Characteristics of the participants
Five hundred sixteen eligible participants aged 60 to 90 years old (Mean age = 67.6 ± 6.6 yrs.) participated in the study. As seen in Table 1, the sample includes 57.9% females, and 68.6% participants were aged from 60 to 69 years. Most of the elderly were married (83.7%) and reported living with their spouse, partner, or children (90.7%). Nearly half (46.5%) of the old adults received college or above education, and more than half (57.9%) reported an average household income level. A total of 92.6% were pensioners/retired. 52.1% of the elderly participants were identified as overweight or obese (BMI ≥ 26 kg/m2). In terms of medical history, about half of the participants (50.8%) suffered from chronic diseases (e.g., heart diseases, diabetes, or cancer). A few participants reported that their acquaintances (e.g., family members, friends, or neighbors) had been confirmed with COVID-19 (9.7%). According to QoL levels, the majority of the participants (78.5%) reported high-level QoL, while 6.0% of the elderly reported low-level QoL and 15.5% of the elderly indicated middle-level QoL during the outbreak of COVID-19. The means of behaviors are shown in Table 1 (mean PA stage = 3.83 (1.54); mean FVI stage = 3.77 (1.49); mean PB = 3.61(0.40).
Table 1: Descriptive characteristics of the study sample (n = 516)
Variable
|
N (%)
|
|
|
Gender, n (%)
|
|
|
Male
|
217 (42.1%)
|
|
Female
|
299 (57.9%)
|
|
Living situation, n (%)
|
|
|
Live alone
|
48 (9.3%)
|
Live with others
|
468 (90.7%)
|
|
Age group, n (%)
|
|
|
60-69 years old
|
354 (68.6%)
|
|
70-79 years old
|
128 (24.8%)
|
|
80 years old and above
|
34 (6.6%)
|
|
Marital status, n (%)
|
|
|
Single
|
14 (2.7%)
|
|
Married
|
432 (83.7%)
|
|
Divorced or widowed
|
70 (13.6%)
|
|
Educational level, n (%)
|
|
|
Primary school or below
|
45 (8.7%)
|
|
Middle or High school
|
231 (44.8%)
|
|
College or above
|
240 (46.5%)
|
|
Professional status, n (%)
|
|
|
Unemployed
|
22 (4.3%)
|
|
Pensioner or Retired
|
478 (92.6%)
|
|
Employed
|
16 (3.1%)
|
|
Economic situation, n (%)
|
|
|
Below average
|
113 (21.9%)
|
|
Average
|
299 (57.9%)
|
|
Above average
|
104 (20.2%)
|
|
Body mass index (BMI), n (%)
|
|
|
BMI < 18.5 kg/m2
|
19 (3.7%)
|
|
18.5 kg/m2 ≤ BMI < 23 kg/m2
|
228 (44.2%)
|
|
23 kg/m2 ≤ BMI < 26 kg/m2
≤ BMI < 26 kg/m2
|
206 (39.9%)
|
|
BMI ≥ 26 kg/m2
|
63 (12.2%)
|
|
Chronic diseases, n (%)
|
|
|
Yes
|
262 (50.8%)
|
|
No
|
254 (49.2%)
|
|
Infected cases of acquaintances, n (%)
|
|
|
Yes
|
50 (9.7%)
|
|
No
|
466 (90.3%)
|
|
QoL, mean (SD): 3.76 (0.61)
|
|
|
Low
|
31 (6.0%)
|
|
Middle
|
80 (15.5%)
|
|
High
|
405 (78.5%
|
|
Lifestyle behaviors
|
|
|
PA stage, mean (SD): 3.83 (1.54)
|
|
|
FVI stage, mean (SD): 3.77 (1.49)
|
|
|
Preventive behaviors, mean (SD): 3.61 (0.40)
|
|
|
Note. SD = standard deiation. PA= Physical Activity, FVI= Fruit and vegetable intake
Characteristics of Quality of Life
As shown in Table 2, older adults’ QoL differed significantly for different characteristics. There were no significant differences in QoL across gender (t514 = -0.26, p = 0.796), BMI intervals (F3, 512 = 1.96, p = 0.119) and infected cases of acquaintances (t514 = -1.61, p = 0.109). The QoL was significantly higher for participants who had better economic situations (t2, 513 = 14.52, p < 0.001) and reported no chronic diseases (t514 = -5.43, p < 0.001). Old adults who were married (F2, 513 = 5.18, p < 0.01) with better education (F2, 513 = 6.98, p < 0.01) reported significantly better QoL. The poorer QoL was identified among those who lived alone (t514 = -2.43, p < 0.05) and were aged over 80 years old (F2, 513 = 4.38, p < 0.05). The employed old adults reported better QoL compared with those who were unemployed, pensioners and those who retired elderly (F2, 513 = 4.25, p < 0.05).
