Available online at www.sciencedirect.com
Nurs Outlook xxx (2014) 1e9
www.nursingoutlook.org
Sleep quality and depression of nursing home older adults in wheelchairs after exercises Kuei-Min Chen, RN, PhDa,*, Hsin-Ting Huang, BSa, Yin-Yin Cheng, RN, MSa, Chun-Huw Li, RN, MSb, Ya-Hui Chang, BSa a
b
College of Nursing, Kaohsiung Medical University, Kaohsiung, Taiwan Department of Nursing, Yuhing Junior College of Health Care and Management, Kaohsiung, Taiwan
article info
abstract
Article history: Received 20 May 2014 Revised 12 August 2014 Accepted 20 August 2014
Background: Sleep disturbances and depression are costly and potentially disabling conditions that affect a considerable proportion of older adults. The purpose of this study was to test the effectiveness of 6 months of elastic band exercises on sleep quality and depression of wheelchair-bound older adults in nursing homes. Methods: One hundred twenty-seven older adults from 10 nursing homes participated in this cluster randomized controlled trial, and 114 completed the study. Participants were randomly assigned to two groups: experimental group (five nursing homes, n ¼ 59) and control group (five nursing homes, n ¼ 55). A 40minute wheelchair-bound senior elastic band exercise program was implemented 3 times per week for 6 months. Sleep quality and depression of the participants were examined at baseline, after 3 months, and at the end of the 6month study. Discussion: Participants in the experimental group had longer sleep durations, better habitual sleep efficiencies, and less depression than the control group at 3 months of the study and maintained them throughout the rest of the 6-month study. Conclusions: Nursing home directors could recruit volunteers to learn the program and lead the elderly residents in wheelchairs in practicing the wheelchair-bound senior elastic band exercises regularly in the facilities.
Keywords: Cluster randomized control trial Depression Elastic band Nursing home Older adults Sleep quality Wheelchair exercise
Cite this article: Chen, K.-M., Huang, H.-T., Cheng, Y.-Y., Li, C.-H., & Chang, Y.-H. (2014, -). Sleep quality and depression of nursing home older adults in wheelchairs after exercises. Nursing Outlook, -(-), 1-9. http://dx.doi.org/10.1016/j.outlook.2014.08.010.
Introduction Sleep disturbances, such as altered sleep duration and increased sleep fragmentation, are common among older adults. In Taiwan, 69.3% of institutional older
adults were reported to have poor sleep quality (Lin, Su, & Chang, 2003). The three most often reported sleep disturbances are difficulty initiating sleep, difficulty maintaining sleep, and early morning awakening (Ohayon, 2002). Sleep disturbances in older adults are ascribed to the inactive lifestyles of repetitive daily
Supported by the National Science Council Taiwan (NSC99-2628-B-037-066-MY3). * Corresponding author: Kuei-Min Chen, College of Nursing, Kaohsiung Medical University, 100 Shih-Chuan 1st Rd, Sanming District, Kaohsiung, Taiwan 80708. E-mail address: [email protected] (K.-M. Chen). 0029-6554/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.outlook.2014.08.010
2
Nurs Outlook xxx (2014) 1e9
routines, lack of physical exercise, and poor sleep practices (e.g., excessive time spent in bed and sleeping during the day) (Foley, Ancoli-Israel, Britz, & Walsh, 2004). Wheelchair-bound older adults in nursing homes were even more disposed to these situations. The percentage of older adults using wheelchairs in nursing homes is over 50% of that population (Kaye, Kang, & LaPlante, 2002). Wheelchair use is one of the obstacles to physical activity (Rimmer, 2005), which might lead to further disability and mortality in older adults (Hirvensalo, Rantanen, & Heikkinen, 2000). Inactive lifestyles could further lead to tiredness, daytime drowsiness, fatigue, depression, greater anxiety, irritability, mood disturbances, immunosuppression, inability to participate in activities, and an overall poor quality of life (Buysse et al., 2008; Ouslander et al., 2006; Stenholm et al., 2010). Among these symptoms, depression is a costly and potentially disabling condition affecting a considerable proportion of older adults. Lin, Wang, and Huang (2007) reported a depression prevalence rate of 81.8% among elderly nursing home residents. Gueldner et al. (2001) claimed that older adults who live in nursing homes have a higher prevalence of depression than those living in the community. No other daytime behavior has been associated with better nighttime sleep than exercise (Youngstedt & Kline, 2006). Lin et al. (2003) found that older adults who exercised regularly had a better sleep quality. Physically active individuals had a longer total sleep time, better stage IV of the sleep cycle, and less rapid eye movement sleep duration than sedentary individuals (Youngstedt, O’Connor, & Dishman, 1997). Moreover, older adults who participated in aerobic exercise training were associated with a decreased risk for clinical depression or anxiety (Mather et al., 2002). Physical activity is associated with a decreased state of depression and anxiety and enhanced psychological well-being of individuals (Vuillemin et al., 2005). Compared with individual exercises, group exercises resulted in higher levels of interest and exercise participation, more positive emotional effects (e.g., increased social interaction and lowered levels of depression), and less financial