Failure” group at baseline. Number and type of physiotherapy sessions as well as number of hospitalisations was found to be associated with the extent of improvement in physical function. Conclusions: The study suggests that current rehabilitation framework at these nursing homes has the potential to enhance the physical function of residents who are in the earlier stage of their nursing home stay. Modifiable variables such as number and types of physiotherapy sessions appear to be significant factors associated with these outcomes. Hence, further studies examine the optimum frequency and type of physiotherapy sessions for this population.

Key words: Physical function, nursing home, longitudinal.



Age-related decline in physical function among older people is well documented (1). Although such functional decline has not been well defined, Schwartz (2) has described the slippery slope of functional decline in older people as consisting of “Fun”, “Function”, “Frailty” and “Failure” categories. “Fun” category refers to older people who are physically able to perform any activity at will; “Function” means those who are physically able to perform only those activities within their limited physical capacity; “Frailty” refers to those who require assistance with basic and instrumental activities of daily living (ADL); and “Failure” refers to those are bedbound and completely care-dependent.
Functional decline in older people can occur rapidly, resulting in functional losses in mental status and ability to accomplish ADL (3). Functional decline is also frequently associated with institutionalization in the older persons (4). Residents in nursing homes (NH), for instance, have been shown to experience functional decline (5).
Muscle strength loss is a critical factor contributing to functional decline. Older people experience an average decrease in muscle strength by approximately 1-2% every year (1, 6). In NH residents, reduced hand grip strength (16.5 ± 7.7 kg) and low physical activity (79.4% classified as having low physical activity <600 MET minutes per week) have been observed (7). Low mean walking speeds (0.37± 0.26 m/s; 75% of residents had speeds <0.5m/s) are also prevalent among NH residents (8). This is in fact linked to reduced muscle strength and balance in NH residents (9-11), with evidence that such reduction could be either reversed or slowed down with supervised and individually tailored physical exercises (12, 13).
The baseline physical and cognitive impairments among NH residents are key predictors of functional decline (5). Other contributing factors include poor postural balance, low body mass index, impaired vision, impaired hearing, social isolation and depression (14). Functional decline is associated with low health-related quality of life in older people (15-18) and NH residents (19). Given the evidence of functional decline in NH residents, reversing and delaying this decline should be an important NH care goal (7, 8, 12, 13). Increasingly, NH is incorporating physical exercise programs in the care of their residents. However, whether decline in physical function change as a result of including physical exercises in NH care remains to be investigated. Therefore, the aims of this study were to firstly determine the change among NH residents over nine months in their physical function and functional category according to Schwartz (2); and secondly to identify the factors associated with changes in their physical function.



The study protocol was approved by the University Institutional Review Board (Singapore Institute of Technology, Project No. 20170027).


This study adopted a longitudinal design, involving three NHs with a total capacity of 800 beds. The NHs are located in the east, west and central part of Singapore.


Residents who scored at least 11 points on the Mini-Mental State Examination (MMSE) (20) and were able to follow simple verbal instructions (in order to participate in the assessments and physiotherapy sessions) were enrolled into the study. Residents requiring cardiac precautions, with untreated deep vein thrombosis or pulmonary edema, Parkinson’s Disease Hoehn and Yahr Stage 4 and above, with existing contractures, or were on palliative care or non-weight bearing status were excluded. Altogether 278 residents were included in the final analysis. Informed consent were sought and obtained.


Participants underwent a battery of tests measuring their physical function. These tests included sit and reach test for flexibility, single leg stance for balance, 5-times chair stand for lower extremity muscle strength, 10-meter walk test for gait speed and Modified Barthel Index (MBI) for ADL. Trained research assistants who had completed standardized training and were assessed to be competent performed these tests.
All participants performed active range of movement exercises to upper and lower limbs for 5 minutes as warm-up prior to the physical tests. The participants were allowed to rest for 3 minutes between tests.


Sit and reach test

This is a test of back and hamstring muscle flexibility (21, 22). Residents sat on their bed, with legs extended and back supported at 90 degrees. The soles of the feet were placed flat against a box. Both knees were maintained in extension. They were then instructed to reach forward with their hands towards their toes as far as possible. The best distance of the tips of the fingers to the toes in three attempts was then measured in centimetres and recorded.


Single leg stance

This test measures the ability to maintain postural control while standing on one leg (23). With eyes open and hands on the hips, residents were instructed to stand on a single leg while maintaining the other leg off the floor for as long as possible. The test was stopped if the patient was not able to maintain their hands on the hips or standing on one leg. Both legs were tested. The best performance out of three attempts was recorded in seconds for each leg.

Gait Speed

10-meter walk test

This is a test of mobility, power, and quality of aging (24). Residents were instructed to walk 10 meters at their preferred walking speed. They were given two meters of space for acceleration and deceleration. The duration to complete the walk was measured in seconds. The best time out of three trials were recorded.

Lower Extremity Strength

5-times chair stand

This test required residents to be seated in the middle of a chair (17 inches height, placed against a wall), without an arm rest, with their back straight and feet approximately shoulder-width apart. The feet were on the floor at an angle slightly posterior to the knees. Arms were crossed and held against the chest. The residents were then instructed to rise to a full stand and return back to their initial seated position five times (25). The time taken to complete this was measured in seconds.

Functional markers

Participants’ performance in the physical tests was categorized as fun, functional, frail or fail (2, 26).

Activities of daily living

The MBI measures ADL and includes ratings of bowel and bladder control, personal hygiene, toilet use, feeding, transfer, mobility, dressing, stairs climbing and bathing. The total score on the MBI ranged from 0-100; the higher the score, the better the physical function for ADL (27).

Other variables

Variables that could influence changes in the participants’ physical function were collected: age, length of stay in the NH, number of medications prescribed and comorbidities were collected at the beginning of the study. Comorbidities were categorized using the Charlson Comorbidity Index (CCI): the higher the score, the more likely the predicted outcome would result in mortality or higher resource use (28). Additionally, the resident’s psycho-emotional status was measured using the Geriatric Depression Scale (GDS) (29). The GDS is a 30-item questionnaire in which the participants were asked to rate their responses to statements such as “Do you feel pretty worthless the way you are now?” The higher the score on the GDS, the poorer the psycho emotional status of the resident. Scores above 10 indicates likely depression.
All measurements were made at the start of study, 3rd month, 6th month and 9th month. Throughout the duration of the study, participants continued with the existing programs in the NH. They attended a combination of individual physiotherapy sessions, group sessions and exercise circuit training conducted by the physiotherapy assistants under the supervision of physiotherapists in the respective NH. The number of sessions attended by each participant was according to his or her needs and abilities. The number of sessions attended by each participant was recorded.

Data analyses

There were less than 5% missing data from the physical function outcome measures collected throughout the duration of the study. In this study, we chose the method of retaining all data by conducting a “mean imputation” to fill missing data (30). Each missing value was replaced with the mean of the observed values for that variable. Comparisons of the standard deviation and distribution pre- and post- missing value imputation showed that both were not distorted with “mean imputation”.
IBM SPSS® Statistics version 25 (IBM, New York) was used for data analyses, using two-tailed tests with significance value set at p < .05. Paired sample t-tests were conducted to determine whether the absolute differences in physical function over the four time points were statistically significant. As multiple pairwise comparisons were conducted, the criteria for statistical significance were adjusted using the Bonferonni correction method. Improvements in absolute score (in seconds or centimeters) might not necessarily make any functional changes to a person’s physical ability. To test whether the functional categories of residents improve, a Wilcoxon signed-rank test was conducted with resident’s functional category status at baseline compared to that at 9th month.
Bivariate correlations and multiple regression (using stepwise approach) were conducted to examine the relationship between key factors (age, length of stay, number of physiotherapy sessions, CCI, number of hospitalisations, GDS and falls) and the dependent variable, which is the changes in performance on the physical functional tests. Variables with a p<0.05 were considered for inclusion in subsequent multivariable model.
A one-way ANOVA followed by post-hoc pairwise comparison using the Tukey test were conducted to determine which of the above factors were significantly different between residents who improved versus those who maintained or declined in their physical function.



Participant profile

Out of the 278 residents enrolled, six were transferred to other care facilities, 24 died, 15 were discharged, and 233 completed the study. From these, 11 residents were unable to complete more than three of the five physical outcome measures at all-time points. Therefore, data from 222 residents (91 females and 131 males) were analysed in this study.
Participants ranged from 52 to 104 years of age (mean M = 77.4, standard deviation SD = 9.4). Their CCI scores were 0-13 (M = 6.04, SD = 1.85). At the start of the data collection, their MBI scores ranged from 2 to 94 (M = 47.9, SD = 22.0) and their GDS scores from 0-14 (M = 5.5, SD = 3.5). Participants had between 1-27 different types of medications (M = 9.94, SD = 4.03). Participants’ length of stay ranged from 8 to 248 days (M = 64, SD = 48) at the point of recruitment. During the study period, participants were hospitalised 0-7 times (M = 0.84, SD = 1.37) and had 0-4 upper respiratory tract infection episodes (M = 0.29, SD = 0.62).
The top five reasons for admission into the NH were lack of caregiver (27.5%), family unable to support (27%), family unable to cope (15.8%), strained relationship with family (6.8%) and person unsafe to be home alone (2.3%).
These 222 participants were fairly representative of the overall resident population (n = 698) in the three NHs with the exception of average length of stay. The overall NH resident population demographics are as follows – age: ranged from 49 to 104 years of age (mean M = 79.2, SD = 9.1); CCI scores: ranged from 0 to 17 (M = 6.72, SD = 1.89); MBI scores: ranged from 0 to 94 (M = 46.8, SD = 22.7); number of hospitalisations: ranged from 0 to 11 (M = 0.91, SD = 1.41). The overall NH resident population had a longer average length of stay: ranged from 6 to 304 (M = 77, SD = 61).

Physiotherapy sessions (with therapy assistants) received


93% of the participants received these sessions, ranging from 1 to 151 sessions (Median = 20) per person throughout the 9-month period.

Group therapy

96% of the participants received these sessions, ranging from 1 to 264 sessions (Median = 137.5) per person throughout the 9-month period.

Exercise Circuit

81.5% of the participants received these sessions, ranging from 1 to 123 sessions (Median = 6) per person throughout the 9-month period.

Absolute changes in physical function status

Table 1 below summarises the changes in physical function of residents throughout the study.
Over the first 3-month period, residents significantly improved in flexibility, balance, and overall physical function but not lower extremity muscle strength (time taken to complete 5 chair stands) and gait speed. The largest gain in function appears to occur within this first 3-month period (between baseline and 3rd month), with the exception of gait speed, which residents showed a steady increase in performance over time, reaching significance at 9th month. The significant gains made between baseline and 3rd month on the sit and reach test was not maintained at 9th month. Below are further details of the analyses between time points for the various physical functional outcome measures.

Table 1
Summary of physical functional and psycho-emotional measures over the 9-month period



Sit and Reach

There was a significant increase in sit and reach test scores between baseline and 3rd month [2.69; 95% confidence interval [CI] = 1.05, 4.34, p = .001]. Participants were able to reach forward by 2.7 cm closer to target between baseline and 3rd month. This gain was however not maintained at the end of the 9-month period.


Single Leg Stance

There was a significant increase in single leg stance on the right leg between baseline and 3rd month [8.46; 95% CI = 6.81, 10.12, p < .001]. Participants continued to improve between 3rd and 6th month [3.28; 95% CI = 1.65, 4.91, p < .001]. There were no significant difference between 6th and 9th month, but performance at 9th month was still significantly better than that at baseline [11.16; 95% CI = 9.35, 12.96, p < .001].
There was a significant increase in single leg stance on the left leg between baseline and 3rd month [7.75; 95% CI = 5.71, 9.8, p < .001]. Participants continued to improve between 3rd and 6th month [3.47; 95% CI = 1.88, 5.06, p < .001]. There were no significant difference between 6th and 9th month, but performance at 9th month was still significantly better than that at baseline [10.24; 95% CI = 8.22, 12.27, p < .001].

