3.1 Demographic characteristics of elderly adults living alone
As shown in Table 1, a total of 3,726 elderly adults living alone were distributed in this study, with only 236 (6.3%) having willingness of institutional care. Of all respondents, the majority of elderly were female (50.4%), at the ages of 70–79 (48.7%), having education level of primary or below (62.8%), rural (57.9%), monthly income below 2000 RMB (69.4%), have medical insurance (93.2%), have endowment insurance (94.7%), have two children (46.3%), annual medical examinations (83.8%).
The results of the chi-square test showed that education (χ2=0.024), source of life support (χ2<0.001), knowledge of the combination with medical care (χ2 < 0.001), medical insurance (χ2=0.032), the number of children (χ2 < 0.001), satisfaction with the current medical service (χ2 < 0.001), loneliness (χ2 < 0.001), and frequency of medical examinations (χ2<0.001) were significantly different, as shown in Table 1.
Table 1 Sociodemographic characteristics of elderly adults living alone
Variables
|
Participants
(N=3726)
|
Non- Institutional care(93.7%)
|
Institutional care(6.3%)
|
χ2
|
P
|
Predisposing factors
|
Gender
|
Male
|
1848(49.6%)
|
1716(92.9%)
|
131(7.1%)
|
3.792
|
0.052
|
Female
|
1878(50.4%)
|
1774(94.5%)
|
105(5.5%)
|
|
|
Age
|
≤69
|
1345(36.1%)
|
1261(93.8%)
|
84(6.2%)
|
4.850
|
0.088
|
70-79
|
1815(48.7%)
|
1711(94.3%)
|
104(5.7%)
|
|
|
≥80
|
566(15.2%)
|
519 (91.7%)
|
47(8.3%)
|
|
|
Education
|
Primary or below
|
2339(62.8%)
|
2208(94.4%)
|
131(5.6%)
|
5.097
|
0.024
|
Junior high school
|
647(17.4%)
|
600(92.7%)
|
47(7.3%)
|
|
|
High school or above
|
740(19.9%)
|
683(92.3%)
|
57(7.7%)
|
|
|
Life support
|
Self
|
1496(40.2%)
|
1405(93.9%)
|
91(6.1%)
|
158.115
|
<0.001
|
Children
|
1823(48.9%)
|
1761(96.6%)
|
62(3.4%)
|
|
|
Government
|
407(10.9%)
|
325(79.9%)
|
82(20.1%)
|
|
|
Knowledge of the combination with medical care
|
Never heard
|
2334(62.6%)
|
2229(95.5%)
|
105(4.5%)
|
54.497
|
<0.001
|
Heard but not understand
|
813(21.8%)
|
757(93.1%)
|
56(6.9%)
|
|
|
Have gained some understanding
|
512(13.7%)
|
447(87.3%)
|
65(12.7%)
|
|
|
Know well
|
67(1.8%)
|
58(86.6%)
|
9(13.4%)
|
|
|
Enabling variables
|
Region
|
Rural
|
2156(57.9%)
|
2018(93.6%)
|
138(6.4%)
|
0.076
|
0.783
|
Urban
|
1570(42.1%)
|
1473(93.8%)
|
97(6.2%)
|
|
|
Monthly income (RMB)
|
≤2000
|
2586(69.4%)
|
2432(94.0%)
|
154(6.0%)
|
1.925
|
0.382
|
2001-4000
|
639(17.1%)
|
592(92.6%)
|
47(7.4%)
|
|
|
≥4001
|
501(13.4%)
|
467(93.2%)
|
34(6.8%)
|
|
|
Have medical insurance
|
No
|
254(6.8%)
|
246(96.9%)
|
8(3.1%)
|
4.599
|
0.032
|
Yes
|
3472(93.2%)
|
3245(93.5%)
|