and time burdens (Shin, Lee, & Jang, 2007). Among the various exercises, resistance training stems the decline of muscular strength, which is one of the most important issues related to aging in human beings (Burton & Sumukadas, 2010). Elastic band exercises have been used widely in rehabilitative medicine and health enhancement (Colado et al., 2010; Ribeiro, Teixeira, Brochado, & Oliveira, 2009). It is relatively inexpensive and provides a practical form of training that could be considered in program design for older adults with mobility limitations (Webber & Porter, 2010). Studies indicated that elastic band exercises improved maximal voluntary thigh muscle strength (Binder et al., 2005); increased isometric, isotonic, and isokinetic muscle strength (Ribeiro et al., 2009; Webber & Porter, 2010; Woo, Hong, Lau, & Lynn, 2007); improved muscle strength and flexibility (Latham & Liu, 2010; Nelson
et al., 2007); promoted muscle hypertrophy (Yasuda et al., 2014); improved movement time (Webber & Porter, 2010); improved knee extension and hip extension strength (Dancewicz, Krebs, & McGibbon, 2003); and sit-to-stand performance of older adults (Chen et al., 2009b). Furthermore, elastic band resistance training improved functional ability (Topp, Boardley, Morgan, Fahlman, & McNevin, 2005), increased flexibility and range of joint motion (Sugimoto & Blanpied, 2006; Swank, Funk, Durham, & Roberts, 2003), and enhanced gait and balance of older adults (Topp, Mikesky, Dayhoff, & Holt, 1996). Previous studies have shown the physical effectiveness of elastic band exercises; however, considerably fewer studies have emphasized their psychological benefits. In a pilot testing of a group of older adults in wheelchairs, participants self-reported having more muscle strength in their hands and legs; increased body flexibility and range of joint motion; and, more importantly, slept better and felt more energetic after 4 weeks of elastic band exercises (Chen, Tseng, Chang, Huang, & Li, 2013). Therefore, in addition to the physiological health benefits, elastic band exercises could have potential psychological health benefits. This study aimed to test the effectiveness of 6 months of elastic band exercises on sleep quality and depression of wheelchair-bound older adults in nursing homes.
Methods Design A cluster randomized controlled trial was used. Ten nursing homes that participated voluntarily were randomly assigned to either the experimental or control group using a black box drawing. Data were examined at three points in time: baseline, at 3 months, and at the end of the 6-month study. These time points were specifically chosen to reflect the most frequent intervention periods suggested in the literature (Chen et al., 2009a) and for the purpose of understanding the trends and changes in the 6-month period.
Setting and Participants This study was conducted in 10 nursing homes in Kaohsiung, Southern Taiwan. These nursing homes were privately funded, government accredited, and equipped with 49 beds, which resembled the majority of nursing homes in Taiwan. Based on the nursing home registry in Kaohsiung, the principal investigator called the directors of the nursing homes to determine their willingness to participate in the study; 10 nursing home directors volunteered to participate. Inclusion criteria were as follows: (a) aged 65 and over, (b) using wheelchairs for mobility, (c) living in the facility for at least 3 months, (d) cognitively intact (Short Portable Mental Status Questionnaire score 8) (Pfeiffer, 1975),
Nurs Outlook xxx (2014) 1e9
and (e) heavily or moderately dependent on others to perform activities of daily living (ADLs) (Barthel Index score of 21e90). Exclusion criteria were (a) having severe or acute cardiovascular, musculoskeletal, or pulmonary illnesses or (b) suffering from a spinal cord injury with no rehabilitation potential. Based on the statistical software Sample Power 2.0 (SPSS Inc., Chicago, IL), the required sample size was 56 participants for each group (power ¼ .80, alpha ¼ .05, R2 of covariate in medium level ¼ .13, effect size in medium level ¼ .25). To consider the possibility of dropouts during the study, a convenience sample of 127 participants was recruited in the beginning and randomly assigned to two groups based on the nursing homes where they lived: the experimental group (five nursing homes, n ¼ 64) and the control group (five nursing homes, n ¼ 63). A total of 115 participants completed the study (attrition rate ¼ 9.45%); nine participants dropped out by the end of the 3-month interval (experimental group: n ¼ 3, control group: n ¼ 6), and three more participants dropped out by the end of
3
the 6-month interval (experimental group: n ¼ 1, control group: n ¼ 2). Reasons for dropping out included deceased (experimental group: n ¼ 3, control group: n ¼ 3) and discharged from nursing homes (experimental group: n ¼ 1, control group: n ¼ 5). To ensure the consistency of the intervention received by the participants, one experimental group participant with an attendance rate less than 50% was further excluded, which yielded a total of 114 participants in the final data analysis (experimental group: n ¼ 59, control group: n ¼ 55) (Figure 1). The Mann-Whitney U and the Pearson chisquare tests indicated no significant differences in the demographic profiles of the participants remaining in the study and those who dropped out (all p > .05).