Gait Speed

10-meter walk test

There was a significant increase in gait speed only between baseline and 9th month [.13; 95% CI = .01, .24, p = .035].

Lower extremity Strength

Chair stand test

There was a significant improvement in time taken (in seconds) to perform five chair stands only between 3rd and 6th month [-1.66; 95% CI = -.22, -3.09, p = .024].

Overall physical function

Modified Barthel Index

There was a significant difference in MBI score between baseline and each of the 3 subsequent time points. Baseline vs 3rd month [5.07; 95% CI = 2.61, 7.52, p < .001]; baseline vs 6th month [5.38; 95% CI = 2.9, 7.87, p < .001]; baseline vs 9th month [5.51; 95% CI = 3.02, 8.0, p < .001].

Functional categorical changes in physical abilities

Changes in rank

For most of the outcome measures, most of the residents maintained their functional category, with the exception of single leg stance where most residents (58%) had improved. For the rest of the outcome measures, more residents had shown improvements than a decline in function. A summary of the results can be found in Table 2 below.

Table 2
Wilcoxon signed rank test indicating changes in functional status from baseline to 9th month


Upon examination of the data, it was found that the group that showed greatest improvements were those in the “Failure” category. Table 3 summarises the shift in the functional categories for each outcome measure in this group of residents.

Table 3
Changes in functional status over a 9-month period for residents who were “Failures” at baseline


Difference between participants who improved, declined and maintained function

The key factors that differentiated those who improved and those who declined in function were number of group therapy sessions and hospitalisation throughout the study period. This indicates that physiotherapy interventions and health status were two important factors. Table 4 summarises the differences that were significant.

Table 4
Significant differences between the improved, declined and maintained group for the different physical outcome measures


Factors associated with improvements in physical function of NH residents

How much the above factors contributed to the improvements in physical function was assessed with multiple regression. Table 5 summarises the relationships that were significant.
Improvements in physical function were higher with more physiotherapy interventions provided. Conversely, the worse their health i.e. greater number of hospitalisations, the lower the improvement in physical function. These factors accounted for between 1.9 to 3.7% of variance in improvements in various physical functions.
Interestingly, the CCI scores of nursing home residents were neither significantly correlated nor contributory to a significant amount of variance in the various physical functions measured in this study. CCI was however positively correlated with age. The older the residents were, the higher their CCI scores, r = .271, p < .001.

Table 5
Factors that significantly associated with improvements in physical function between baseline and 9th month



In this study, participants were fairly representative of the overall NH resident population with the exception of their average length of stay at point of recruitment. The difference in average length of stay suggests that our findings may be more applicable to residents who are in the earlier stage of their nursing home stay.
Our study shows that with physiotherapy and physical exercises incorporated in NH care, 29% of the residents have demonstrated improvements, 57% maintained their functional abilities without functional decline, whereas only 14% have shown decline in their functional abilities across the outcome measures over a 9-month period. These findings are similar to that of a recent systematic review of literature from 1983 till 2011, showing that most residents either maintain or improve their functional status (10).
Improvements were recorded in balance, gait speed and overall physical function, but not flexibility and strength. Analyses of the results showed that the group with the greatest improvements were those in the “Failure” category. Improvements from “Failure” category to the “Fun”, “Functional” and “Frail” categories were made by 13.6% of residents on the gait speed test, 15.7% on the flexibility test, 62.7% on the balance test, and 70.9% on the chair stand test. This implies that quantitative improvements in physical test scores could be translated to physical functional gains. Findings from this study agree with those of clinical trials examining the role of intensive physical exercise programs in NH residents (12, 13).
In this study, the total number of medications, the number of medical comorbidities (CCI), upper respiratory tract infections, and bowel and bladder continence did not have any significant relationship to the physical outcome measures of the NH residents. This finding contradicts some authors’ speculation that comorbidities are predictive of functional decline in NH residents (9). One study shows that medications and comorbidities could predict functional decline (11), however, it has not included any physical exercise program as in the present study.
The number of group therapy sessions and hospitalisation throughout the study period were also found to differentiate between those who had improved and those who had declined in physical function. This suggests that therapy and health status are two important factors contributing to improvements in physical function amongst the NH residents. Previous studies supported the association between number of physiotherapy sessions and extent of improvement in physical function (31-34). Conversely, the worse their health, as indicated by hospitalisation episodes in this study, the lower the improvement in physical function. Whilst the health status of the NH residents are dependent on many factors, some of which might be non-modifiable, the number of physiotherapy sessions can be increased in the NH setting. In Singapore, currently there are 72 NHs with 12,000 beds; the Ministry of Health targets to increase the number of NH beds to 17,000 by the year 2020 (35). The scaling-up reflects the increasing demand for NH beds. However, unless the NH residents’ physical function improves to the point of discharge, the number of beds will not necessarily meet the demand from an ageing population. Incorporating and optimizing the number of physiotherapy sessions could potentially be one strategy to contribute to improvement of NH residents’ physical function.
There are policy implications. The finding that the greatest improvement is shown among those whose baseline functional category is labelled “Failure” challenges the traditional view that those who are in failure category of functional mobility, are often considered low on the potential for rehabilitation. This finding offers implications for NH administrators and policy-makers where funding rehabilitative services in NHs could be reviewed.

Study limitations

Although data were collected on the amount of individual, group and exercise circuit therapy sessions with therapy assistants, other physical activities carried out at the NH accommodation level, including occupational therapy, nursing and volunteer-led activities, were not accounted for. Hence, the extent to which these other activities contributed to the physical outcomes could not be ascertained. In addition, this study excluded residents with moderate to severe cognitive impairments. Hence, results cannot be generalised to all NH residents, given that many of them have cognitive impairment. There was a 16% attrition rate of participants secondary to discharge from NH (back to their own home) and deaths. The performance of this pool of residents was not represented in the study analyses.

Future directions

Future studies could investigate the optimal amount and frequency of physical exercises for NH residents. This knowledge could help NH organizations allocate health care resources more effectively. Other studies could examine the reasons for hospitalization and how the health status of NH residents could be improved. Further studies could also incorporate economic evaluation such as cost-effectiveness analysis.



Our study suggests that incorporating physical exercises into the daily activities of NH residents in the earlier stage of their stay has the potential to enhance the physical function of these residents. Given that physical exercises is a modifiable factor and can be easily incorporated into the daily care plans of residents in the NH, further studies could examine the optimum frequency and type of physiotherapy sessions for residents who are in the earlier stage of their nursing home stay to maximise this potential.


Funding: Support for this study was provided under the Agency for Integrated Care grant number ILTCRG2016/4. Funding was provided between August 2017 and March 2019. The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript.

Acknowledgements: The authors would like to acknowledge the following individuals for their support in this research: Alyah Dinah Binte Zalzuli, Jenalynne Tomas Montemayor, Lee Pei Pei, Stacie Teow Pei Pei, Aye Aye Maw, Kyi Kyi Khaing, Angel Lim, Shwe Sin Win Khin, Chiew Yian Nee and Lim Jia Min.

Ethical standards: The study protocol was reviewed and approved by the Institutional Review Board of Singapore Institute of Technology.



1. Milanović Z, Pantelić S, Trajković N, Sporiš G, Kostić R, James N. Age-related decrease in physical activity and functional fitness among elderly men and women. Clinical Interventions in Aging. 2013;8:549.
2. Schwartz RS. Sarcopenia and physical performance in old age: introduction. Muscle & Nerve: Official Journal of the American Association of Electrodiagnostic Medicine. 1997;20:10-12.
3. Raîche M, Hébert R, Dubois M-F, Dubuc N. Yearly transitions of disability profiles in older people living at home. Archives of Gerontology and Geriatrics. 2012;55:399-405.
4. Gaugler JE, Duval S, Anderson KA, Kane RL. Predicting nursing home admission in the US: a meta-analysis. BMC Geriatrics. 2007;7:13.
5. Banaszak-Holl J, Liang J, Quinones A, Cigolle C, Lee I-C, Verbrugge LM. Trajectories of functional change among long stayers in nursing homes: does baseline impairment matter? Journal of Aging and Health. 2011;23:862-882.
6. Rantanen T, Masaki K, Foley D, Izmirlian G, White L, Guralnik J. Grip strength changes over 27 yr in Japanese-American men. Journal of Applied Physiology. 1998;85:2047-2053.
7. Senior HE, Henwood TR, Beller EM, Mitchell GK, Keogh JW. Prevalence and risk factors of sarcopenia among adults living in nursing homes. Maturitas. 2015;82:418-423.
8. Keogh JW, Senior H, Beller EM, Henwood T. Prevalence and risk factors for low habitual walking speed in nursing home residents: an observational study. Archives of Physical Medicine and Rehabilitation. 2015;96:1993-1999.
9. Colón-Emeric CS, Whitson HE, Pavon J, Hoenig H. Functional decline in older adults. American Family Physician. 2013;88:388.
10. Palese A, Menegazzi G, Tullio A, Fuso MZ, Hayter M, Watson R. Functional decline in residents living in nursing homes: A systematic review of the literature. Journal of the American Medical Directors Association. 2016;17:694-705.
11. Sibbritt DW, Byles JE, Regan C. Factors associated with decline in physical functional health in a cohort of older women. Age and Ageing. 2007;36:382-388.
12. Frändin K, Grönstedt H, Helbostad JL, Bergland A, Andresen M, Puggaard L, et al. Long-term effects of individually tailored physical training and activity on physical function, well-being and cognition in Scandinavian nursing home residents: a randomized controlled trial. Gerontology. 2016;62:571-580.
13. Telenius EW, Engedal K, Bergland A. Effect of a high-intensity exercise program on physical function and mental health in nursing home residents with dementia: an assessor blinded randomized controlled trial. PloS One. 2015;10:e0126102.
14. Bürge E, von Gunten A, Berchtold A. Factors favoring a degradation or an improvement in activities of daily living (ADL) performance among nursing home (NH) residents: a survival analysis. Archives of Gerontology and Geriatrics. 2013;56:250-257.
15. Cress ME, Buchner DM, Questad KA, Esselman PC, DeLateur BJ, Schwartz RS. Continuous-scale physical functional performance in healthy older adults: a validation study. Archives of Physical Medicine and Rehabilitation. 1996;77:1243-1250.
16. Wood RH, Reyes-Alvarez R, Maraj B, Metoyer KL, Welsch MA. Physical fitness, cognitive function, and health-related quality of life in older adults. Journal of Aging and Physical Activity. 1999;7:217-230.
17. Sato S, Demura S, Kobayashi H, Nagasawa Y. The relationship and its change with aging between ADL and daily life satisfaction characteristics in independent Japanese elderly living at home. Journal of Physiological Anthropology and Applied Human Science. 2002;21:195-204.
18. Stewart KJ, Turner KL, Bacher AC, DeRegis JR, Sung J, Tayback M, et al. Are fitness, activity, and fatness associated with health-related quality of life and mood in older persons? Journal of Cardiopulmonary Rehabilitation and Prevention. 2003;23:115-121.
19. Olsen C, Pedersen I, Bergland A, Enders-Slegers M-J, Jøranson N, Calogiuri G, et al. Differences in quality of life in home-dwelling persons and nursing home residents with dementia–a cross-sectional study. BMC Geriatrics. 2016;16:137.
20. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12:189-198.
21. Baltaci G, Un N, Tunay V, Besler A, Gerçeker S. Comparison of three different sit and reach tests for measurement of hamstring flexibility in female university students. British Journal of Sports Medicine. 2003;37:59-61.
22. Australian College of Sport & Fitness. Flexibility test – sit and reach. http://acsf.edu.au/pdf/Flexibility_Test_Sit_and_Reach.pdf. Accessed 10 May 2017
23. Bohannon RW. Single limb stance times: a descriptive meta-analysis of data from individuals at least 60 years of age. Topics in Geriatric Rehabilitation. 2006;22:70-77.
24. Peters DM, Fritz SL, Krotish DE. Assessing the reliability and validity of a shorter walk test compared with the 10-Meter Walk Test for measurements of gait speed in healthy, older adults. Journal of Geriatric Physical Therapy. 2013;36:24-30.
25. Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. Journal of Gerontology. 1994;49:M85-M94.
26. VanBeveren PJ, Avers D. Geriatric Physical Therapy. In: Guccione AA, Wong R, Avers D (eds). Geriatric Physical Therapy. 3rd edn. 2012. Elsevier Mosby,St Louis, MO, pp. 64-86
27. Duffy L, Gajree S, Langhorne P, Stott DJ, Quinn TJ. Reliability (inter-rater agreement) of the Barthel Index for assessment of stroke survivors: systematic review and meta-analysis. Stroke. 2013;44:462-468.
28. Buntinx F, Niclaes L, Suetens C, Jans B, Mertens R, Van den Akker M. Evaluation of Charlson’s comorbidity index in elderly living in nursing homes. Journal of Clinical Epidemiology. 2002;55:1144-1147.
29. Sheikh JI, Yesavage JA. Geriatric Depression Scale (GDS): recent evidence and development of a shorter version. Clinical Gerontologist: The Journal of Aging and Mental Health. 1986.
30. Little RJA, Rubin DB. Statistical Analysis with Missing Data. 2nd ed. 2002. Wiley-Interscience,New Jersey
31. Peiris CL, Taylor NF, Shields N. Extra physical therapy reduces patient length of stay and improves functional outcomes and quality of life in people with acute or subacute conditions: a systematic review. Archives of Physical Medicine and Rehabilitation. 2011;92:1490-1500.
32. Webber SC, Porter MM, Menec VH. Mobility in older adults: a comprehensive framework. The Gerontologist. 2010;50:443-450.
33. Howe TE, Rochester L, Neil F, Skelton DA, Ballinger C. Exercise for improving balance in older people. Cochrane Database of Systematic Reviews. 2011.
34. Rogers ME, Rogers NL, Takeshima N, Islam MM. Methods to assess and improve the physical parameters associated with fall risk in older adults. Preventive Medicine. 2003;36:255-264.
35. Ministry of Health Singapore. Health Facilities, MOH 2016. https://www.moh.gov.sg/content/moh_web/home/statistics/Health_Facts_Singapore/Health_Facilities.html. Accessed 10 May 2017

