4.1▓Sample characteristic
The average time of accomplished the questionnaire was 22.9 minutes (SD 26.8, range 3-240 minutes). Fifty one (33.8%) participants accomplished the questionnaires by themselves without any help, while sixteen participants required some assistance from interviewers (10.6%). About half of all participants (55.6%) which were 84 participants were interviewed. Two hundred and eighteen participants were screened and one hundred and fifty one participants had completed the questionnaires including Short Portable Mental Status Questionnaire (SPMSQ), WHOQOL-BREF Taiwan version, and social demographic characteristics, Barthel Index of Activities of Daily Living (ADL) Activity, Instrumental Activities of Daily Living Scale (IADL), and Taiwan Depression Questionnaire (TDQ). The response rate was 71.0%. Table 4-1 summarizes the social demographic characteristics of the participants.
The mean ages for the home-dwelling, senior apartment, and institute-based senior home were 74.0 (SD 5.4), 81.5 (SD 6.2), and 77.5 (SD 8.5) years, respectively with significant different p<0.01 (F=20.7, p=.000). It showed that the age of the seniors in apartment was the eldest among three types of living arrangement. There were more female in senior apartments and home-dwelling 72.5% and 67.4% compare to the institution based senior housing which was 48%. Mostly, the elderly people were educated regardless the different living arrangements. The senior apartments’ comprised of the elderly people educated with university and above was 18.8% and who were literate or at least attended primary schools were 91.2%. For the elderly people who stayed at home were 93.5% literate and graduated from the university or above were 23.9%. The education level of the elderly people staying in the institution-based home were 92.0% at least attended primary school and only graduated from the university and above were 10.0%.
The participants in senior apartment were 87.5% married but 70.0% of them had lost
- 40 -
their partners and became widowed. For the home-dwelling participants were mostly 95.6%
married and 30.4% were widowed. The participants in institute-based home were 72%
married with 44% were widowed.
There were only 1.3% of the participants in senior apartment, 8.8% at home-dwelling, and 8.0% of institution-based participants were still working. Others were housewife, no working or retired. With the individual income, family income more than 20,000 NTD, there were 57.7%, 68.1% of participants in senior apartment, 43.4%, 68.2% of participants at home-dwelling, and 20%, 44% of participant staying at institution-based home. Mostly the residents in institution-based home had less income trends.
About half of the residents were Buddhist in senior apartment (51.3%) and home-dwelling (53.3%). The non-religious participants were 8.8% staying in senior apartment, 2.2% staying at home and 16.0% in institution-based home. Generally, there were 93.8%
(senior apartment), 63.0% (home-dwelling), and 84.0% (institution-based home) participants had at least one chronic disease. The 18.8% (senior apartment), 4.3% (home-dwelling), 8.0%
(institution-based home) of the participants’ health status were poor or very poor. The rest were either good, very good or neither good nor poor. Their life conditions were 50%, 63.1%, 56.0% happy and very happy in senior apartment, home-dwelling, institution-based home respectively. All the participants were non-smoker except 4.3% of the participants were smokers who stay at home. Besides, the 89.9 % and 76.1 % of participants staying in senior apartment and at home were non-alcoholic and all the participants 100% in institution were non-alcoholic.
The predictors including age, marriage, disease, health status, individual income, family income, alcohol consumption were significantly different among the three different living arrangements (refer table 4-1). The p-value of age (p<.0001), marriage (p<.0001), disease (p<.0001), health status (p<.0001), family income (p=.006) and alcohol consumption were less than 0.001. The alcohol consumption (p=.039), is significantly different among the three types living with p less than 0.05.