Ethical Considerations After approval by the institutional review board of the university hospital and the nursing home administrators, potential participants were contacted by the staff
Figure 1 e Flow of the participants.
4
Nurs Outlook xxx (2014) 1e9
of the nursing homes. From those older adults who were willing to participate, the staff of the nursing homes made referrals to the research team. The research assistants met with the participants individually, explained the study, and then had them sign written informed consent forms.
Intervention The wheelchair-bound senior elastic band (WSEB) exercise program developed by Chen et al. (2013) was applied as the intervention for the experimental group participants in addition to their regular daily activities. The program includes basic and advanced levels, and both levels have three phases: warm-up, aerobic motion, and harmonic stretching. The basic level has four elastic band exercises in each phase. Two relatively challenging exercises were added in each phase of the basic level to form the advanced WSEB program (six elastic band exercises in each phase) (Chen et al., 2013). The basic level was taught and practiced for the first 3 months followed by the advanced level for another 3 months. The WSEB group exercise intervention was performed three times per week 40 minutes per session for 6 months. The program was led by two trained and certified instructors who were staff or volunteers from the facilities. To ensure intervention consistency across the intervention groups and interrater reliability among the instructors, a prerecorded CD made by the principal investigator verbally guided the intervention process while the instructors demonstrated and led the group exercises. The instructors were asked to follow the CD to lead the group exercise and demonstrate the exercises as trained. Furthermore, the instructors were asked to record any signs or symptoms of discomfort that occurred during exercises; no special concerns occurred. The attendance rate of the participants was 94.51%. The control group participants continued with their regular daily activities.
Data Collection The study was conducted from March through September 2012. Data were collected by three research assistants who were trained for consistency in delivering the questions in the questionnaires. The face-toface individually structured interviews took place in a small lounge or private and quiet room in the nursing home. Sleep quality was measured by the Pittsburgh Sleep Quality Index (PSQI), which is a widely used, 18item, self-reported questionnaire that measures sleep quality and sleep disturbances of an individual for the past month (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Seven component scores (i.e., subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, the use of sleeping medications, and daytime dysfunction) are generated and are summed to yield one global score. The higher the score, the worse the sleep quality. A PSQI global score greater than 5 is associated with clinically significant sleep disturbances, such as
insomnia or major mood disorders (90% sensitivity and 87% specificity) (Buysse et al., 1989). The Chinese PSQI, which is well documented for reliability and validity (Chen et al., 2010; Lu, Lin, Chen, Tsang, & Su, 2013), was used in this study. A Cronbach alpha of .70 was obtained from this sample. The depression of the participants was measured by the 18-item Taiwanese Depression Questionnaire (TDQ), which is a scale used to measure the emotional feelings of the respondents for the past week (Lee, Yang, Lai, Chiu, & Chau, 2000). The participants were asked to rate their emotions on a Likert-type scale, ranging from 0 to 3 (0 ¼ the described symptoms have happened for .05).
Baseline Comparisons Between the Two Groups Results of the independent t-tests indicated that participants in the experimental group had worse overall sleep quality (t ¼ 2.06, p ¼ .042), sleep latency (t ¼ 3.18, p ¼ .002), and habitual sleep efficiencies (t ¼ 2.28, p ¼ .025) than the control group. The experimental group participants also had more sleep disturbances than the control group participants (t ¼ 2.31, p ¼ .023). No significant differences were found in the remaining variables (all p >.05) at baseline.