What do you want to do ?

New mail

What do you want to do ?

New mail

What do you want to do ?

New mail

What do you want to do ?

New mail

What do you want to do ?

New mail

What do you want to do ?

New mail



I. Røen1, J. Šaltytė Benth1,2,3  , Ø. Kirkevold1,4,5, I. Testad6,7,8, G. Selbæk1,4,9, K. Engedal4, S. Bergh1,4


1. Centre for Old Age Psychiatric Research, Innlandet Hospital Trust, Ottestad, Norway; 2. Institute of Clinical Medicine, Campus Ahus, University of Oslo, Oslo, Norway; 3. Health Services Research Unit, Akershus University Hospital, Lørenskog, Norway; 4. Norwegian National Advisory Unit on Ageing and Health, Vestfold Hospital Trust, Tønsberg, Norway; 5. Norwegian University of Science and Technology (NTNU), Department of Health Science in Gjøvik, Gjøvik, Norway; 6. Centre for Age-Related Medicine – SESAM, Stavanger University Hospital, Stavanger, Norway; 7. Department of Old Age Psychiatry, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, UK; 8. University of Exeter Medical School, St Luke’s Campus, University of Exeter, UK. 9. Faculty of Medicine, University of Oslo, Oslo, Norway. Corresponding author: Irene Røen, Centre for Old Age Psychiatric Research, Innlandet Hospital Trust, p.b. 68, 2312 Ottestad, Norway.  E-mail address: irroee@sykehuset-innlandet.no, Phone: +47 90652165

Jour Nursing Home Res 2019;5:8-19
Published online Februay 11, 2019, http://dx.doi.org/10.14283/jnhrs.2019.2



Background: There is no cure for dementia and appropriate care should be offered to improve or maintain quality of life for those living with dementia. Objectives: To identify groups of residents following similar trajectories of quality of life after nursing home admission, to examine which resident, staff, and organizational characteristics at baseline differ between the identified groups, and to assess the associations between the trend in quality of life and the same characteristics measured at baseline and over the study period. Design: A prospective, observational, longitudinal cohort design over 30 months. Setting: Nursing homes in Norway. Participants: Residents admitted to nursing homes. Measurements: Resident data on quality of life, dementia, pain, activities of daily living, physical health, neuropsychiatric symptoms, medication, and demographic characteristics were obtained by interviews. Unit characteristics and the staff data on person-centered care, psychosocial factors, and job satisfaction were obtained by questionnaires and interviews. The physical environment of the unit was assessed by structured observation. Results: 694 residents admitted to a nursing home and 1161 staff from 175 nursing home units participated. Three resident groups following similar trajectories in quality of life were identified by growth mixture model; good quality of life (53.6%), moderate quality of life (32.9%), and poor quality of life (13.4%). All groups’ quality of life decreased over time. More pain, more severe dementia, and more affective symptoms at baseline were associated with belonging to the poor quality of life group. Overall decline in quality of life was associated with more severe dementia, more pain, poorer function in activities of daily living, more severe neuropsychiatric symptoms among residents, and poorer job satisfaction among staff. Conclusion: Reducing pain, reducing NPS, improving activities of daily living for the residents, and improving the staff’s job satisfaction may be factors of importance to improve the residents’ quality of life.

Key words: Quality of life trajectories, nursing home, dementia, job satisfaction, physical environment.




Dementia is a chronic syndrome that can be caused by a variety of brain disorders, most frequently Alzheimer’s disease. It is characterized by cognitive decline, impaired functioning in daily life activities, deterioration of emotional control, and change of social behavior or motivation, and is one of the factors most strongly associated with admission to a nursing home (NH) (1).
There is no cure for any of the brain disorders causing dementia, and appropriate care should be offered to improve or maintain the quality of life (QoL) of persons with dementia (2). QoL is a multidimensional concept encompassing the emotional, physical, social, and environmental domains of a person’s wellbeing (3). Several studies have investigated QoL in NH patients, both self reported and proxy reported (family and staff), where age, ADL, dementia severity, pain, psychiatric disorders, pulmonary diseases and neuropsychiatric symptoms (NPS), are found to be associated with reduced QoL in NH patients with dementia (4, 5). Previous studies, following QoL in persons with dementia over time in NH, found QoL to be rather stable (6-8). Attention towards improving quality of NH care and QoL for NH residents is needed, and staff knowledge and skills should be reinforced to maintain or improve the residents’ QoL. A review from 2016 found that when the staff were trained to interact empathetically and humanely with the patients in their care, the residents experienced fewer depressive symptoms, less functional dependence, better food intake, and less psychotropic medications were prescribed. The review concluded that associations exist between potentially adjustable staff variables and Quality of care onwards to QoL (9).
Person-centered care (PCC) is regarded as good quality of care and is a guiding principle in dementia care (10, 11). An increasing amount of literature has evaluated resident outcomes of PCC, showing significant benefits such as decrease in NPS, reduced prescription of psychotropic medication, improved mood, improved QoL, and cost-effectiveness in providing care to persons with dementia in long-term care (11, 12).
Additionally, a recent literature review concluded that the physical environment of care settings is important in improving the residents’ QoL and in improving quality of care practices (13).
The previous longitudinal studies on QoL in nursing homes (6-8) have small cohort size, follow their participants over a short period of time, and to a lesser extent investigate associations with important patient, staff- and NH-variables. Therefore, we designed a study to identify groups of residents following similar trajectories of QoL after admission to NH, over a period of 30 months; and, to examine how resident, NH staff, and unit characteristics measured at baseline were associated with the group-belonging. Additionally, we aimed at assessing the associations between the same characteristics and the overall trend in QoL, with the characteristics measured simultaneously as QoL whenever possible.




This is a longitudinal observational study of patients in 47 NHs in Norway, previously described in the “Resource Use and Disease Course in dementia – Nursing Home (REDIC-NH) study” (14). Resident baseline data were collected at admission to the NH (within one month of admission), and follow-up data were collected biannually for 30 months. The baseline data were collected between March 2012 and November 2014, and the last follow-up data were collected in May 2017.
NH demographics and staff characteristics were obtained through questionnaires distributed to the staff and the head nurse of the NH units; these included both standardized questionnaires and questions developed for this study by the reseach group. The members of the research group are highly experienced in both clinical practice and NH research, and the questions developed were based on previous experience, and literature regarding organizational and structural factors in NH (i.e. staff level and education, leadership, management, physical environment and culture). The physical environment of the units was assessed by structured observation between October 2013 and December 2014, presented in a previous study (15).

Ethics approval and consent to participate

The residents’ capacity to consent to participate in the study was evaluated by the NH staff and physician, in close collaboration with relatives. Written consent for participation was obtained from all participants with the capacity to consent, and for participants lacking the capacity to consent, the next of kin gave consent on behalf of the residents. Data from the staff and the head nurses were collected anonymously.The Regional Ethics Committee for Medical Research in South-Eastern Norway approved the study (2011/1738a).

Participant inclusion criteria

Nursing home resident

All residents 65 years or older, regardless of degree of cognitive function, were eligible for inclusion in the study. In addition, we included residents younger than 65 years with established dementia, as their symptoms and functional decline over time resembles elderly patients in NH. Residents with an expected stay in the NH under four weeks or an expected life expectancy of less than six weeks were excluded.

Nursing home staff

NH staff who knew the residents in the unit and the organizational structure of the unit well, were eligible for inclusion in the study.

Sample characteristics and measurements

Nursing home residents

Demographic characteristics of the residents included age, gender, marital status, and medication. The residents’ QoL was assessed with the Quality of life in Late-Stage Dementia scale (QUALID), a standardized and validated proxy-based questionnaire with a sum score ranging from 11 to 55, with lower scores indicating a better QoL (16). QUALID is not validated for persons without dementia, but as the aim of the study was to follow the participants over 30 months we expected that some of the participants without dementia at baseline, would develop dementia during the study period. In addition, some participants would have severe dementia and a proxy-based QoL assessment tool would be useful. Therefore, we judged QUALID a sensible assessment tool in our study. Dementia at baseline was diagnosed independently by GS, KE and SB using all available collected information. In case of disagreement the cases were discussed until consensus was reached. Dementia severity was assessed with the Clinical Dementia Rating scale (CDR), a global rating scale covering six domains of cognitive and functional performance (17). The CDR sum of boxes (CDR-SOB) was calculated by adding the domain scores, which range from 0 – 18, with a higher score indicating more severe dementia (18). Pain was assessed with the Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID-2), which consists of 10 items, each item score ranging from 0 to 10, with a higher score indicating more severe pain (19). ADL function was assessed with the Physical Self-Maintenance Scale (PSMS), a six-item scale ranging from 6 to 30, with a higher score indicating a lower level of functioning (20). General physical health was assessed using the General Medical Health Rating (GMHR) scale, a one-item global rating scale with four categories (excellent, good, fair, poor) (21). Neuropsychiatric symptoms (NPS) were assessed using the 12-item Neuropsychiatric Inventory nursing home version (NPI-NH) (22). An NPI item score was calculated by multiplying frequency (0-4) with severity (0-3), producing an item score (0-12), with a higher score indicating more severe NPS. NPI sub-syndrome scores were calculated based on a previous factor analysis: NPI agitation (agitation/aggression, disinhibition, and irritability, range 0-36), NPI psychosis (delusions and hallucinations, range 0-24), and NPI affective (depression and anxiety, range 0-24) (23).

Nursing home staff

NH staff characteristics were obtained through questionnaires, and included age, gender, Norwegian as first language, number of years of health-related education, experience in the current job, and percentage of full-time position.
Person-centered care was assessed with the Person-centered Care Assessment Tool (P-CAT), consisting of 13 items that are formulated to measure staff perceptions of the practice in the unit where they work. The total score ranges from 13 to 65, with higher scores indicating a higher level of PCC (24).
Work-related psychosocial factors were assessed with 32 of the 129 items in the General Nordic Questionnaire for Psychosocial and Social Factors at Work (QPS-Nordic), covering essential social and psychological factors at work (25). The 32 items are distributed across 10 scales; each scale consists of 3 or 4 items, giving a subscale score of 3-15 or 4-20, with higher scores indicating better work-related psychosocial factors.
NH staff’s job satisfaction was measured with a single question: “How would you describe your general experience of your job satisfaction?” with seven possible answers: “very bad – bad – unsure – quite good – good – excellent”.