- 41 -
Table 4-1▓Demographic characteristics of participants
Senior Apartment (N=80)
N (%) or Mean (SD) Home-dwelling (N=46)
- 42 -
Table 4-1▓Demographic characteristics of participants (continued)
Senior Apartment
N (%) or Mean (SD) Home-dwelling
Technician & assistant Technical labor Affairs staff
Serving & salesman Farmers
- 43 -
Table 4-1▓Demographic characteristics of participants (continued)
Senior Apartment
N (%) or Mean (SD) Home-dwelling
- 44 -
Table 4-1▓Demographic characteristics of participants (continued)
Senior Apartment
N (%) Home-dwelling
- 45 -
Table 4-1▓Demographic characteristics of participants (continued)
Senior Apartment
N (%) Home-dwelling
Average cigarette /day before quitting 6-10/ day
- 46 -
Table 4-1▓Demographic characteristics of participants (continued)
Senior Apartment
N (%) or Mean (SD) Home-dwelling
- 47 -
4.2▓Objective health conditions
The level of dependence or functional status and mental health (depression state) is summarized in Table 4-2. The mean score of the TDQ was highest in senior apartment (8.4, SD 7.0), and second in institution-based home (6.2, SD 7.6). The participants who stay at home had significantly less depression symptoms compared to the other two types living arrangements with p-value of 0.021. However, the participants in the senior apartment and home-dwelling had the mean scores of level of independency with the functional ability (ADL) 99.2 (SD 3.2) and 98.9 (SD 4.2). The mean score for the ADL in institute was 98.0 (SD 4.3), all these high scores indicated only they had mild dependency. The mean IADL score for the participants in the institution-based home was the lowest (16.8, SD 3.4), that means they need more assistance compare to the one who stayed in senior apartment (20.5, SD 2.9) or were home-dwelling (21.6, SD 3.3). The different need of assistant (from IADL score) was significant among the three types of living places arrangements with p-values less than 0.001. However, the level of dependency had no significant different with p=.376.
Overall, the objective health conditions indicated that the participants had mild dependency and mild depression.
- 48 -
Table 4-2▓Scores of objective health: Mental health and level of dependence measures
Senior Apartment
N=80
Home-Dwelling N=46
Institute-Based N=25
Statistics Mean (SD) Mean (SD) Mean (SD) F p Mental Health
*TDQ 8.4 (7.0) 5.1 (4.9) 6.2 (7.6) 4.0 .021
Functional Status
**IADL
***ADL
20.5 (2.9) 99.2 (3.1)
21.6 (3.3) 98.9 (4.2)
16.8 (3.4) 98.0 (4.3)
19.6 .000 1.0 .376
*TDQ: Taiwanese Depression Questionnaire
**IADL: Instrumental Activities of Daily Living
***ADL: Barthel Index of Activities of Daily Living
- 49 -
4.3▓Quality of Life assessed by the WHOQOL-BREF
Descriptive results for the four domains of the quality of life are shown in table 4-3.
Among the 28 items, enough money for spending (Q12), the sexual activity (Q21), and support from friends (Q22) did not have 100% response rate, especially Q21. The interviewers had tried the second approach. Most of the questions had been answered for the second attempt, only the sexual activity questions were not accomplished. There were two residents in senior apartment refused to answer and one resident in institution based home moved so they did not answer Q21. The reasons the elderly people refused to answer this questions were due to the privacy or they were single or widowed. They felt uneasy and sensitive to answer this question.
For the total of one hundred and fifty one elderly people, the lowest score was Q14-opportunity for leisure activities (mean score 3.1, SD 1.0) and the second lowest was Q5-enjoy life (mean score 3.2, SD 0.2). Q13-opportunities for acquiring new daily information and skills (mean score, 3.2, SD 0.8) was the third lowest scores. The top three highest scores were Q3-pain and discomfort (mean score 4.1, SD 0.9); Q23-home environment (mean score 3.9, SD 0.6), Q24-conveniences for medical services (mean score 3.9, SD 0.5).