Interaction Effects Between Time Points and Groups Results of a mix-design, two-way ANOVA indicated that there were significant interaction effects among three time points and two groups in the following variables: overall sleep quality (F ¼ 5.84, p ¼ .007), sleep duration (F ¼ 7.79, p ¼ .001), habitual sleep efficiency (F ¼ 8.57, p ¼ .001), and depression state (F ¼ 13.21, p < .001).
Changes among Three Time Points in Each Group All of the variables, except overall sleep quality (F ¼ 1.85, p ¼ .171), had significant changes in the experimental group (Table 1). After the WSEB exercises, participants had longer sleep duration and habitual sleep efficiencies
(F ¼ 7.12, p ¼ .002 and F ¼ 4.27, p ¼ .025; respectively) and less depression (F ¼ 4.38, p ¼ .022) than their baseline (Table 1). Similarly, all of the variables, except sleep duration (F ¼ 1.80, p ¼ .175), had significant changes in the control group (Table 2). However, without WSEB exercises, the overall sleep quality, habitual sleep efficiencies, and depression of the control group participants worsened significantly (F ¼ 5.02, p ¼ .015; F ¼ 4.63, p ¼ .019; and F ¼ 11.27, p < .001, respectively) (Table 2).
Group Differences in Each Time Point Participants in the experimental group had longer sleep durations and less depression than the control group at the 3-month point of the study (F ¼ 11.09, p < .001 and F ¼ 28.10, p < .001, respectively) (Table 3) and maintained those differences throughout the remainder of the 6month study (F ¼ 10.07, p ¼ .002 and F ¼ 18.85, p < .001, respectively) (Table 4). The experimental group also had better habitual sleep efficiencies than the control group as soon as the 3-month point of the study (F ¼ 7.46, p ¼ .007) (Table 3); however, these differences were not maintained throughout the remainder of the 6-month study (F ¼ 3.66, p ¼ .058) (Table 4). Finally, although the overall sleep quality showed no significant difference between the two groups at the 3-month point of the study (F ¼ 3.93, p ¼ .050) (Table 3), by the end of 6 months of the study, participants in the experimental group had a significantly better overall sleep quality than the control group (F ¼ 4.83, p ¼ .030) (Table 4).
Discussion Results indicated that both experimental and control group participants had PSQI total scores greater than 5 at baseline (experimental ¼ 8.14, control ¼ 6.47), indicating that participants had clinically significant sleep disturbances, such as insomnia or major mood disorders (Buysse et al., 1989). These results were similar to Lin et al. (2003) in that more than half of the
Table 1 e Changes among Baseline, Post-test 1 (3 Months), and Post-test 2 (6 Months) on the Variables with Significant Interaction Effects in the Experimental Group (n [ 59) Variables Baseline Post-test 1 Post-test 2 F ( p) Post hoc* M
SD
M
SD
M
SD
Sleep quality total score Sleep duration
8.14 1.02
3.84 1.17
7.51 0.68
4.20 1.06
7.68 0.68
4.24 1.06
1.85 (.171) 7.12 (.002)jj
Habitual sleep efficiency Depression statez
1.37 7.63
1.31 8.40
0.98 4.92
1.20 5.73
1.12 5.44
1.21 7.80
4.27 (.025)x 4.38 (.022)x
y
d Base > post Base > post Base > post Base > post
1 2 1 1
d ¼ post hoc analysis was not performed because of nonsignificant F values; SD ¼ standard deviation. * Bonferroni post hoc test (adjusting a levels of multiple tests). y Sleep quality was measured by the Pittsburgh Sleep Quality Index (PSQI). A higher score indicated a worse situation. z Depression state was measured by the Taiwanese Depression Questionnaire (TDQ). A higher score indicated a worse situation. x p < .05. jj p < .01.