Unit characteristics

A unit in a NH was defined as a group of residents who live together with a common living area and who have their own care staff during the daytime. Data were collected about the unit size (number of residents); the daytime staff/resident ratio (the number of NH staff working per resident during the daytime); the type of unit (special care unit [SCU] or regular unit [RU]); the number of hours the nursing home physician was working per resident per week; the number of units per head nurse; and whether the unit had a nursing professional development specialist.
The physical environment of the unit was assessed with the Special Care Unit Environmental Quality Scale (SCUEQS), a summary scale embedded in the Therapeutic Environment Screening Survey for Nursing Homes (TESS-NH) (26). Scores range from 0 – 41, with higher scores indicating a better physical environment.

Statistical methods

Resident, NH staff, and unit characteristics were described as means and standard deviations (SD) or as frequencies and percentages. Participants vs. non-participants were compared by Independent samples t-test or χ2-test. Staff characteristics used in regression models were aggregated to a mean score at each unit. Missing values on staff-rated person-centered care (P-CAT) items were imputed on cases with fewer than 50% missing on the P-CAT scale by generating an empirical distribution for each item and drawing a random number from it. As an exploratory approach, growth mixture model was estimated to identify potential groups of residents that were following similar trajectories in QoL score throughout the study period. According to this method, the groups are identified based on individual trajectories of residents. The number of groups was determined by applying Akaike’s Information Criterion (AIC), where a smaller value indicates a better model. It was also required, that 95% confidence intervals of trajectories were non-overlapping and within-group probabilities high, preferably 0.80 or higher. Nominal regression analysis was performed to assess if resident and NH staff characteristics and/or factors related to the physical environment (as measured at baseline) were associated with group-belonging. No clustering within the NH-unit was present, and hence no adjustment was implemented into the nominal regression model. The associations between the same characteristics as measured at baseline or all assessments (whenever available) and the overall trend in QoL were assessed by estimating a linear mixed model with random effects for residents, units, and the interaction between the two. The fixed effects for time coded as dummies and for characteristics were included. The multiple models were further reduced by AIC. Regression models were estimated on cases with no missing values on covariates.
Statistical analyses were performed in IBM SPSS V25 and SAS 9.4. Results with p-values below 0.05 were considered statistically significant.



Six-hundred and ninety-four residents and 1161 NH staff from 175 NH units from 47 NH, from 4 counties in Norway were included in the study.
To compare age and gender of participants vs. non- participants, 38 of the 47 NH collected data on all residents eligible for inclusion. Of 1331 eligible residents in the 38 NH, 607 were included and 724 did not participate (205 declined participation, 191 died before inclusion took place, and 328 was not included for unknown reasons). The mean age of participants was 84.4 years (SD 7.5), while for non-participants it was 83.6 years (SD 9.3) (p = 0.048); 64.4% of participants and 56.6% of non-participants were women (p = 0.004).
Table 1 presents resident, unit, and staff characteristics at baseline. The residents had a mean age of 84.4 years, 64% were women, 52.6% had poor/fair physical health, and the mean CDR-SOB was 10.3.

Table 1
Characteristics of residents, unit, and staff variables at baseline

N varies between variables due to missing data; * not married; including singles, widowed, and residents divorced or separated opposed to married; including residents being married or living with a partner; ** Nursing home unit was defined as a group of residents living together with a common living area and having their own care staff during the daytime; *** Staff variables aggregated at unit  n= 160/175; SD= standard deviation; GMHR= General Medical Health Rating Scale (excellent, good, fair, poor); CDR-SOB= Clinical Dementia Rating Scale sum of boxes (range 0 – 18); MOBID-2= Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (range 0 – 10); PSMS= Physical Self-Maintenance Scale (range 6 – 30); NPI agitation= sum of agitation/aggression, disinhibition, and irritability (range 0 – 36); NPI psychosis= sum of delusions and hallucinations (range 0 – 24); NPI Affective= sum of depression and anxiety (range 0 – 24); SCUEQS= Special Care Unit Environmental Quality Scale (range 0 – 41); P-CAT= Person-centered Care Assessment Tool (range 13-65); QPS-Nordic= General Nordic Questionnaire for Psychosocial and Social Factors at Work; QPS-Nordic subscales each consist of 3 or 4 items (range 3-15 or 4-20)


Figure 1
QUALID scale sum scores by trajectory groups

QUALID: Quality of life in Late-Stage dementia scale, ranging from 11 to 55


Trajectories of QoL

Eight residents were excluded from the growth mixture model due to missing baseline QoL data, leaving 686 residents for the analysis. Three QoL groups of residents following similar trajectories were identified (Table 2 and figure 1). All three groups exhibited a decline in QoL over 30 months. The group with best QoL named “good QoL” (n = 368, 53.6%) experienced a small deterioration in QoL from 16.1 (see intercept table 2) to 18.2 points at 30-month follow-up. Another group named “moderate QoL” (n = 226, 32.9%) experienced a deterioration in QoL from 21.0 to 25.5 points at 30-month follow-up. The last group, named “poor QoL” (n = 92, 13.4%), experienced a stable QoL, from 30.6 to 31.5 points, during the first 24 months, and thereafter remaining stable until 30 months. Table 2 presents the demographic and clinical characteristics within the three groups.

Table 2
Parameter estimates of growth mixture model and residents’ characteristics within three groups of QoL *

* only cases with no missing item at baseline QUALID sum included; ** not married; including singles, widowed, and residents divorced or separated opposed to married; including residents being married or living with a partner; QoL= Quality of Life; SD= standard deviation; GMHR= General Medical Health Rating Scale (excellent, good, fair, poor); CDR-SOB= Clinical Dementia Rating Scale sum of boxes (range 0 – 18); MOBID-2= Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (range 0 – 10); PSMS= Physical Self-Maintenance Scale (range 6 – 30); NPI agitation= sum of agitation/aggression, disinhibition, and irritability (range 0 – 36); NPI psychosis= sum of delusions and hallucinations (range 0 – 24); NPI Affective= sum of depression and anxiety (range 0 – 24)


Table 3 presents the results of the nominal regression model, with group-belonging as the outcome variable and characteristics measured at baseline as explanatory variables. In the AIC-reduced multiple model, more severe dementia, more pain (MOBID-2), and more severe affective symptoms at baseline were associated with higher odds of belonging to the group named moderat QoL and poor QoL groups compared to the group named good QoL, while more severe symptoms of agitation at baseline was associated with higher odds of belonging to the group named  poor QoL as compared to the group named  good QoL.


Table 3
Nominal regression model with QoL (QUALID score) group-belonging as outcome variable

Good QoL group as reference. N=561 (N=305 in good QoL group, N=182 in moderate QoL group, N=74 in poor QoL group), cases with at least one missing value on covariates were excluded


Variables associated with QUALID score over time

Table 4 presents the results of the linear mixed model for the associations between the residents’ QUALID score and the resident, staff, and unit characteristics measured simultaneously or, if not possible, at baseline. More severe dementia, more pain, lower ADL-function, and more severe NPS (except for the NPI psychosis sub-syndrome) were associated with decreasing QoL in the AIC-reduced multiple model. Better job satisfaction among the staff was associated with increasing resident QoL during the observation period.

Table 4
Linear mixed model with QoL (QUALID score, ranging from 11 to 55) sum as dependent variable
(N=1942, based on the number of residents assessed with QUALID at the six time points).
Cases with at least one missing value on covariates were excluded

GMHR= General Medical Health Rating Scale (excellent, good, fair, poor); CDR-SOB= Clinical Dementia Rating Scale sum of boxes (range 0 – 18); MOBID-2= Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (range 0 – 10); PSMS= Physical Self-Maintenance Scale (range 6 – 30); NPI agitation= sum of agitation/aggression, disinhibition, and irritability (range 0 – 36); NPI psychosis= sum of delusions and hallucinations (range 0 – 24); NPI Affective= sum of depression and anxiety (range 0 – 24); SCUEQS= Special Care Unit Environmental Quality Scale (range 0 – 41); P-CAT= Person-centered Care Assessment Tool (range 13-65); QPS-Nordic= General Nordic Questionnaire for Psychosocial and Social Factors at Work; QPS-Nordic subscales each consist of 3 or 4 items (range 3-15 or 4-20)



In the present study, we followed 694 nursing home residents for 30 months after admission to a nursing home. Three resident groups following distinct trajectories in QoL, as measured with the QUALID sum score, were identified. The majority of the residents belonged to the group named good QoL over time. This group and the group named poor QoL had the most stable QoL through the study period, although all three groups exhibited some decline in QoL. Furthermore, the study showed that more severe dementia, more pain, poorer ADL-function, and more severe NPS, as assessed throughout the study period, were associated with an overall decrease in QoL.
Several studies have shown an association between different resident characteristics and QoL (27, 28). The pattern is clear: more severe dementia, poorer ADL function, and more severe NPS—especially more severe depression and anxiety symptoms—are all associated with worse QoL (29). Our results are in line with these previous findings. Thus, to improve the residents’ QoL, it is important to treat or prevent pain and NPS, especially depression and anxiety, and to improve or maintain ADL function.
We found that higher job satisfaction among the staff at baseline was associated with an overall increase in the residents’ QoL. Previous studies have reported that job satisfaction is associated with quality of care (30), but to our knowledge no other studies have investigated the influence of staff job satisfaction on residents’ QoL. This is an important finding, underpinning the urgency of focusing on job satisfaction and retention of NH staff to meet the increasing need for qualified NH staff, due to an aging society. We expected that a higher level of PCC would improve the residents’ QoL, as a guiding principle for good quality of care (10, 11), but found no such association in the present study. However, in a previous study based on the same data, we found that the staff’s job satisfaction was positively associated with PCC (15), and efforts to improve the staff’s job satisfaction may improve both the PCC given at the ward and the residents’ QoL.
The importance of the physical environment is increasingly recognized as a therapeutic resource in NH (13). However, the present study was not able to demonstrate an association between the physical environment and the residents’ QoL. This may be due to the homogeneity between NH units included in the study, the characteristics of the assessment tool used, or the sample size. However, four staff and unit characteristics (staff at daytime/resident ratio, nurse educator at unit, QPS-N quantitative demands and fair leadership) were kept in the AIC-reduced model as important covariates, indicating that this factors contribute to the better model fit.
Interestingly, except for the staff’s job satisfaction, the only characteristics associated with the residents’ QoL were individual resident characteristics. This association was found both in the growth mixture and the linear mixed models, which strengthens the result. Other studies have demonstrated that there are associations between staff variables and the quality of care, influencing the resident’s QoL (9). Studies have also shown that PCC are positively associated with the staff’s job satisfaction (30), which may improve NPS in persons with dementia and can reduce NPS in people with dementia living in NH (12). Improving ADL function would be benefitical for the patients and improving their QoL.

Strengths and limitations

A strength of the study is that residents are included at admission to the NH, and the longitudinal follow-up design of the study with analysis of the residents’ characteristics and QoL at the same time points. The linear mixed model analysis captures how the residents’ characteristics change over the follow-up period simultaneously with their QoL changes. The same characteristics associated with poor QoL were found for both baseline characteristics and longitudinally assessed characteristics, and the two independent analyses confirming the same associations istrengthens the result.
There are several limitations in the study. The complex causal pathways between unit characteristics, staff characteristics, and resident characteristics are a limitation when designing a study that investigates which factors are associated with the residents’ QoL. Another limitation is the fact that the residents did not rate their own QoL; rather, we had to rely on proxy report by the NH staff, which could possibly lead to biased data. When proxies rate a person as having reduced function in ADL and NPS, they may also assume that the person’s QoL is reduced. It is reported that people with dementia rate their QoL higher than proxies do (2). However, the proxies assessed the residents’ QoL with a standardized and validated questionnaire, that is used internationally to rate QoL, which should help to reduce the subjectivity in the rating.
Patient characteristics were assessed every six month, but units and staff characteristics were assessed only once, at baseline, which is a limitation in the study. While the unit characteristics would be rather stable over the follow-up period, staff characteristics could change due to increased or decreased knowledge or new staff recruiting.
Finally, limitations due to the data collection procedure should be noted, as a high number of project nurses collected data on resident characteristics, which could in turn excess heterogeneity of the data. Furthermore, the study cohort may not accurately represent the general NH population, as information from 38 of the 47 NH revealed that there were more women in the included-residents group than in the eligible-but-not-included residents group (14, 15).