Table 2: Characteristics of quality of life (n = 516)
Variable
|
QoL Mean (SD)
|
F/t
|
P
|
Gender, n (%)
|
|
t514 = -0.26
|
0.796
|
Male
|
3.75 (0.60)
|
|
|
Female
|
3.77 (0.62)
|
|
|
Living situation, n (%)
|
|
t514 = -2.43
|
< 0.05
|
Live alone
|
3.51 (0.77)
|
|
|
Live with others
|
3.78 (0.59)
|
|
|
Age group, n (%)
|
|
F2, 513 = 4.38
|
< 0.05
|
60-69 years old
|
3.80 (0.60)
|
|
|
70-79 years old
|
3.70 (0.61)
|
|
|
80 years old and above
|
3.51 (0.68)
|
|
|
Marital status, n (%)
|
|
F2, 513 = 5.18
|
< 0.01
|
Single
|
3.64 (0.82)
|
|
|
Married
|
3.80 (0.57)
|
|
|
Divorced or widowed
|
3.54 (0.74)
|
|
|
Educational level, n (%)
|
|
F2, 513 = 6.98
|
< 0.01
|
Primary school or below
|
3.44 (0.78)
|
|
|
Middle or High school
|
3.77 (0.59)
|
|
|
College or above
|
3.81 (0.58)
|
|
|
Professional status, n (%)
|
|
F2, 513 = 4.25
|
< 0.05
|
Unemployed
|
3.41 (0.68)
|
|
|
Pensioner or Retired
|
3.77 (0.60)
|
|
|
Employed
|
3.90 (0.58)
|
|
|
Economic situation, n (%)
|
|
F2, 513 = 14.52
|
< 0.001
|
Below average
|
3.50 (0.68)
|
|
|
Average
|
3.83 (0.57)
|
|
|
Above average
|
3.86 (0.58)
|
|
|
Body mass index (BMI)
|
|
|
|
BMI < 18.5 kg/m2
|
3.82 (0.630
|
F3, 512 = 1.96
|
0.119
|
18.5 kg/m2 ≤ BMI < 23 kg/m2
|
3.72 (0.61)
|
|
|
23 kg/m2 ≤ BMI < 26 kg/m2
≤ BMI < 26 kg/m2
|
3.84 (0.59)
|
|
|
BMI ≥ 26 kg/m2
|
3.67 (0.61)
|
|
|
Chronic diseases, n (%)
|
|
t514 = -5.43
|
< 0.001
|
Yes
|
3.62 (0.61)
|
|
|
No
|
3.90 (0.58)
|
|
|
Infected cases of acquaintances
|
|
|
|
Yes
|
3.63 (0.67)
|
t514 = -1.61
|
0.109
|
No
|
3.78 (0.60)
|
|
|
Note. SD = standard deviation.
|
|
|
|
Association of PA Stage, FVI Stage and Preventive Behaviors with QoL
Based on the characteristics of QoL, 6 significant demographic variables (living situation, age group, marital status, educational level, professional status, and economic situation) were entered as predictors in Model 1. Dummy variables were applied for all polynomial predictors. Model 1 explained 9% of the variance in QoL (p < 0.001). Medical history of chronic diseases and infected cases of acquaintances were entered as covariates into Model 2 contributing to the explanation of 5% of the variance in QoL (ΔR2 = 0.05, p < 0.001). After controlling demographics and covariates, PA stage, FVI stage and preventive behaviors the lifestyle behaviors were entered to Model 3, contributing to a significant improvement in the variance explanation (ΔR2 = 0.06, p < 0.001). Model 3 accounted for 20% explanation power of the variance in QoL. The economic situation (β average vs. below average = 0.17, p < 0.01, 95%CI = 0.08 to 0.33; βabove average vs. below average = 0.15, p < 0.01, 95%CI = 0.07 to 0.39), chronic diseases (β = 0.19, p < 0.001, 95%CI = 0.14 to 0.34), FVI stage (β = 0.21, p < 0.001, 95%CI = 0.05 to 0.12) and preventive behaviors (β = 0.10, p < 0.05, 95%CI = 0.03 to 0.29) can significantly predict the QoL of old adults. Details of hierarchical regression analysis is shown in Table 3.
