Baseline characteristics of the 16,651 study subjects (6,797 males, 9,854 females) are described in Table 1. The living conditions and prevalence of health-related factors and comorbidities were as follows: living alone (8.5%), smoking (12.5%), alcohol consumption (55.3%), hypertension (46.3%), diabetes (11.2%), dyslipidemia (57.3%), and obesity (24.0%)
Table 1. Baseline characteristics of study subjects
|
Total subjects
|
Males
|
Females
|
Number
|
16,651
|
6,797
|
9,854
|
Age (years)
|
62.7 (8.4)
|
64.1 (7.9)
|
61.7(8.5)
|
Body mass index (kg/m2)
|
23.0 (3.2)
|
23.6 (3.0)
|
22.6 (3.3)
|
Living alone (%)
|
8.5
|
8.0
|
8.9
|
Education period)
|
|
|
|
≦ 9 years (%)
|
14.9
|
17.6
|
13.0
|
10-12 years (%)
|
55.2
|
55.5
|
55.2
|
≧ 13 years (%)
|
29.9
|
26.9
|
32.0
|
Smoking (%)
|
12.5
|
22.8
|
5.1
|
Alcohol consumption (%)
|
55.3
|
79.9
|
38.0
|
Hypertension (%)
|
46.3
|
55.9
|
39.7
|
Diabetes (%)
|
11.2
|
20.7
|
9.2
|
Dyslipidemia (%)
|
57.3
|
55.7
|
58.7
|
Obesity (%)
|
24.0
|
29.1
|
20.5
|
|
|
|
|
|
|
Mean (SD)
Status of social support
Table 2 shows the results of the five component social support questions. Overall, the prevalence of the lack of someone 1) whom you can consult when you are in trouble was 12.7%; 2) whom you can consult when your physical condition is not good was 10.6%; 3) who can help you in daily homework was 15.0%; 4) who can take you to a hospital when you don't feel well was 7.7%; 5) who can take care of you when you are ill in bed was 11.4%. For most of the five components, men had a higher prevalence of lacking social support than women. Logistic regression analysis of factors related to a lack of social support components showed that male sex, age under 65, and living alone were independently associated with the lack of almost all five social support components (Table 3). Further, smoking was associated with the lack of someone 3) who can help you in daily homework and 5) who can take care of you when you are ill in bed. In contrast, alcohol consumption was associated with having someone 1) whom you can consult when you are in trouble, 2) whom you can consult when your physical condition is not good, and 5) who can take care of you when you are ill in bed. Comorbidity factors were not associated with the lack of social support components.
Table 2. The prevalence of the lack of social support components
|
Total subjects
|
Males
|
Females
|
1) No one whom you can consult when you are in trouble (%)
|
12.7
|
19.1
|
8.2
|
2) No one whom you can consult when your physical condition is not good (%)
|
10.6
|
15.1
|
7.6
|
3) No one who can help you in daily housework (%)
|
15.0
|
19.1
|
12.3
|
4) No one who can take you to a hospital when you don’t feel well (%)
|
7.7
|
9.5
|
6.4
|
5) No one who can take care of you when you are ill in bed (%)
|
11.4
|
10.7
|
12.1
|
Table 3. Multivariate logistic regression analysis to predict the lack of social support components.
|
1) Consult at trouble
|
2) Consult at physical condition
|
3) Support for daily housework
|
4) Support for transportation to hospital
|
5) Take care of you when sick
|
|
OR (95%CI)
|
OR (95%CI)
|
OR (95%CI)
|
OR (95%CI)
|
OR (95%CI)
|
Male gender
|
2.78 (2.45-3.15)
|
2.53 (2.21-2.90)
|
1.84 (1.64-2.07)
|
1.89 (1.61-2.22)
|
0.92 (0.80-1.05)
|
<65 years
|
1.23 (1.10-1.39)
|
1.37 (1.22-1.54)
|
1.12 (1.01-1.25)
|
1.54 (1.34-1.78)
|
1.35 (1.19-1.54)
|
Living alone
|
1.87 (1.60-2.18)
|
2.48 (2.12-2.90)
|
4.71 (4.13-5.34)
|
10.7 (9.23-12.4)
|
10.2 (8.90-11.6)
|
Education period
|
|
|
|
|
|
≦ 9 years (%)
|
Reference
|
Reference
|
Reference
|
Reference
|
Reference
|
10-12 years (%)
|
1.06 (0.91-1.24)
|
1.09 (0.91-1.30)
|
0.89 (0.77-1.03)
|
0.93 (0.76-1.15)
|
0.96 (0.80-1.15)
|
≧ 13 years (%)
|
1.14 (0.96-1.35)
|
1.21 (1.01-1.47)
|
0.95 (0.81-1.12)
|
1.08 (0.86-1.35)
|
1.13 (0.93-1.38)
|
Smoking
|
1.10 (0.95-1.27)
|
1.08 (0.92-1.26)
|
1.18 (1.03-1.36)
|
1.17 (0.93-1.36)
|
1.34 (1.13-1.60)
|
Alcohol consumption
|
0.81 (0.72-0.91)
|
0.84 (0.74-0.95)
|
0.90 (0.81-1.01)
|
0.88 (0.75-1.03)
|
0.84 (0.74-0.96)
|
Hypertension
|
1.04 (0.93-1.16)
|
1.15 (1.02-1.30)
|
0.98 (0.88-1.09)
|
1.01 (0.88-1.17)
|
1.00 (0.88-1.13)
|
Diabetes
|
1.06 (0.91-1.24)
|
1.02 (0.85-1.21)
|
1.11 (0.91-1.12)
|
1.01 (0.81-1.24)
|
0.99 (0.82-1.19)
|
Dyslipidemia
|
1.02 (0.91-1.13)
|
1.00 (0.89-1.12)
|
0.99 (0.89-1.10)
|
0.90 (0.78-1.04)
|
0.88 (0.78-0.99)
|
Obesity
|
1.06 (0.94-1.19)
|
1.08 (0.94-1.23)
|
1.02 (0.90-1.14)
|
0.96 (0.81-1.13)
|
1.08 (0.94-1.24)
|
CI: confidence interval, OR: odds ratio.