Overall, the majority of residents belonged to the group named good QoL over the observation period of 30 months. Residents in the group named poor QoL was associated with having more pain, more severe dementia, more affective symptoms, and living in a unit with poorer staff job satisfaction at baseline, as well as more pain, poorer ADL function, and more severe NPS measured simultaneously. As dementia is a chronic disease, focus on symptom relief and QoL is important. Efforts that focus on reducing pain, reducing NPS, and improving ADL function for the resident, as well as improving the job satisfaction of the staff may be important factors to improve resident QoL.


Conflict of interest: None.

Description of authors’ roles: I. Røen designed the study, collected the data and wrote the paper. J. Šaltytė Benth performed the statistical analysis and assisted with writing the paper. Ø. Kirkevold was involved in designing the study and in the statistical analysis, and assisted in writing the paper. I. Testad critically revised the manuscript. G. Selbæk was involved in designing the study and critically revised the manuscript. K. Engedal critically revised the manuscript. S. Bergh was involved in designing the study, supervised the data collection and assisted with writing the paper. All authors read and approved the final version of the manuscript.

Funding: The Norwegian Health Directorate provided funding for the data collection. The first author’s Ph. D. study was funded by the Research Counsil of Norway.The sponsors had no role in the design and conduct of the study; in the collection, analysis, and interpretation of data; in the preparation of the manuscript; or in the review or approval of the manuscript. This work was supported by the by the Norwegian Health Directorate and the Research Council of Norway.

Acknowledgments: The REDIC-NH study was administrated by the Centre for Old Age Psychiatric Research, Innlandet Hospital Trust, and was initiated by the Norwegian Health Directorate, which also provided funding for the data collection. The first author’s Ph. D. study was funded by the Research Council of Norway.

Contributions of others who did not merit authorship but participated in the research: Nursing Homes in Hedmark, Oppland, Hordaland, and Nord-Trøndelag counties participated in the study. We would like to thank the residents and their next of kin for participating in the study and for giving us their information. We would also like to thank the nursing home managers for their cooperation, the staff members in the nursing home that filled out the questionnaires, and the research nurses that collected the data.



1.    Wang SY, Shamliyan TA, Talley KM, Ramakrishnan R, Kane RL. Not just specific diseases: systematic review of the association of geriatric syndromes with hospitalization or nursing home admission. Archives of gerontology and geriatrics. 2013;57(1):16-26.
2.    Moyle W, Fetherstonhaugh D, Greben M, Beattie E. Influencers on quality of life as reported by people living with dementia in long-term care: a descriptive exploratory approach. BMC geriatrics. 2015;15:50.
3.    Lawton MP. Quality of life in Alzheimer disease. Alzheimer disease and associated disorders. 1994;8 Suppl 3:138-50.
4.    Holopainen A, Siltanen H, Pohjanvuori A, Makisalo-Ropponen M, Okkonen E. Factors associated with the quality of life of people with dementia and with quality of life-improving interventions: Scoping review. Dementia (London, England). 2017:1471301217716725.
5.    Klapwijk MS, Caljouw MA, Pieper MJ, van der Steen JT, Achterberg WP. Characteristics Associated with Quality of Life in Long-Term Care Residents with Dementia: A Cross-Sectional Study. Dementia and geriatric cognitive disorders. 2016;42(3-4):186-97.
6.    Oudman E, Veurink B. Quality of life in nursing home residents with advanced dementia: a 2-year follow-up. Psychogeriatrics : the official journal of the Japanese Psychogeriatric Society. 2014;14(4):235-40.
7.    van der Zon A, Wetzels RB, Bor H, Zuidema SU, Koopmans R, Gerritsen DL. Two-Year Course of Quality of Life in Nursing Home Residents with Dementia. The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry. 2018.
8.    Reimer MA, Slaughter S, Donaldson C, Currie G, Eliasziw M. Special care facility compared with traditional environments for dementia care: a longitudinal study of quality of life. Journal of the American Geriatrics Society. 2004;52(7):1085-92.
9.    Anderson K, Bird M, MacPherson S, Blair A. How do staff influence the quality of long-term dementia care and the lives of residents? A systematic review of the evidence. International psychogeriatrics. 2016;28(8):1263-81.
10.    Brooker D. What is person-centered care in dementia? Reviews in Clinical Gerontology. 2004;13(3):215-22.
11.    Li J, Porock D. Resident outcomes of person-centered care in long-term care: a narrative review of interventional research. International journal of nursing studies. 2014;51(10):1395-415.
12.    Testad I, Corbett A, Aarsland D, Lexow KO, Fossey J, Woods B, et al. The value of personalized psychosocial interventions to address behavioral and psychological symptoms in people with dementia living in care home settings: a systematic review. International psychogeriatrics. 2014;26(7):1083-98.
13.    Chaudhury H, Cooke HA, Cowie H, Razaghi L. The Influence of the Physical Envirnment on Residents With Dementia in Long-Term Care Settings: A Review of the Empirical Literature. The Gerontologist. 2017;Vol. 00(00):1–13.
14.    Roen I, Selbaek G, Kirkevold O, Engedal K, Testad I, Bergh S. Resourse Use and Disease Couse in dementia – Nursing Home (REDIC-NH), a longitudinal cohort study; design and patient characteristics at admission to Norwegian nursing homes. BMC health services research. 2017;17(1):365.
15.    Roen I, Kirkevold O, Testad I, Selbaek G, Engedal K, Bergh S. Person-centered care in Norwegian nursing homes and its relation to organizational factors and staff characteristics: a cross-sectional survey. International psychogeriatrics. 2017:1-12.
16.    Weiner M, Martin-Cook K, Svetlik D, Saine K, Foster B, Fontaine C. The quality of life in late-stage dementia (QUALID) scale 2000 [12818023]. 114-6].
17.    Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clinical scale for the staging of dementia. The British journal of psychiatry : the journal of mental science. 1982;140:566-72.
18.    O’Bryant SE, Waring SC, Cullum CM, Hall J, Lacritz L, Massman PJ, et al. Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: a Texas Alzheimer’s research consortium study. Archives of neurology. 2008;65(8):1091-5.
19.    Husebo BS, Strand LI, Moe-Nilssen R, Husebo SB, Snow AL, Ljunggren AE. Mobilization-Observation-Behavior-Intensity-Dementia Pain Scale (MOBID): development and validation of a nurse-administered pain assessment tool for use in dementia. J Pain Symptom Manage. 2007;34(1):67-80.
20.    Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. The Gerontologist. 1969;9(3):179-86.
21.    Lyketsos CG, Galik E, Steele C, Steinberg M, Rosenblatt A, Warren A, et al. The General Medical Health Rating: a bedside global rating of medical comorbidity in patients with dementia. Journal of the American Geriatrics Society. 1999;47(4):487-91.
22.    Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology. 1994;44(12):2308-14.
23.    Selbaek G, Engedal K. Stability of the factor structure of the Neuropsychiatric Inventory in a 31-month follow-up study of a large sample of nursing-home patients with dementia. International psychogeriatrics. 2012;24(1):62-73.
24.    Edvardsson D, Fetherstonhaugh D, Nay R, Gibson S. Development and initial testing of the Person-centered Care Assessment Tool (P-CAT). International psychogeriatrics. 2010;22(1):101-8.
25.    Dallner M, Elo, A.-L., , Gamberale F, Hottinen V, Knardahl S, Lindström K. Validation of the general Nordic questionnaire (QPSNordic) for psychological and social factors at work. Copenhagen; 2000.
26.    Sloane PD, Mitchell CM, Weisman G, Zimmerman S, Foley KM, Lynn M, et al. 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. The journals of gerontology Series B, Psychological sciences and social sciences. 2002;57(2):S69-78.
27.    Mjorud M, Rosvik J, Rokstad AM, Kirkevold M, Engedal K. Variables associated with change in quality of life among persons with dementia in nursing homes: a 10 months follow-up study. PLoS One. 2014;9(12):e115248.
28.    Naylor MD, Hirschman KB, Hanlon AL, Abbott KM, Bowles KH, Foust J, et al. Factors Associated With Changes in Perceived Quality of Life Among Elderly Recipients of Long-Term Services and Supports. Journal of the American Medical Directors Association. 2016;17(1):44-52.
29.    Beerens HC, Zwakhalen SM, Verbeek H, Ruwaard D, Hamers JP. Factors associated with quality of life of people with dementia in long-term care facilities: a systematic review. International journal of nursing studies. 2013;50(9):1259-70.
30.    van den Pol-Grevelink A, Jukema JS, Smits CH. Person-centred care and job satisfaction of caregivers in nursing homes: a systematic review of the impact of different forms of person-centred care on various dimensions of job satisfaction. International journal of geriatric psychiatry. 2012;27(3):219-29.



S.L. Tsai1, J.F. Stocker2, C.H. Tsai3, S.H. Yeh4


1. Department of Nursing, Chang Gung University of Science and Technology, Chiayi Campus, Chiayi, 61363, Taiwan; 2. Department of Medicine, National Cheng Kung University, Tainan city, 70101 Taiwan; 3. Department of Life-and-Death Studies, Nanhua University, Dalin Township, Chiayi, Chiayi County 62249, Taiwain; 4. Institute of Long-Term Care, Mackay Medical College, New Taipei City 252, Taiwan. Corresponding author: Shu-Hui Yeh, Institute of Long-Term Care, Mackay Medical College, No.46, Sec. 3, Zhongzheng Rd., Sanzhi Dist., New Taipei City 252, Taiwan, Tel: 886-2-26360303 ext. 1800, E-mail: yehshuhui@mmc.edu.tw

Jour Nursing Home Res 2019;5:1-7
Published online January 28, 2019, http://dx.doi.org/10.14283/jnhrs.2019.1



Background: The number of people living in long-term care (LTC) facilities has been rising in many parts of the world, and most current residents will end their lives in LTC facilities. The perceptions of residential care and practices in most current research of nursing homes (NHs) in Taiwan are based on evidence from an ego-centric rather than socio-cultural-centric model. Objectives: This study was designed to address the overlooked cultural aspect in NHs research and thereby advance understanding of how the NH staff in an East Asian setting perceive NH resident death and dying. Design: A qualitative study was designed in line with the hermeneutic phenomenological method. Data were collected via in-depth semi-structured interviews. Setting: The research was conducted in five hospital affiliated nursing homes and seven independent nursing homes in central and southern Taiwan. Participants: Through purposive sampling, twenty-five participants were recruited for interview, twelve registered nurses (RNs) and thirteen nursing aides (NAs). Measurements: An interview guide was used to produce the digitally-recorded contents, which was then transcribed verbatim and translated. The hermeneutic phenomenological analysis was conducted such that authors went back and forth through every interview text (parts) and the research questions (whole) until they reached a comprehensive understanding of the subject matter in terms of reduced, emerging themes. Results: Four themes were identified in the data analysis. They were ‘impact of a resident’s death,’ ‘reflections on entangled feelings,’ ‘insufficiencies.’ and ‘tremendous pressure of informing death.’ Conclusion: This qualitative study confirmed previous findings of non-Asian studies about the significance of ‘assessment of dying’ and ‘family communication’ in quality NH care. In addition, NH nurses were in need of palliative training in dying care. The nurses’ felt pressure due to family requests to send residents home for their ‘last breath,’ which was both the nurses’ most challenging work of care and the most culturally grounded aspect of it.

Key words: Death and dying, nursing home, nursing staff, phenomenology.