Table 3: Hierarchical multiple linear regression analysis of demographics, covariate and lifestyle behaviors with QoL (n=516)
|
Model1
|
Model2
|
Model3
|
Variable
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
Block 1: Demographics
|
|
|
|
|
|
|
|
|
|
Living situation
|
|
|
|
|
|
|
|
|
|
Live alone
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Live with others
|
0.12(0.10)
|
(-0.08, 0.33)
|
0.06
|
0.13(0.10)
|
(-0.07, 0.32)
|
0.06
|
0. 90(0.10)
|
(-0.10, 0.28)
|
0.04
|
Age group
|
|
|
|
|
|
|
|
|
|
60-69 years old
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
70-79 years old
|
-0.04(0.06)
|
(-0.16, 0.09)
|
-0.03
|
0.01(0.06)
|
(-0.11, 0.13)
|
0.01
|
0.02(0.06)
|
(-0.09, 0.14)
|
0.02
|
80 years old and above
|
-0.25(0.11)
|
(-0.47, -0.04)
|
-0.10*
|
-0.20(0.11)
|
(-0.41, 0.01)
|
-0.08
|
-0.13(0.11)
|
(-0.34, 0.07)
|
-0.05
|
Marital status
|
|
|
|
|
|
|
|
|
|
Single
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Married
|
0.13(0.16)
|
(-0.19, 0.45)
|
0.08
|
0.11(0.16)
|
(-0.20, 0.42)
|
0.07
|
0.09(0.15)
|
(-0.22, 0.39)
|
0.05
|
Divorced or widowed
|
0.06(0.18)
|
(-0.29, 0.40)
|
0.03
|
0.06(0.17)
|
(-0.28, 0.39)
|
0.03
|
0.04(0.17)
|
(-0.29, 0.36)
|
0.02
|
Educational level
|
|
|
|
|
|
|
|
|
|
Primary school or below
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Middle or High school
|
0.16(0.11)
|
(-0.05, 0.37)
|
0.13
|
0.21(0.11)
|
(0.00, 0.42)
|
0.17*
|
0.14(0.10)
|
(-0.06, 0.34)
|
0.12
|
College or above
|
0.13(0.11)
|
(-0.09, 0.35)
|
0.11
|
0.19 (0.11)
|
(-0.02, 0.41)
|
0.16
|
0.11 (0.11)
|
(-0.10, 0.32)
|
0.09
|
Professional status
|
|
|
|
|
|
|
|
|
|
Unemployed
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Pensioner or Retired
|
0.16(0.14)
|
(-0.11, 0.44)
|
0.07
|
0.21(0.14)
|
(-0.06, 0.48)
|
0.09
|
0.17(0.13)
|
(-0.09, 0.43)
|
0.07
|
Employed
|
0.24(0.21)
|
(-0.16, 0.65)
|
0.07
|
0.26(0.20)
|
(-0.13, 0.65)
|
0.07
|
0.29(0.19)
|
(-0.09, 0.67)
|
0.08
|
Economic situation
|
|
|
|
|
|
|
|
|
|
Below average
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
Average
|
0.30(0.07)
|
(0.16, 0.43)
|
0.24***
|
0.26(0.07)
|
(0.13, 0.39)
|
0.21***
|
0.21(0.07)
|
(0.08, 0.33)
|
0.17**
|
Above average
|
0.32(0.09)
|
(0.16, 0.49)
|
0.21***
|
0.29(0.08)
|
(0.12, 0.45)
|
0.19***
|
0.23(0.08)
|
(0.07, 0.39)
|
0.15**
|
Block 2: Covariates
|
|
|
|
|
|
|
|
|
|
Chronic diseases
|
|
|
|
|
|
|
|
|
|
Yes
|
--
|
--
|
--
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
No
|
--
|
--
|
--
|
0.26(0.05)
|
(0.16, 0.36)
|
0.21***
|
0.24(0.05)
|
(0.14 0.34)
|
0.19***
|
Infected cases of acquaintances
|
|
|
|
|
|
|
|
|
|
Yes
|
--
|
--
|
--
|
Reference
|
N/A
|
N/A
|
Reference
|
N/A
|
N/A
|
No
|
--
|
--
|
--
|
0.12(0.09)
|
(-0.05, 0.29)
|
0.06
|
0.10 (0.09)
|
(-0.07, 0.27)
|
0.05
|
Block 3: Lifestyle behaviors
|
|
|
|
|
|
|
|
|
|
PA stage
|
--
|
--
|
--
|
--
|
--
|
--
|
0.01 (0.02)
|
(-0.03, 0.04)
|
0.02
|
FVI stage
|
--
|
--
|
--
|
--
|
--
|
--
|
0.08 (0.02)
|
(0.05 0.12)
|
0.21***
|
Preventive behaviors
|
--
|
--
|
--
|
--
|
--
|
--
|
0.16 (0.07)
|
(0.03 0.29)
|
0.10*
|
Moderating Effect of Socioeconomic Status
Correlation analyses revealed that educational level (r = 0.13, p < 0.01), professional status (r = 0.12, p < 0.01) and economic situation (r = 0.20, p < 0.001) were significantly associated with QoL. In addition, except PA stage (r = 0.02, p = 0.449), FVI stage (r = 0.21, p < 0.001) and preventive behaviors (r = 0.10, p < 0.05) were significantly related to QoL.