Association between social support and mortality.
During the 5-year follow-up period (median 3.4 years), there were 166 total deaths (116 males, 50 females) and 38 cardiovascular deaths (27 males, 11 females). First, we performed a Kaplan-Meier survival analysis of the association between each component of social support and mortality. A significant association with all-cause and cardiovascular mortality was observed only with the transport support to hospital, but not with the other components. The Kaplan-Meier analysis in Fig. 1 compared the survival curves between the two groups (yes/no) regarding the presence of "Support for transportation to hospital when sick", based on the answer to question 4). It showed that the all-cause and cardiovascular mortality of those without transport support to the hospital were significantly higher than those with such support.
Similarly, unadjusted Cox proportional hazard analysis showed that only a lack of support for transportation to hospital had a significant association with all-cause mortality and cardiovascular mortality (HR, 2.01, 95% CI, 1.26-3.05 for all-cause mortality, and HR, 3.30, 95% CI, 1.41-6.87 for cardiovascular mortality) (Table 4). These associations remained significant after adjustment for possible confounding factors (HR, 1.88, 95% CI, 1.15-2.93 for all-cause mortality, and HR, 2.73, 95% CI, 1.09-5.95 for cardiovascular mortality) (Table 4). In this multivariate model, the independent risk factors for all-cause mortality, other than support for transport to the hospital, were older age, men, diabetes, and smoking. On analysis by sex, the significant association with all-cause mortality was observed in males (adjusted HR, 2.07, 95% CI, 1.18-3.40) but not in females (adjusted HR, 1.46, 95%CI, 0.44-3.62). We further examined the association between all-cause mortality and structural aspects of social support (living alone) using the same multivariate model. However, this factor did not show an independent association with all-cause mortality.
Table 4. The association between the lack of social support components and mortality.
|
Unadjusted
|
Adjusted*
|
All-cause mortality
|
HR (95%CI)
|
P value
|
HR (95%CI)
|
P value
|
1) No one whom you can consult when you are in trouble
|
1.44 (0.94-2.12)
|
0.09
|
1.15 (0.73-1.74)
|
0.53
|
2) No one whom you can consult when your physical condition is not good
|
1.27 (0.77-1.94)
|
0.34
|
1.18 (0.71-1.85)
|
0.50
|
3) No one who can help you in daily housework
|
1.38 (0.93-2.01)
|
0.11
|
1.18 (0.78-1.76)
|
0.41
|
4) No one who can take you to a hospital when you don’t feel well
|
2.01 (1.26-3.05)
|
<0.01
|
1.88 (1.15-2.93)
|
0.01
|
5) No one who can take care of you when you are ill in bed
|
1.02 (0.60-1.61)
|
0.94
|
1.09 (0.62-1.77)
|
0.76
|
|
|
|
|
|
Cardiovascular mortality
|
HR (95%CI)
|
P value
|
HR (95%CI)
|
P value
|
1) No one whom you can consult when you are in trouble
|
1.88 (0.80-3.90)
|
0.14
|
1.38 (0.55-3.02)
|
0.47
|
2) No one whom you can consult when your physical condition is not good
|
1.31 (0.45-3.05)
|
0.59
|
0.94 (0.28-2.39)
|
0.90
|
3) No one who can help you in daily housework
|
1.54 (0.66-3.20)
|
0.30
|
1.18 (0.47-2.57)
|
0.70
|
4) No one who can take you to a hospital when you don’t feel well
|
3.30 (1.41-6.87)
|
<0.01
|
2.73 (1.09-5.95)
|
0.03
|
5) No one who can take care of you when you are ill in bed
|
1.27 (0.43-2.94)
|
0.64
|
1.11 (0.33-2.83)
|
0.84
|
CI: confidence interval, HR: hazard ratio.
*Adjusted for age, gender, education period, hypertension, diabetes, dyslipidemia, obesity, alcohol consumption, smoking.