Concomitant with the growing elderly populations in many countries, the number of people living in long-term care (LTC) facilities has been rising in many parts of the world, and most current residents will end their lives in LTC facilities (1). Nursing homes (NH), as a particular type of LTC, will be important for end-of-life care for the foreseeable future. However, the research literature indicates that NHs often provide sub-standard care due to high staff turnover, high nurse-patient ratios and stressful environments (2). The result is end-of-life care of questionable quality for too many residents. In 2018 Taiwan is faced with a similar situation where the ratio number of elderly has reached 14.5%, the threshold of an aged society (3). The need for long-term care will also soar up to a historical high level.
In 2015 Taiwan was ranked 6th out of 80 countries worldwide, and 1st in Asia, in end-of-life care (4). While this could indicate end-of-life palliative care in Taiwanese NHs is well-developed, evidence shows that this might not be the case. A comparison of 2012 and 2016 figures revealed a marked increase in the number of NHs (447-511, 14%) and residents (27,056-36,020, 33%) (3). According to the official regulation, there were no on-site physicians, and the ratio of registered nurses (RNs) and nursing aides (NAs) was about 1:3, with one RN for every 15 residents, and one NA for every 5 residents (5). Importantly, palliative care training is not provided to NH staff in Taiwan. Therefore, NH personnel whose main responsibility is end-of-life care lack appropriate skills in facing the death of residents. In fact, most end-of-life care in Taiwan takes place in hospices (mostly within hospitals), at home [with the help of foreign caregivers (due to the shortage of less expensive manpower)] or in emergency rooms.
A significant number of studies have focused on the provision of palliative and end-of-life care in NHs in North American or European countries (6). Researchers have paid little attention, however, to the end-of-life palliative care carried out specifically by NH nursing staff, or even wondered whether or not they have received such training—perhaps under the assumption that every facility has an end-of-life care trainer (7,8).  Therefore, the main objective of the current research is to explore NH nursing staff perceptions of death when faced with dying residents in Taiwanese NHs where nursing staff receive no palliative training.
There is a growing body of research focused on the emotional and stress-related experiences of NH nursing staff surrounding resident death and dying. The literature indicates that negative emotions and a sense of personal loss are common when residents die (9). The impact of resident death has been categorized as a type of ‘moral distress’ (10). Other studies have treated resident death as a clinical issue from which nursing roles in the dying care were examined (11). Further, specific clinical problems have been noted as NH nursing staff attend to the dying of residents. Among them, ‘assessing dying’ has been considered very important for NH carers (12). Also, communicating about death is a major source of stress, apparently leading many nurses to seek emotional detachment.
A variety of intervention programs have been proposed in order to provide systematic training for ‘assessing dying.’ The Liverpool Care Pathway (LCP) and the Gold Standards Framework for Care Homes (GSFCH) are notable examples in the UK. NHs have implemented programs either individually or in an integrated fashion (13, 14). Other programs have focused on palliative care skills, such as in a project in Canadian NHs (12), or by improving pedagogical methods, such as in a Danish project (13). Nonetheless, the clinical effectiveness of these approaches remains a debatable issue. Some major challenges in implementing these programs have been identified and discussed (15).
Another area of research deserving more attention is the cultural dimension. There is limited research available to help caregivers in NHs deliver high quality care in Chinese cultural contexts because research perceptions of residential care and practices in East-Asian societies are typically constructed by evidence from ego-centric rather than socio-cultural-centric studies (Lee 2010) (16). There is relatively scarce nursing science literature focusing on the Chinese cultural context. This issue of death demonstrates the importance of addressing the sociocultural values of specific societies in residential care provision. In the current research, we explore our assumption that elucidating Taiwanese NH nursing staff perceptions of death when faced with dying residents in NHs is to some extent shaped by local Taiwanese cultural values.
Chan and Kayser-Jones drew upon the few nursing home studies that deal with cultural issues to develop their study of the experiences of care for terminally ill ethnic Chinese residents in the US, identifying various factors that influence end-of-life care (17). According to this research, the most significant factors influencing the care Chinese residents received were communication barrier, dislike of Western food, and the differing cultural beliefs and customs. These findings help develop and implement interventions to improve the terminal care of Chinese elderly individuals in nursing homes. Another study described the culturally-embedded meaning of death through coping acts of NH nursing staff seen in normally undocumented care given to dying residents (18). That study revealed the import of behaviors of socialization, interaction, communication and superstition between staff and dying residents.
East Asian countries influenced by Chinese culture share a death-denying cultural context (19) in which people tend to adopt religious ceremonies that help them avoid bad luck or death. Past research has shown that Chinese death beliefs, which mostly consisted of traditional religious thoughts, are related to death anxiety. It was further confirmed these ˝superstition˝ were the only factor that could predict death anxiety in the Chinese context (20). As an example, in Taiwan, the “zhougian” ritual is an exorcism constitutive of a process of regaining the lost spirit and soul of a person who has been frightened (21). This cultural aspect is by and large overlooked in NH research. As the secondary objective, this study will also attempt to explore how the NH staff in an East Asian setting perceives NH resident death and dying.



This qualitative study analyzes 25 NH nursing staff members’ perceptions of the essential meaning of death through narrative reports of their end-of-life care experiences. It adopted a  research approach informed by interpretive phenomenology. This approach was selected because the researchers were concerned about the subjective meaning of residents’ death to the nursing staff. Hermeneutic phenomenology aids in a deep understanding of the human condition by structuring the study of “how people interpret their lives and make meaning of what they experience” (22). This method allows the exploration of the essential meaning of how NH nursing staff perceives resident dying and death as they provide end-of-life care.
Five Hospital Affiliated Nursing Homes & seven Independent Nursing Homes in central and southern Taiwan were included. Purposive sampling was used. The inclusion criteria were as follows: Participants (i) provided direct care for dying residents, (ii) worked in an NH in the targeted region, (iii) had worked on the current unit as a full-time NH nursing staff member for at least 3 months, and (iv) could communicate in Chinese fluently.
The research was approved by the Institutional Review Broad (IRB) of a hospital. Participants were interviewed after they had each signed an informed consent form. They were informed that they were free to withdraw at any time. To ensure anonymity, pseudonyms were adopted.
Data were collected through semi-structured interviews in Chinese. An interview guide (see Table 1) was used to conduct a 50-minute dialogue, on average, with each participant. The digitally-recorded content was then transcribed verbatim and translated. The hermeneutic phenomenological analysis followed that of Cohen (22). The authors (SLT, CHT) went back and forth through every interview text (parts) and the research questions (whole) until they reached a comprehensive understanding of the subject matter in terms of reduced, emerging themes.
To ensure research rigor, the study also followed Packer and Addison’s ‘four evaluation approaches’ for interpretive research (23). The research addressed the risk of bias by remaining close to the original text, and by uncovering biases for scrutiny. Texts with analyzed themes were shown to participants to make sure they fit their subjective experience. Then, the text was read inter-subjectively among the author-researchers to reach a coherent interpretation of the meaning of resident death for NH nursing staff.



There were 25 participants, including 12 RNs and 13 NAs. Ages ranged from 22 to 60 years old (average age = 45 years). Relevant characteristics are listed in Table 1. Four themes were identified in the data analysis. They were ‘impact of a resident’s death,’ ‘reflections on entangled feelings,’ ‘insufficiencies,’ and ‘tremendous pressure of informing death.’ The essential meaning of NH resident death and dying for nursing staff can be summarized as being a strong impact that has made them aware of the need for proper education on end-of-life care, especially incorporating cultural background.

Table 1
Background of participants

*IHN( independent nursing home); *HANH( hospital affiliated nursing home)


Impact of a resident’s death

This first theme was revealed in a common situation where participants felt ill-prepared for a resident’s death. The RNs and NAs in this study had received no palliative training. As a result, participants reported substantial emotional shock when faced with the unexpected death of a resident. A new NA, Kwei, described her shock:
It’s just that a grandma over 90, who was pleased with us, suddenly stopped breathing one morning. There was no heart-beat, no blood pressure. I was frightened to death. I hurried to get help from the nurses. (Kwei, NA)
Unexpected death also troubled experienced nursing aides. Ya, despite being in her 9th year on the job, struggled to care for a resident with paralysis and breathing trouble due to the latter’s old age and lack of educational background (graduate of elementary school) and proper training in palliative care. Major shock holds true for registered nurses, too. Lili offered this experience:
A four-year old boy was diagnosed with cerebral paralysis (CP) and had just checked in … for one or two days. I remembered he came in the daytime. When I was on the night shift, I found his skin color was abnormal. It had turned purple. We immediately gave him CPCR and sent him to the hospital. It was too late, though. He was pronounced dead on arrival. I kept sobbing on my way back to the NH. (Lili, RN)
NH nursing staff also described suffering from nightmares, especially when the dying process was a struggle or not peaceful. Ding’s nightmares hinted at trauma suffered during her first resident death:
I had no experience dealing with that. I had nightmares in which I heard him calling my name (in my ears) all night. I was afflicted . . . . That was my first dying case. His last sounds were full of suffering and sorrow. (Ding, NA)
She thought the death might be her fault. The patient did not have the good death Ding had expected for him. The result was that she was haunted by the resident’s death for a long time. It had to do with her sense of guilt resulting from the lack of dying care experience.

Table 2
Interview guide


Reflections on entangled feelings

The second theme to emerge was ‘reflections on entangled feelings.’ Gradually, nursing staff found their worldviews becoming jaded, leading to pessimism regarding dying. This rather negative attitude towards life was perceived when dealing with residents’ circumstances:
Life is short. Watching them die like this made me feel worthless. A pet’s death may be a little more valuable than a resident’s. I just felt that people seemed to have no value or dignity as they became old. (Shu, NA)
Some participants blamed themselves because they felt guilty. However, they often worried most about possible lawsuits resident family members may file. Chung described the following:
When we informed the family about the resident’s death in the evening, they couldn’t accept it because they had just paid a visit to the resident in the morning. They suspected this death might have to do with our negligence or malpractice. I was really afraid of being sued for that and doubted my own care. The mental pressure was truly huge. (Chung, RN)
Still, some NH nursing staff managed to find some positive meaning out of their dying care in their clinical circumstances. Liu’s description puts it succinctly: ‘I altered my attitude toward life and came to realize the changeability of life’ (Liu, NA). Such thoughts apparently worked well at reducing anxiety. Hsian said:
They are ill in bed, so letting them go is better for them. This is life. This is his fate. He was suffering from chronic disease, and death would be his escape route. (Hsian, NA)
NH staff also reinforced their self-worth by telling themselves that they were doing the right thing for residents no matter what happened:
In my opinion, this job is charity work. We do a lot that the residents’ families do not know about. I tell myself, ‘I am doing well. I am helping them.’ (Lili, RN)


The third theme was awareness of their own insufficiencies. This was revealed in part when participants expressed a strong need to gain palliative care skills. Nursing staff often found ways to improve death care. The most prominent learning avenue was through gaining experience:
I knew I had to check on the resident’s vital signs after I found he/she had some physical syndrome. Because I have had such an experience before, I knew how to deal with it when a similar situation occurred. (Feng, NA)
Assessing signs of dying was the training they most desired. For example, Hsian described constantly checking the vital signs of a resident with liver cancer during the terminal stage because she was afraid that the resident would suddenly die. Had there been related training, Hsian may have had less anxiety. Ming said, ‘[we] hope to know how to take good care of them’ (Ming, RN) during the terminal stage.
Truth-telling and grief-comforting were also considered important to learn. This training would involve communication with bereaved families, a resident’s roommates, and comforting NH colleagues. In contrast, the usual response was avoidance:
In facing an empty bed, everyone sensed the atmosphere of grief. It seemed I had to do something, but I didn’t know what to say, how to comfort them [roommates of the resident]. The deceased resident seemed to mirror their own future. (Susu, R.N.)
It is in this context that NH nursing staff most strongly expressed the need to add dying care into their training program. With no physicians in NHs, nursing staff are expected to assume greater responsibilities than those in European and North American countries, especially around signs of dying and the death event. Monitoring the dying is a difficult task for already over-worked staff, and any improvement in this area would seem to rest upon providing specific palliative training.