6
Nurs Outlook xxx (2014) 1e9
Table 2 e Changes among Baseline, Post-test 1 (3 Months), and Post-test 2 (6 Months) on the Variables with Significant Interaction Effects in the Control Group (n [ 55) Variables Baseline Post-test 1 Post-test 2 F ( p) Post hoc* y
Sleep quality total score Sleep duration Habitual sleep efficiency Depression statez
M
SD
M
SD
M
SD
6.47 0.91 0.84 7.95
4.71 1.16 1.20 9.07
7.11 1.07 1.20 10.69
4.86 1.22 1.31 9.83
7.64 1.11 1.22 11.53
4.91 1.18 1.30 9.85
5.02 (.015)x 1.80 (.175) 4.63 (.019)x 11.27 (
Nurs Outlook xxx (2014) 1e9
www.nursingoutlook.org
Sleep quality and depression of nursing home older adults in wheelchairs after exercises Kuei-Min Chen, RN, PhDa,*, Hsin-Ting Huang, BSa, Yin-Yin Cheng, RN, MSa, Chun-Huw Li, RN, MSb, Ya-Hui Chang, BSa a
b
College of Nursing, Kaohsiung Medical University, Kaohsiung, Taiwan Department of Nursing, Yuhing Junior College of Health Care and Management, Kaohsiung, Taiwan
article info
abstract
Article history: Received 20 May 2014 Revised 12 August 2014 Accepted 20 August 2014
Background: Sleep disturbances and depression are costly and potentially disabling conditions that affect a considerable proportion of older adults. The purpose of this study was to test the effectiveness of 6 months of elastic band exercises on sleep quality and depression of wheelchair-bound older adults in nursing homes. Methods: One hundred twenty-seven older adults from 10 nursing homes participated in this cluster randomized controlled trial, and 114 completed the study. Participants were randomly assigned to two groups: experimental group (five nursing homes, n ¼ 59) and control group (five nursing homes, n ¼ 55). A 40minute wheelchair-bound senior elastic band exercise program was implemented 3 times per week for 6 months. Sleep quality and depression of the participants were examined at baseline, after 3 months, and at the end of the 6month study. Discussion: Participants in the experimental group had longer sleep durations, better habitual sleep efficiencies, and less depression than the control group at 3 months of the study and maintained them throughout the rest of the 6-month study. Conclusions: Nursing home directors could recruit volunteers to learn the program and lead the elderly residents in wheelchairs in practicing the wheelchair-bound senior elastic band exercises regularly in the facilities.
Keywords: Cluster randomized control trial Depression Elastic band Nursing home Older adults Sleep quality Wheelchair exercise
Cite this article: Chen, K.-M., Huang, H.-T., Cheng, Y.-Y., Li, C.-H., & Chang, Y.-H. (2014, -). Sleep quality and depression of nursing home older adults in wheelchairs after exercises. Nursing Outlook, -(-), 1-9. http://dx.doi.org/10.1016/j.outlook.2014.08.010.
Introduction Sleep disturbances, such as altered sleep duration and increased sleep fragmentation, are common among older adults. In Taiwan, 69.3% of institutional older
adults were reported to have poor sleep quality (Lin, Su, & Chang, 2003). The three most often reported sleep disturbances are difficulty initiating sleep, difficulty maintaining sleep, and early morning awakening (Ohayon, 2002). Sleep disturbances in older adults are ascribed to the inactive lifestyles of repetitive daily
Supported by the National Science Council Taiwan (NSC99-2628-B-037-066-MY3). * Corresponding author: Kuei-Min Chen, College of Nursing, Kaohsiung Medical University, 100 Shih-Chuan 1st Rd, Sanming District, Kaohsiung, Taiwan 80708. E-mail address: [email protected] (K.-M. Chen). 0029-6554/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.outlook.2014.08.010
2
Nurs Outlook xxx (2014) 1e9
routines, lack of physical exercise, and poor sleep practices (e.g., excessive time spent in bed and sleeping during the day) (Foley, Ancoli-Israel, Britz, & Walsh, 2004). Wheelchair-bound older adults in nursing homes were even more disposed to these situations. The percentage of older adults using wheelchairs in nursing homes is over 50% of that population (Kaye, Kang, & LaPlante, 2002). Wheelchair use is one of the obstacles to physical activity (Rimmer, 2005), which might lead to further disability and mortality in older adults (Hirvensalo, Rantanen, & Heikkinen, 2000). Inactive lifestyles could further lead to tiredness, daytime drowsiness, fatigue, depression, greater anxiety, irritability, mood disturbances, immunosuppression, inability to participate in activities, and an overall poor quality of life (Buysse et al., 2008; Ouslander et al., 2006; Stenholm et al., 2010). Among these symptoms, depression is a costly and potentially disabling condition affecting a considerable proportion of older adults. Lin, Wang, and Huang (2007) reported a depression prevalence rate of 81.8% among elderly nursing home residents. Gueldner et al. (2001) claimed that older adults who live in nursing homes have a higher prevalence of depression than those living in the community. No other daytime behavior has been associated with better nighttime sleep than exercise (Youngstedt & Kline, 2006). Lin et al. (2003) found that older adults who exercised regularly had a better sleep quality. Physically active individuals had a longer total sleep time, better stage IV of the sleep cycle, and less rapid eye movement sleep duration than sedentary individuals (Youngstedt, O’Connor, & Dishman, 1997). Moreover, older adults who participated in aerobic exercise training were associated with a decreased risk for clinical depression or anxiety (Mather et al., 2002). Physical activity is associated with a decreased state of depression and anxiety and enhanced psychological well-being of individuals (Vuillemin et al., 