227(6.5%)
|
|
|
Have endowment insurance
|
No
|
198(5.3%)
|
185(93.4%)
|
13(6.6%)
|
0.024
|
0.878
|
Yes
|
3528(94.7%)
|
3306(93.7%)
|
222(6.3%)
|
|
|
Number of children
|
0
|
176(4.7%)
|
133(75.6%)
|
43(24.4%)
|
123.927
|
<0.001
|
1
|
530(14.2%)
|
477(90.0%)
|
53(10.0%)
|
|
|
2
|
1726(46.3%)
|
1646(95.4%)
|
80(4.6%)
|
|
|
≥3
|
1294(34.7%)
|
1235(95.4%)
|
59(4.6%)
|
|
|
Satisfaction with current health services
|
Very Unsatisfied
|
23(0.6%)
|
21(91.3%)
|
2(8.7%)
|
21.569
|
<0.001
|
Unsatisfactory
|
89(2.4%)
|
75(84.3%)
|
14(15.7%)
|
|
|
Fair
|
560(15.0%)
|
517(92.3%)
|
43(7.7%)
|
|
|
Satisfied
|
2475(66.4%)
|
2321(93.8%)
|
154(6.2%)
|
|
|
Very satisfied
|
579(15.5%)
|
557(96.2%)
|
22(3.8%)
|
|
|
Needs factors
|
Feeling of solitude
|
Often
|
382(10.3%)
|
341(89.3%)
|
41(10.7%)
|
15.622
|
0.001
|
Sometimes
|
1120(30.1%)
|
1047(93.5%)
|
73(6.5%)
|
|
|
Occasionally
|
1638(44.0%)
|
1550(94.6%)
|
88(5.4%)
|
|
|
Never
|
586(15.7%)
|
553(94.4%)
|
33(5.6%)
|
|
|
Number of chronic diseases
|
0
|
863(23.2%)
|
811(94.0%)
|
52(6.0%)
|
0.932
|
0.818
|
1
|
1342(36.0%)
|
1251(93.2%)
|
91(6.8%)
|
|
|
2
|
888(23.8%)
|
836(93.2%)
|
52(5.9%)
|
|
|
≥3
|
633(17.0%)
|
593(93.7%)
|
40(6.3%)
|
|
|
Frequency of medical examinations
|
Never
|
230(6.2%)
|
213(92.6%)
|
7(7.4%)
|
21.749
|
<0.001
|
Once a year
|
3124(83.8%)
|
2949(94.4%)
|
175(5.6%)
|
|
|
Twice a year
|
270(7.2%)
|
241(89.3%)
|
29(10.7%)
|
|
|
Three or more Times a year
|
40(1.1%)
|
34(85.0%)
|
6(15.0%)
|
|
|
Readily
|
62(1.7%)
|
54(87.1%)
|
8(12.9%)
|
|
|
3.2 Factors associated with willingness for institutional care among elderly adults living alone
The dependent variable was the willingness to choose institutional care for the elderly. The variables with P < 0.05 in the χ2 test in Table 1 were included in the binary logistic regression analyses, and the results are shown in Table 2. Of the predisposing factors, sources of life support were children (OR (95% Cl) = 0.623 (60.442 to 0.876)) and government (OR (95% Cl) =3.630 (2.564 to 5.139)); knowledge of the combination with medical care (OR (95% Cl) = 1.739 (1.489 to 2.031)) were significantly associated with the institutional care willingness.
Of the enabling factors, number of children (OR (95% Cl)= 0.658 (0.566~0.765)), satisfaction with current health services (OR (95% Cl)= 0.759 (0.626~0.921)), and the needs factors of feeling of solitude (OR (95% Cl)= 0.796 (0.677~0.934)), frequency of medical check-up (OR (95% Cl)= 1.302 (1.073~ 1.580)) had a statistically significant difference on the impact on the willingness of the elderly adults living alone to choose institutional care (see table 2).