Tremendous pressure of informing death

The fourth theme is ‘tremendous pressure of informing death.’ In Taiwanese NHs, resident families demand that nursing home staff inform them of imminent death so patients can be discharged to their homes as expected in accordance with Taiwanese Taoist-Buddhist folk custom. However, this is extremely challenging. A senior nurse said, ‘We can assess the resident’s general condition based on our experience, but we are not God’ (Yang, RN). Ming’s description helps illustrate this point:
The pressure was really incredible because we didn’t know the actual timing of their last breath. We were afraid to trouble family members if we informed them too early. We were also afraid not to be able to help the residents reach home in time if we informed them too late. (Ming, RN)
All participants acted instinctively, resorting to cultural resources first because they had not been equipped with palliative nursing training. For instance, Ding’s peculiar behavior well illustrates this point:
Since the first time encountering the death of a resident and having nightmares, I have smiled to the residents, but do not introduce myself and do not tell them my name. I just tell them to call me ‘Miss’. (Ding, NA)
She justified her reaction and shaped it based on the folk belief that by means of calling names ghosts can collect human spirits. ‘Praying to deities,’ ‘wearing a charm,’ and ‘reciting sutra or mantra’ were among the common practices of the interviewees, practicing Taiwanese folk religions. One registered nurse was frightened and said she ‘felt much better after I went to pray at a temple’ (Zhi, RN). Some participated in a ritual called ‘zhougian,’ which literally means ‘recalling the frightened soul.’ Having had nightmares about a deceased resident, Lonyu participated in ‘zhougian’ to pacify her mind:
I talked to my father-in-law, a spirit medium. He performed a ‘zhougian’ ritual for me. The quality of my sleep has been greatly improved since then. (Lonyu, RN)
Many NH nursing staff recommend this folk ritual to novice caregivers who are fearful of the dying process. For them, ‘this could at least comfort the soul’ (of the nurse) (Susu, RN).



This paper explicitly explores the cultural elements of the experiences of ethnic-Chinese Taiwanese from a hermeneutic-phenomenological perspective, and explores how this maps to, and contrasts with European and North American cultural values and analyses of this issue. The meaning of the death and dying of residents in NHs from a nursing staff perspective in Taiwan can be understood in terms of ‘impact of a resident’s death,’ ‘reflections on entangled feelings,’ ‘insufficiencies,’ and ‘tremendous pressure of informing death.’
‘Impact of a resident’s death’ indicates the disturbance felt by nursing staff when a resident is dead. This characterization fits well with ‘emotional labor’24 or ‘emotional burden’25 found in the literature, but nursing staff’s responses and adaptations in Taiwan also reflect some unique qualities of the Taiwanese cultural context. The meaning of death in Taiwanese nursing homes carries slightly different meanings when compared to European and North American countries. No physician on site, no palliative training, and cultural expectations that the dying resident will ideally die at the ancestral home define the basic setting for end-of-life care of NHs in Taiwan. A primary responsibility of NH nurses is to make a judgement on when the dying resident should be sent home or to the hospital. Poor performance in this area might be due to insufficient training on ‘assessing dying,’ which is an ambiguous and legally tenuous area of nursing practice. The pressure on nursing staff is compounded by family demands to have the resident’s final hours be at home rather than at the NH or hospital.26 Taiwan Ministry of Health and Welfare statistics in 2018 show that 36% of people died at home, a number much higher than in Western countries.27
Dying assessment is a common challenge acknowledged by NH nursing staff internationally. It is identified as an ‘important critical skill,’14 a ‘pre-established condition for commencing on a care pathway,’15 or a condition for arousing ‘anxiety’ around doing things right.15, 28 With cultural background knowledge, one begins to understand the unique meaning of dying assessment experienced by Taiwanese caregivers. To be specific, anxiety identified in other developed systems of care is about whether the NH nursing staff can do as prescribed in the adopted program such as LCP, GSFCH, etc. However, Taiwanese NH nurses feel anxiety simply because they have no such prescribed pattern to follow. By the same token, the second theme, ‘reflections on entangled feelings’ also revealed how reactions are circumscribed by cultural dynamics.
While some NH administrators expressed the view that NHs can provide effective care without palliative care training, most studies support the view that NH staff need to enhance palliative care skills.29 Nevertheless, since little is known about the care provided by nursing staff with no palliative training for NHs in European and North American countries, it is possible that the established training programs in European and North American countries would fall short if offered as-is in Taiwan. Without palliative training, some studies suggest NH staff would default to culturally-based ritual-like behavior not prescribed by the nursing profession. Examples include ‘opening the window to let the spirit go’ in an NH in the US,18 or ‘made a sign of the cross’ as a patient passed away in an ICU in Rome.30 Such rituals are embedded in the local culture. In this study, when a nurse wished not to be called by name, she rushed to Taoist ritual exorcisms for help and recited a sutra or mantra to pacify her mind demonstrating typical non-prescribed reactions of NH nursing staff in Taiwan. These typify the cultural resources relied on by staff to improve care capacity and lift the burden place on them in the absence of palliative training.
Regarding insufficiences felt by NH nursing staff, researchers suggest NH nursing staff should be offered continuous education and support regarding palliative care (31). Monitoring the dying is a difficult task for already over-worked staff, and any improvement in this area would seem to rest upon providing specific palliative training. Though not described here, a recent study provides an illustration of the need for palliative training for nursing assistants (32). Though participants in the present study unanimously hope for palliative training, continuous education for NH nursing staff in Taiwan has many obstacles to overcome such as labor shortages, tuition fees, a high turnover rate, and a lack of a pedagogical approach to training that includes consideration of cultural background.
It appears that the fear NH nurses have of lawsuits is closely related to the family’s request of sending the dying resident home for the last breath. It has partly to do with their lack of proper palliative training. However, the findings suggest that the cultural dimension of the meaning of death and dying should be evaluated prior to any development of palliative training. Further, it suggests that education applied in Taiwan, where no palliative end-of-life care in NHs is traditionally provided, should be conducted with sensitivity to cultural nuances such as those elucidated in this study in order to give it the greatest chance of success.
Truth-telling and grief-comforting were also considered important. This training would involve communication with bereaved families, a resident’s roommates, and comforting NH colleagues. Bereaved Taiwanese family members have a unique way of expressing grief in that family ethics always plays a more important role than individual concerns. Therefore, any training program for Taiwan must take into consideration such cultural behaviors, as suggested by a study of cultural factors in NHs.20 The current study further points to a strong need to develop such a culturally well-rounded program. The benefits of greater understanding of embedded cultural factors in NHs warrants further research.
Finally, the major limitation of a hermeneutic phenomenological study is that, while it can give a better understanding of what the issues are and how they come to form the current situation, it will not necessarily aid in prediction.



This qualitative study explored how NH nurses in Taiwan perceive the meaning of NH dying and death. It confirmed the previous findings of non-Asian studies about the significance of ‘assessment of dying’ and ‘family communication’ in quality NH care. It also found that the palliative training of nurses for caring the dying resident is much in need. Attention should be paid to the pressures put on nurses’ by family requests, based on folk belief to send dying residents home for the last breath. While this unique situation of dying care holds true for NH nurses in Taiwan, palliative training programs for NH nurses existing in countries such as U. K. and U.S. are still worthy of application, in modified form, to meet the explicitly-expressed needs of nurses in Taiwan.


Acknowledgement: The thanks of the authors go to those participating NHs nursing staff for the time they spent and the experiences they shared in this research.

Conflict of Interest: No conflict of interest has been declared by the authors.

Ethical standards: The research was approved by the Institutional Review Broad (IRB) of a hospital. Each of the participants signed an informed consent form.



1.    Luta X, Panczak R, Clough-Gorr K, et al. Dying among older adults in Switzerland: who dies in hospital, who dies in a nursing home? BMC Palliative Care 2016;15:1-11
2.     Eliopoulos C., Guest Editorial: Cultural transformation in nursing homes. International journal of older people nursing. 2010;5:1-2.
3.    Ministry of Health and Welfare. Available from URL: https://1966.gov.tw/LTC/cp-3636-38462-201.html.Accessed 20 May 2018
4     Economist Intelligence Unit. The quality of death: Ranking end-of-life care across the world: Economist Intelligence Unit; 2010. Available from URL: http://graphics.eiu.com/upload/eb/qualityofdeath.pdf. Accessed 20 May 2018
5     Laws & Regulations Database of The Republic of China. The Criteria of Classification of establishing Nursing Institute. Available from URL: http://law.moj.gov.tw/LawClass/LawAll.aspx?PCode=L0020035. Accessed 21 May 2018
6.    Kinley J, Froggatt K, Bennett M. The effect of policy on end-of-life care practice within nursing care homes: A systematic review. Palliative Medicine 2013;27:209-220
7.     Seymour J. Looking back, looking forward: the evolution of palliative and end-of-life care in England. Mortality. 2012;17:1-17
8.     Kaasalainen S, Ploeg J, McAiney C, Martin LS, Donald F, Martin-Misener R, et al. Role of the nurse practitioner in providing palliative care in long-term care homes. International journal of palliative nursing. 2013;19:477-85
9 .    Jenull B & Brunner E., Death and dying in nursing homes: a burden for the staff? Journal of Applied Gerontology, 2008:27,166-180.
10.    Lokker M, Swart S, Rietjens J, van Zuylen L, Perez R, van der Heide A. Palliative sedation and moral distress: A qualitative study of nurses. Applied Nursing Research. 2018;40:157-161
11.     Hov R, Athlin E, Hedelin B. Being a nurse in nursing home for patients on the edge of life. Scand J Caring Sci. 2009;23:651-9
12.     Ellershaw J, Ward C, Neuberger RJ. Care of the dying patient: the last hours or days of lifeCommentary: a “good death” is possible in the NHS. Bmj. 2003;326:30-4
13.     Raunkiær M, Timm H. Development of palliative care in nursing homes: evaluation of a Danish project. International journal of palliative nursing. 2010;16:613-20
14.     Hockley J, Watson J, Oxenham D, Murray SA. The integrated implementation of two end-of-life care tools in nursing care homes in the UK: an in-depth evaluation. Palliat Med. 2010;24:828-38. PubMed PMID: 20663812
15.     Duffy A, Payne S, Timmins F. The Liverpool Care Pathway: does it improve quality of dying? British Journal of Nursing. 2011;20:942-6
16.     Lee D.T. Guest Editorial: Privacy and dignity in residential care homes: cross-cultural issues. International journal of older people nursing. 2010;5:3-4
17     Chan J, Kayser-Jones J. The experience of dying for Chinese nursing home residents: cultural considerations. Journal of gerontological nursing. 2005;31:26-32
18 .    Oliver DP, Porock D, Oliver DB. Managing the secrets of dying backstage: The voices of nursing home staff. OMEGA-Journal of Death and Dying. 2006;53:193-207
19.     Dong F, Zheng R, Chen X, Wang Y, Zhou H, Sun R. Caring for dying cancer patients in the Chinese cultural context: A qualitative study from the perspectives of physicians and nurses. European Journal Of Oncology Nursing .2016;21:189-196
20.     Wong S. Does Superstition Help? A Study of the Role of Superstitions and Death Beliefs on Death Anxiety amongst Chinese Undergraduates in Hong Kong. Omega: Journal Of Death & Dying .2012;65:55-70
21.     Chang H. Dajia and folk healing culture – an example of Zhougian syndrome, In Rituals, Temple Festival, and Community-Dajia, Religious Belief, and Folk Culture (Lee PR & Zhu RG eds.), Preparation Center of Institute of Chinese Literature and Philosophy, Academia Sinica, Taipei, 1996;427-457
22.     Cohen MZ, Kahn DL, Steeves RH. Hermeneutic Phenomenological Research: A Practical Guide for Nurse Researchers. SAGE Publications; 2000.
23.     Packer MJ, Addison RB. Introduction. In M.J. Packer & R. B. Addison (Eds.), Entering the Circle: Hermeneutic Investigation in Psychology. Albany: State University of New York Press, 1989;13-16
24.     Bailey C, Murphy R, Porock D. Professional tears: developing emotional intelligence around death and dying in emergency work. J Clin Nurs. 2011;20:3364-72
25.     Sorensen R, Iedema R. Emotional labour: clinicians’ attitudes to death and dying. Journal of Health Organisation and Management. 2009;23:5-22
26.     Tsao SH.A Study of End-of-Life Ethics on Funerary Customs «Hall Cleaning»-A Case about Person Breathed His Last on the Way of Home from Hospital. Monthly review of Philosophy and culture. 2006;33:57-68
27.     Ministry of Health and Welfare.Available from URL: https://dep.mohw.gov.tw/DOS/cp-3352-36826-113.html. Accessed 20 May 2018
28.     Dee JF, Endacott R. Doing the right thing at the right time. J Nurs Manag. 2011;19:186-92
29.     Seymour JE, Kumar A, Froggatt K. Do nursing homes for older people have the support they need to provide end-of-life care? A mixed methods enquiry in England. Palliat Med. 2011;25:125-138
30.     Goopy S. … that the social order prevails: death, ritual and the ‘Roman’nurse. Nursing inquiry. 2006;13:110-117
31.     Kisvetrová, H., Klugar M, & Kabelka, L., Spiritual support interventions in nursing care for patients suffering death anxiety in the final phase of life. International Journal of Palliative Nursing, 2013: 19(12),599-605)
32.     Sahlberg-Blom E, Hårsmar A, Österlind J. Assistant nurses’ descriptions of signs of dying among older people in nursing homes. Vard I Norden. Nursing Science And Research In The Nordic Countries. 2014; 33: 20-24