2005). Compared with individual exercises, group exercises resulted in higher levels of interest and exercise participation, more positive emotional effects (e.g., increased social interaction and lowered levels of depression), and less financial and time burdens (Shin, Lee, & Jang, 2007). Among the various exercises, resistance training stems the decline of muscular strength, which is one of the most important issues related to aging in human beings (Burton & Sumukadas, 2010). Elastic band exercises have been used widely in rehabilitative medicine and health enhancement (Colado et al., 2010; Ribeiro, Teixeira, Brochado, & Oliveira, 2009). It is relatively inexpensive and provides a practical form of training that could be considered in program design for older adults with mobility limitations (Webber & Porter, 2010). Studies indicated that elastic band exercises improved maximal voluntary thigh muscle strength (Binder et al., 2005); increased isometric, isotonic, and isokinetic muscle strength (Ribeiro et al., 2009; Webber & Porter, 2010; Woo, Hong, Lau, & Lynn, 2007); improved muscle strength and flexibility (Latham & Liu, 2010; Nelson
et al., 2007); promoted muscle hypertrophy (Yasuda et al., 2014); improved movement time (Webber & Porter, 2010); improved knee extension and hip extension strength (Dancewicz, Krebs, & McGibbon, 2003); and sit-to-stand performance of older adults (Chen et al., 2009b). Furthermore, elastic band resistance training improved functional ability (Topp, Boardley, Morgan, Fahlman, & McNevin, 2005), increased flexibility and range of joint motion (Sugimoto & Blanpied, 2006; Swank, Funk, Durham, & Roberts, 2003), and enhanced gait and balance of older adults (Topp, Mikesky, Dayhoff, & Holt, 1996). Previous studies have shown the physical effectiveness of elastic band exercises; however, considerably fewer studies have emphasized their psychological benefits. In a pilot testing of a group of older adults in wheelchairs, participants self-reported having more muscle strength in their hands and legs; increased body flexibility and range of joint motion; and, more importantly, slept better and felt more energetic after 4 weeks of elastic band exercises (Chen, Tseng, Chang, Huang, & Li, 2013). Therefore, in addition to the physiological health benefits, elastic band exercises could have potential psychological health benefits. This study aimed to test the effectiveness of 6 months of elastic band exercises on sleep quality and depression of wheelchair-bound older adults in nursing homes.
Methods Design A cluster randomized controlled trial was used. Ten nursing homes that participated voluntarily were randomly assigned to either the experimental or control group using a black box drawing. Data were examined at three points in time: baseline, at 3 months, and at the end of the 6-month study. These time points were specifically chosen to reflect the most frequent intervention periods suggested in the literature (Chen et al., 2009a) and for the purpose of understanding the trends and changes in the 6-month period.
Setting and Participants This study was conducted in 10 nursing homes in Kaohsiung, Southern Taiwan. These nursing homes were privately funded, government accredited, and equipped with 49 beds, which resembled the majority of nursing homes in Taiwan. Based on the nursing home registry in Kaohsiung, the principal investigator called the directors of the nursing homes to determine their willingness to participate in the study; 10 nursing home directors volunteered to participate. Inclusion criteria were as follows: (a) aged 65 and over, (b) using wheelchairs for mobility, (c) living in the facility for at least 3 months, (d) cognitively intact (Short Portable Mental Status Questionnaire score 8) (Pfeiffer, 1975),
Nurs Outlook xxx (2014) 1e9
and (e) heavily or moderately dependent on others to perform activities of daily living (ADLs) (Barthel Index score of 21e90). Exclusion criteria were (a) having severe or acute cardiovascular, musculoskeletal, or pulmonary illnesses or (b) suffering from a spinal cord injury with no rehabilitation potential. Based on the statistical software Sample Power 2.0 (SPSS Inc., Chicago, IL), the required sample size was 56 participants for each group (power ¼ .80, alpha ¼ .05, R2 of covariate in medium level ¼ .13, effect size in medium level ¼ .25). To consider the possibility of dropouts during the study, a convenience sample of 127 participants was recruited in the beginning and randomly assigned to two groups based on the nursing homes where they lived: the experimental group (five nursing homes, n ¼ 64) and the control group (five nursing homes, n ¼ 63). A total of 115 participants completed the study (attrition rate ¼ 9.45%); nine participants dropped out by the end of the 3-month interval (experimental group: n ¼ 3, control group: n ¼ 6), and three more participants dropped out by the end of
3
the 6-month interval (experimental group: n ¼ 1, control group: n ¼ 2). Reasons for dropping out included deceased (experimental group: n ¼ 3, control group: n ¼ 3) and discharged from nursing homes (experimental group: n ¼ 1, control group: n ¼ 5). To ensure the consistency of the intervention received by the participants, one experimental group participant with an attendance rate less than 50% was further excluded, which yielded a total of 114 participants in the final data analysis (experimental group: n ¼ 59, control group: n ¼ 55) (Figure 1). The Mann-Whitney U and the Pearson chisquare tests indicated no significant differences in the demographic profiles of the participants remaining in the study and those who dropped out (all p > .05).