Among the three types of senior housing, the mean scores of WHOQOL-BREF Taiwan version in senior apartment, home-dwelling, institution based home were 13.7 (SD 2.0), 15.0, (SD 2.5), 14.5 (SD 2.1), (F=5.56, p=.005) respectively for physical domain;13.5 (SD 2.1), 14.6 (SD 2.5), 13.8 (SD 2.7), (F=3.36, p=.037) for psychological domain, 14.4 (SD 1.8), 15.4 (SD 2.5), 14.6 (SD 1.9), (F=4.27, p=.016) for social relationship, and 13.9 (SD 1.3), 14.9 (SD 2.2), 14.2 (SD 2.2), (F=4.55, p=.012) for environment domain (refer table 4-3, graph 4-1).
The mean scores of four domains were significantly different among the three types living arrangement with p value less than 0.05. The senior apartment, home-dwelling, institute-based home had the highest domain score in social relationship and lowest in psychological domain among four domains. Overall, the mean scores of quality of life before the specific chronic
- 50 -
conditions scores were 77. 6 (SD 12.7), 81.6 (SD 14.3), 82.6 (SD 14.1) and the mean scores of the current quality of life were 69.8 (SD 13.1), 78.8 (SD 14.2), 75.3 (SD 16.5) respectively in senior apartment, home-dwelling, and institute-based home. The mean score of the quality of life was generally lower compare to the quality of life before any specific chronic conditions. The participants had significant better quality of life before they had any chronic conditions. With the simple independent t-test, one hundred and fifty one participants had significant better quality of life before suffering from any chronic condition (73.4, SD 14.6 compared to 79.6, SD 13.5, p<.0001).
There was model significance for the ANOVA, 1) F(2,148) = 5.56, p< .05 for physical domain of quality of life; 2) F(2,148)=3.36, p<.05 for psychological domain; 3) F(2,148)=4.27, p<.05; 4) F(2,148)= 4.55, p<.05, indicating at least one significant difference among the means (refer table 4-3). Scheffé posthoc comparisons showed that physical domain was significantly higher in the home dwelling (M = 15.0) than the senior apartment (M =13.7), p =.005 (two-tailed). In turn, physical domain of quality of life was lower in the senior apartment (M =13.7) than either the institution-based, p= .321, or home-dwelling, p
< .01 (two-tailed). There were no significant differences between either the institution-based and home-dwelling, p=.576 or institution-based and senior apartment, p>.05. Psychological domain, social relationship domain, and environment domain of quality of life was significantly higher in the home-dwelling (Mpsychological= 14.6, Msocial relationship= 15.4, Menvironment= 14.9) than senior apartment ((Mpsychological= 13.5, Msocial relationship= 14.4, Menvironment= 13.9) with p= .038, p= .017, p= .012 respectively. However, the psychological domain, social relationship domain, and environment domain of quality of life was no significant difference between both home-dwelling and institution-based p>.05 (ppsychological= .374, psocial relationship= .218, penvironment= .333), or institution-based and senior apartment p>.05 (ppsychological= .854, psocial relationship= .912, penvironment= .715).
- 51 -
Table 4-3▓Mean scores (SD) of Quality of Life by domains
Quality of Life
Senior Apartment (N=80)
Home-Dwelling (N=46)
Institution-Based (N=25)
Statistics F p
Domain 1
Physical 13.7 (2.0) 15.0 (2.5) 14.5 (2.1)
5.56 .005 Domain 2
Psychological 13.5 (2.1) 14.6 (2.5) 13.8 (2.7)
3.36 .038 Domain 3
Social relationship
14.4 (1.8) 15.4 (2.5) 14.6 (1.9)
4.27 .017 Domain 4
Environment 13.9 (1.3) 14.9 (2.2) 14.2 (2.2)
4.55 .012
*QOL scores before chronic
conditions 77. 6 (12.7) 81.6 (14.3) 82.6 (14.1)
6.30 .002
*QOL scores 69.8 (13.1) 78.8 (14.2) 75.3 (16.5) 2.08 .128
*QOL: Quality of Life
- 52 -
Figure 4-1 95%CI means of physical domain of QOL by different living arrangements