D.A. Harris1, G.A. Wellenius2


1. Department of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Toronto, ON, CA; 2. Department of Epidemiology, School of Public Health, Brown University, Providence, RI, USA.  Corresponding author: Daniel Harris, MPH, Department of Epidemiology, Dalla Lana School of Public Health, University of Toronto, (781) 733-5291
daniel.harris@mail.utoronto.ca, daniel_harris2@alumni.brown.edu

Jour Nursing Home Res 2018;4:64-66
Published online December 6, 2018, http://dx.doi.org/10.14283/jnhrs.2018.12



The Centers for Medicare & Medicaid Services (CMS) reported 15,634 certified nursing homes in the United States in 2014. Approximately 1.4 million older adults reside in nursing homes due to a variety of clinical and social factors. Older adults who transition into nursing home care tend to have a greater prevalence of cognitive and physical morbidities, such as cognitive impairment and chronic obstructive pulmonary disease. Given their clinical vulnerabilities, nursing home residents are at an increased risk of adverse events due to climate change. Major hurricane systems over the past several decades have contributed to significant and avoidable loss of life among nursing homes residents. Currently, evidence from both the qualitative and quantitative literatures consistently suggest that the evacuation of nursing homes residents leads to greater morbidity and mortality compared to sheltering in place due to a host of clinical and environmental factors. However, as extreme weather events intensify due to climate change, policy makers, health officials, and nursing homes will need to reassess their disaster plans amidst the increasing risk of facility damage and need for evacuation from worsening storm systems. In this commentary we review the evidence regarding the risks of sheltering in place versus evacuation during extreme weather events and propose that climate-change projections be integrated into the conversation and development of nursing home disaster preparation.

Key words: Climate change, nursing home, evacuation, shelter in place.




Older adults living in nursing homes are at an increased risk of adverse health outcomes as a result of extreme weather events such as hurricanes and severe storms. Since Hurricanes Katrina and Rita in the United States, several studies and reports have compared the health risks of evacuating nursing home residents versus sheltering in place (1-3). These studies have shown that with adequate preparations, sheltering in place tends to be safer compared to evacuation procedures, leading local health departments to advocate and prioritize this method of hurricane response (4). However, if extreme weather events become more frequent and more severe, as currently projected under climate change scenarios (5), will sheltering in place remain the preferred response to extreme weather events? In other words, what is the future generalizability of previous studies’ findings in the context of a changing climate?
Our aims in this commentary are to briefly review the history and population of nursing homes, survey past research measuring the effects of evacuation compared to sheltering in place, and ultimately propose that future legislative decisions and recommendations for nursing homes’ hurricane response be proactively informed by climate science.


The challenges of nursing home evacuation

In 2014, the Centers for Medicare & Medicaid Services (CMS) reported that there were 15,634 certified nursing homes in the United States, serving approximately 1.4 million nursing home residents (6). Notably, among the older adults living in nursing homes, 7.8% were older than 95 years old, 69.3% had three or more impairments of activities of daily living, and 61.4% had moderate to severe cognitive impairment (6). The high prevalence of poor physical and cognitive health lends insight into the medical complexity and vulnerability of older adults living in nursing homes.
The fragility of the nursing home population has facilitated a number of tragedies during past hurricanes. Evacuating a medically complex senior population is associated with several adverse health outcomes (7). As one nurse administrator expressed, maneuvering older adults into buses for transport is not only a formidable task, but also risky, as this population cannot quickly evacuate a bus during a secondary emergency (2). For example, these risks became a reality when twenty-four nursing home residents died when their evacuation bus caught fire during Hurricane Katrina (8).
The quantitative and qualitative evidence regarding the impact of evacuation on nursing home residents has been consistent in highlighting the risks and challenges of evacuating nursing home residents. One qualitative study of twenty nursing home administrators affected by Hurricane Katrina reported consistent themes of inadequate support from government response teams, challenges associated with evacuating a medically acute population, and poor staff retention (2).Similarly, a report by the Office of Inspector General (OIG) of twenty nursing homes in the Gulf States (Alabama, Florida, Louisiana, Mississippi, and Texas), revealed that evacuation led to unforeseen problems with contractual agreements with transportation services (9). Among the thirteen facilities reporting prior contracts with transportation services, five were unable to access the contracted service due to “overbooking”, as several other nursing homes contracted with the same transportation service. The OIG report also noted that nursing homes reported evacuation taking substantially longer than anticipated, with the median journey to a new host facility taking 3 hours. One nursing home detailed an arduous 19-hour evacuation, marred by numerous vehicle failures, ill residents, 100-degree temperatures, and diminishing supplies. A more recent study that quantified the increase in morbidity and mortality associated with evacuation during hurricanes found that 90-days after Hurricanes Katrina, Gustav, Rita and Ike, evacuation was associated with a 2.7% to 5.3% higher risk of death and a 1.8% to 8.3% higher risk of hospitalization compared to non-hurricane periods (3).


Sheltering in place

In contrast to evacuation, prior studies have shown that nursing homes residents who sheltered in place during extreme weather events have had better health outcomes compared to their evacuated peers (3). One qualitative study of nursing home administrators affected by Hurricanes Katrina and Rita found that only one of eleven (9%) nursing homes that sheltered in place reported being concerned about resident morbidity and mortality, whereas six of fifteen (40%) administrators who evacuated reported morbidity and mortality as a concern.
Although the results from these previous studies and reports identify the high risk associated with evacuating this vulnerable population, sheltering in place is not without danger.  The most prevalent concerns from nursing home administrators who sheltered in place during Hurricanes Katrina and Rita were facility damage, reduced staffing, and limited supplies (2). Many of the concerns and risks associated with sheltering in place are related to the local government’s recovery time and sustainability of resources within the nursing home. For example, in one Florida nursing home, staff were able to care for residents during Hurricane Irma, but limited power resources, high temperatures, and a slow emergency response after the storm contributed to the unnecessary loss of eight lives (10). Clearly, sheltering in place is not without risk, with its success often dependent on external factors, such as the storm’s relative impact and community-level disaster planning and execution. However, the many negative findings evaluating evacuation suggests that keeping residents in place is perhaps, on average the lesser of two risks, leading health systems to recommend that nursing homes prepare accordingly and shelter in place (4).


Introducing climate science into disaster preparedness

We propose that a crucial element is missing from the interpretation of past research findings and the implementation of current and future emergency preparedness policies. Increasing temperatures intensify extreme weather events, such as hurricanes, and pose significant risks to population health (especially for vulnerable populations like older adults) (11, 12). Although some nursing homes may have withstood hurricanes in the past, climate projections suggest that further increases in storm system intensity will likely render more catastrophic infrastructural damage (5). Moreover, barriers such as limited supplies and limited support from emergency services will likely worsen as infrastructural damage from severe storms becomes more widespread. Therefore, in the context of a rapidly changing climate, the debate regarding evacuation versus sheltering in place deepens in complexity, requiring substantially greater thought and preparation into the drafting of legislation, allocation of resources, and initiation of disaster plans. Addressing these challenges is not just a matter of policy or legislation. Nursing homes should consider and invest in their own emergency preparedness in advance, in the context of a diverse set of growing climate-related threats. This may include infrastructural improvements to be able to better shelter in place.
As the literature currently stands, the debate between evacuation or sheltering place during a hurricane or severe storm appears to favor the latter; however, as extreme weather events become more common and more severe due to climate change, the relative risks become markedly less clear, highlighting the need to develop policies that extend beyond past experience and take into account the hazards projected for the future. Moreover, the debate of whether or not to evacuate or shelter in place must move beyond the minds of nursing home administrators and be brought into a wider discussion of how to weigh and consider evacuation amidst worsening storm systems, especially as other climate-related threats also exist for this population (e.g., extreme heat, heavy precipitation, and displacement from sea level rise). Now is the time to begin crafting climate-informed policies, and not rely previous findings unlikely to generalize to a changing climate.


Funding: The authors have no funding source(s) to report.

Acknowledgements: The authors would like to acknowledge Geetika Kalloo for reviewing and providing thoughtful feedback on earlier drafts. Authors DAH and GAW contributed equally to the conception, writing, and revision of the commentary.

Conflict of Interest: None to declare.



1.    Dosa DM, Hyer K, Brown LM, et al. The controversy inherent in managing frail nursing home residents during complex hurricane emergencies. J Am Med Dir Assoc, 2008. 9(8): p. 599-604.
2.    Dosa DM, Grossman N, Wetle T, Mor V. To evacuate or not to evacuate: lessons learned from Louisiana nursing home administrators following Hurricanes Katrina and Rita. J Am Med Dir Assoc, 2007. 8(3): p. 142-9.
3.    Dosa D, Hyer K, Thomas K, et al. To evacuate or shelter in place: implications of universal hurricane evacuation policies on nursing home residents. J Am Med Dir Assoc, 2012. 13(2): p. 190 e1-7.
4.    Rhode Island Department of Health. Senior Resiliency Project. 2018; Available from: http://www.health.ri.gov/programs/detail.php?pgm_id=1105.
5.    Hayhoe KJ, Edmonds RE, Kopp AN, et al. Climate models, scenarios, and projections., In: Climate Science Special Report: Fourth National Climate Assessment, Volume I [Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. 2017, U.S. Global Change Research Program: Washington, DC, USA. p. 133-160.
6.    Centers for Medicare & Medicaid Services. Nursing Home Data Compendium of Medium] 2015 Last Update  [cited Access 2015; Available from: https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/CertificationandComplianc/Downloads/nursinghomedatacompendium_508-2015.pdf
7.    Moore C, Wisnivesky J, Williams S, McGinn T. Medical errors related to discontinuity of care from an inpatient to an outpatient setting. J Gen Intern Med, 2003. 18(8): p. 646-51.
8.    Belli A, Falkenberg L. 24 nursing home evacuees die in bus fire. 2005. Houston Chronicle. https://www.chron.com/news/hurricanes/article/24-nursing-home-evacuees-die-in-bus-fire-1946742.php.
9.    Leviston DR. NH Emergency Preparedness and Response during Recent Hurricanes. 2006; Available from: https://oig.hhs.gov/oei/reports/oei-06-06-00020.pdf.
10.    Reisner N, Fink S, Yee V. Eight Dead From Sweltering Nursing Home as Florida Struggles After Irma, in The New York Times 2017.
11.    IPPC. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. 2014; 151]. Available from: https://www.ipcc.ch/report/ar5/syr/.
12.    Luber G, Lemery J. Global climate change and human health: from science to practice. 2015, Jossey-Bass, a Wiley brand: San Francisco, CA.