Ethical Considerations After approval by the institutional review board of the university hospital and the nursing home administrators, potential participants were contacted by the staff
Figure 1 e Flow of the participants.
4
Nurs Outlook xxx (2014) 1e9
of the nursing homes. From those older adults who were willing to participate, the staff of the nursing homes made referrals to the research team. The research assistants met with the participants individually, explained the study, and then had them sign written informed consent forms.
Intervention The wheelchair-bound senior elastic band (WSEB) exercise program developed by Chen et al. (2013) was applied as the intervention for the experimental group participants in addition to their regular daily activities. The program includes basic and advanced levels, and both levels have three phases: warm-up, aerobic motion, and harmonic stretching. The basic level has four elastic band exercises in each phase. Two relatively challenging exercises were added in each phase of the basic level to form the advanced WSEB program (six elastic band exercises in each phase) (Chen et al., 2013). The basic level was taught and practiced for the first 3 months followed by the advanced level for another 3 months. The WSEB group exercise intervention was performed three times per week 40 minutes per session for 6 months. The program was led by two trained and certified instructors who were staff or volunteers from the facilities. To ensure intervention consistency across the intervention groups and interrater reliability among the instructors, a prerecorded CD made by the principal investigator verbally guided the intervention process while the instructors demonstrated and led the group exercises. The instructors were asked to follow the CD to lead the group exercise and demonstrate the exercises as trained. Furthermore, the instructors were asked to record any signs or symptoms of discomfort that occurred during exercises; no special concerns occurred. The attendance rate of the participants was 94.51%. The control group participants continued with their regular daily activities.
Data Collection The study was conducted from March through September 2012. Data were collected by three research assistants who were trained for consistency in delivering the questions in the questionnaires. The face-toface individually structured interviews took place in a small lounge or private and quiet room in the nursing home. Sleep quality was measured by the Pittsburgh Sleep Quality Index (PSQI), which is a widely used, 18item, self-reported questionnaire that measures sleep quality and sleep disturbances of an individual for the past month (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Seven component scores (i.e., subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, the use of sleeping medications, and daytime dysfunction) are generated and are summed to yield one global score. The higher the score, the worse the sleep quality. A PSQI global score greater than 5 is associated with clinically significant sleep disturbances, such as
insomnia or major mood disorders (90% sensitivity and 87% specificity) (Buysse et al., 1989). The Chinese PSQI, which is well documented for reliability and validity (Chen et al., 2010; Lu, Lin, Chen, Tsang, & Su, 2013), was used in this study. A Cronbach alpha of .70 was obtained from this sample. The depression of the participants was measured by the 18-item Taiwanese Depression Questionnaire (TDQ), which is a scale used to measure the emotional feelings of the respondents for the past week (Lee, Yang, Lai, Chiu, & Chau, 2000). The participants were asked to rate their emotions on a Likert-type scale, ranging from 0 to 3 (0 ¼ the described symptoms have happened for .05).
Baseline Comparisons Between the Two Groups Results of the independent t-tests indicated that participants in the experimental group had worse overall sleep quality (t ¼ 2.06, p ¼ .042), sleep latency (t ¼ 3.18, p ¼ .002), and habitual sleep efficiencies (t ¼ 2.28, p ¼ .025) than the control group. The experimental group participants also had more sleep disturbances than the control group participants (t ¼ 2.31, p ¼ .023). No significant differences were found in the remaining variables (all p >.05) at baseline.