Table 2 Factors associated with willingness for institutional care among elderly living alone
Independent variable
|
Comparison
group
|
Reference group
|
β
|
SE
|
P
|
OR (95% Cl)
|
Education
|
|
|
0.002
|
0.11
|
0.983
|
1.002(0.808~1.244)
|
Life support
|
Children
|
Self
|
-0.474
|
0.174
|
0.007
|
0.623(0.442~0.876)
|
|
Government
|
|
1.289
|
0.177
|
0.000
|
3.630(2.564~5.139)
|
Knowledge of the combination with medical care
|
|
|
0.553
|
0.079
|
0.000
|
1.739(1.489~2.031)
|
Have medical insurance
|
Yes
|
No
|
0.525
|
0.380
|
0.168
|
1.690(0.802~3.562)
|
Number of children
|
|
|
-0.418
|
0.077
|
0.000
|
0.658(0.566~0.765)
|
Satisfaction with current health services
|
|
|
-0.275
|
0.098
|
0.005
|
0.759(0.626~0.921)
|
Feeling of solitude
|
|
|
-0.229
|
0.082
|
0.005
|
0.796(0.677~0.934)
|
Frequency of medical examinations
|
|
|
0.264
|
0.099
|
0.007
|
1.302(1.073~1.580)
|
3.3 Validation of Anderson model fit coefficients
In this study, the impacts of predisposing factors, enabling factors, and needs factors on the willingness of elderly adults living alone to select institutional care were investigated further by including them in Anderson's model, as shown in Table 3. Model I includes the predisposing and enabling factors; Model II includes the predisposing and needs factors; Model III includes the enabling and needs factors; and Model IV includes the predisposing, enabling, and needs factors and serves as the reference group for comparing changes in the values of -2LL, Cox & Snell R2, and Nagelkerke R2 in the model. The smallest change in each value in Model I and the largest change in each value in Model III indicate that the predisposition factor has the greatest influence on the decision to institutionalize care for the elderly, followed by the enabling factor, and the needs factor has the least influence.
Table 3 Anderson Model regression results
|
-2LL
|
−2LL change
value*
|
Cox&Snell R2
|
Ccx&SnellR2 change value*
|
Nagelkerke R2
|
Nagelkerke R2
change value*
|
Model Ⅳ
|
1522.906
|
-
|
0.060
|
-
|
0.160
|
-
|
ModelⅠ
|
1536.951
|
14.045
|
0.057
|
-0.003
|
0.151
|
-0.01
|
ModelⅡ
|
1563.644
|
40.738
|
0.050
|
-0.010
|
0.132
|
-0.029
|
ModelⅢ
|
1649.240
|
126.334
|
0.028
|
-0.032
|
0.074
|
-0.087
|
Note: *Represents the change in value compared to model IV
ModelⅠ: Logitic(Yi)=predisposing factors+enabling variables.
Model II: Logitic(Yi)=predisposing factors+needs factors.
Model III: Logitic(Yi)=enabling variables+needs factors.
Model IV: Logistic (Yi)=predisposing factors+enabling variables+needs factors.
3.4 Gender, region, and age-stratified analysis
To identify the key influencing factors affecting the willingness of elderly adults living alone to receive institutional care, stratified analyses were conducted by gender, region, and age.
The study's findings revealed that the factors influencing the willingness to receive institutional care differed slightly between the elderly's genders, with satisfaction with current health services being an important factor influencing men's choice of institutional care. The feeling of solitude and the frequency of medical examinations influence women's willingness of institutional care. In a comparison of elderly of different ages, the frequency of medical examinations was an important factor influencing the willingness of the elderly under 70 years of age, while satisfaction with current health services is an important factor for those aged 70–79 years. When comparing different regions, feelings of solitude and satisfaction with current health services have a significant impact on urban elderly adults' willingness, see Figure 1.
In conclusion, the main source of life support for the elderly, knowledge of the combination with medical care, and the number of children have an impact on the different characteristics of the elderly.