Interaction Effects Between Time Points and Groups Results of a mix-design, two-way ANOVA indicated that there were significant interaction effects among three time points and two groups in the following variables: overall sleep quality (F ¼ 5.84, p ¼ .007), sleep duration (F ¼ 7.79, p ¼ .001), habitual sleep efficiency (F ¼ 8.57, p ¼ .001), and depression state (F ¼ 13.21, p < .001).
Changes among Three Time Points in Each Group All of the variables, except overall sleep quality (F ¼ 1.85, p ¼ .171), had significant changes in the experimental group (Table 1). After the WSEB exercises, participants had longer sleep duration and habitual sleep efficiencies
(F ¼ 7.12, p ¼ .002 and F ¼ 4.27, p ¼ .025; respectively) and less depression (F ¼ 4.38, p ¼ .022) than their baseline (Table 1). Similarly, all of the variables, except sleep duration (F ¼ 1.80, p ¼ .175), had significant changes in the control group (Table 2). However, without WSEB exercises, the overall sleep quality, habitual sleep efficiencies, and depression of the control group participants worsened significantly (F ¼ 5.02, p ¼ .015; F ¼ 4.63, p ¼ .019; and F ¼ 11.27, p < .001, respectively) (Table 2).
Group Differences in Each Time Point Participants in the experimental group had longer sleep durations and less depression than the control group at the 3-month point of the study (F ¼ 11.09, p < .001 and F ¼ 28.10, p < .001, respectively) (Table 3) and maintained those differences throughout the remainder of the 6month study (F ¼ 10.07, p ¼ .002 and F ¼ 18.85, p < .001, respectively) (Table 4). The experimental group also had better habitual sleep efficiencies than the control group as soon as the 3-month point of the study (F ¼ 7.46, p ¼ .007) (Table 3); however, these differences were not maintained throughout the remainder of the 6-month study (F ¼ 3.66, p ¼ .058) (Table 4). Finally, although the overall sleep quality showed no significant difference between the two groups at the 3-month point of the study (F ¼ 3.93, p ¼ .050) (Table 3), by the end of 6 months of the study, participants in the experimental group had a significantly better overall sleep quality than the control group (F ¼ 4.83, p ¼ .030) (Table 4).
Discussion Results indicated that both experimental and control group participants had PSQI total scores greater than 5 at baseline (experimental ¼ 8.14, control ¼ 6.47), indicating that participants had clinically significant sleep disturbances, such as insomnia or major mood disorders (Buysse et al., 1989). These results were similar to Lin et al. (2003) in that more than half of the
Table 1 e Changes among Baseline, Post-test 1 (3 Months), and Post-test 2 (6 Months) on the Variables with Significant Interaction Effects in the Experimental Group (n [ 59) Variables Baseline Post-test 1 Post-test 2 F ( p) Post hoc* M
SD
M
SD
M
SD
Sleep quality total score Sleep duration
8.14 1.02
3.84 1.17
7.51 0.68
4.20 1.06
7.68 0.68
4.24 1.06
1.85 (.171) 7.12 (.002)jj
Habitual sleep efficiency Depression statez
1.37 7.63
1.31 8.40
0.98 4.92
1.20 5.73
1.12 5.44
1.21 7.80
4.27 (.025)x 4.38 (.022)x
y
d Base > post Base > post Base > post Base > post
1 2 1 1
d ¼ post hoc analysis was not performed because of nonsignificant F values; SD ¼ standard deviation. * Bonferroni post hoc test (adjusting a levels of multiple tests). y Sleep quality was measured by the Pittsburgh Sleep Quality Index (PSQI). A higher score indicated a worse situation. z Depression state was measured by the Taiwanese Depression Questionnaire (TDQ). A higher score indicated a worse situation. x p < .05. jj p < .01.
6
Nurs Outlook xxx (2014) 1e9
Table 2 e Changes among Baseline, Post-test 1 (3 Months), and Post-test 2 (6 Months) on the Variables with Significant Interaction Effects in the Control Group (n [ 55) Variables Baseline Post-test 1 Post-test 2 F ( p) Post hoc* y
Sleep quality total score Sleep duration Habitual sleep efficiency Depression statez
M
SD
M
SD
M
SD
6.47 0.91 0.84 7.95
4.71 1.16 1.20 9.07
7.11 1.07 1.20 10.69
4.86 1.22 1.31 9.83
7.64 1.11 1.22 11.53
4.91 1.18 1.30 9.85
5.02 (.015)x 1.80 (.175) 4.63 (.019)x 11.27 (
![[Quality of sleep in nursing home non-demented residents].](/assets/img/hold.png)