In terms of the moderating effects of socioeconomic status between FVI and QoL, Table 4 shows that educational level, professional status, and economic situation significantly predicted old adults’ QoL in model 1 (R2 = 0.22, p < 0.001), FVI stage significantly contributed to model 2 (ΔR2 = 0.14, p < 0.001), the interactions of SES with FVI stage did not significantly contribute to model 3 (ΔR2 = 0.00, p = 0.510). In terms of moderating effects of socioeconomic status between preventive behaviors and QoL, Table 5 shows that economic situation significantly predicted old adults’ QoL in model 1 (R2 = 0.22, p < 0.001), preventive behaviors significantly contributed to model 2 (ΔR2 = 0.06, p < 0.001), the interactions of SES with preventive behaviors significantly contributed to model 3 (ΔR2 = 0.05, p < 0.01). In particular, 2 out of 3 interaction terms (preventive behaviors * educational level, β = -1.3, p < 0.01, 95%CI = -0.51 to -0.09; preventive behaviors * economic situation, β = -0.97, p < 0.05, 95%CI = -0.43 to -0.03) were significantly associated with QoL among old adults.
Table 4: Hierarchical multiple linear regression examining main and interaction effects of socioeconomic status and FVI measures on QoL (n = 516)
|
|
Model 1
|
|
Model 2
|
Model 3
|
|
|
Variable
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
Educational level
|
0.05(0.04)
|
(-0.03, 0.14)
|
0.06
|
0.03(0.04)
|
(-0.05, 0.11)
|
0.03
|
0.14 (0.11)
|
(-0.08, 0.36)
|
0.15
|
Professional status
|
0.17(0.10)
|
(-0.03, 0.37)
|
0.08
|
0.17(0.10)
|
(-0.02, 0.36)
|
0.08
|
0.28 (0.22)
|
(-0.15, 0.7)
|
0.12
|
Economic situation
|
0.16(0.04)
|
(0.07, 0.24)
|
0.17***
|
0.14(0.04)
|
(0.05, 0.22)
|
0.14***
|
0.07 (0.11)
|
(-0.14, 0.29)
|
0.08
|
FVI Stage
|
--
|
--
|
--
|
0.12(0.02)
|
(0.08, 0.15)
|
0.28***
|
0.23 (0.11)
|
(0.01, 0.45)
|
0.56*
|
FVI Stage x Educational level
|
--
|
--
|
--
|
--
|
--
|
--
|
-0.03 (0.03)
|
(-0.09, 0.02)
|
-0.23
|
FVI Stage x Professional status
|
--
|
--
|
--
|
--
|
--
|
--
|
-0.04 (0.06)
|
(-0.15, 0.08)
|
-0.18
|
FVI Stage x Economic situation
|
--
|
--
|
--
|
--
|
--
|
--
|
0.02 (0.03)
|
(-0.04, 0.07)
|
0.11
|
Note: FVI= Fruit and vegetable intake. B = Unstandardized coefficient; SE = Standard error; β = Standardized coefficient. -- = Data do not include in this model. *: p < .05; **: p <.01; ***: p <.001, 2 tailed. Model 1 R2 = 0.22; Model 2 R2 = 0.36; Model 3 R2 = 0.36.
Table 5: Hierarchical multiple linear regression examining main and interaction effects of socioeconomic status and Preventive behaviors on QoL (n = 516)
|
Model 1
|
|
Model 2
|
Model 3
|
|
|
Variable
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
B (SE)
|
95%CI
|
β
|
Educational level
|
0.05(0.04)
|
(-0.03, 0.14)
|
0.06
|
0.03(0.04)
|
(-0.06, 0.11)
|
0.03
|
1.11 (0.39)
|
(0.34, 1.87)
|
1.16**
|
Professional status
|
0.17(0.10)
|
(-0.03, 0.37)
|
0.08
|
0.17(0.10)
|
(-0.02, 0.37)
|
0.08
|
-0.41 (0.76)
|
(-1.91, 1.08)
|
-0.18
|
Economic situation
|
0.16(0.04)
|
(0.07, 0.24)
|
0.17***
|
0.13(0.04)
|
(0.05, 0.22)
|
0.14**
|
0.95 (0.37)
|
(0.24, 1.67)
|
1.01**
|
Preventive behaviors
|
--
|
--
|
--
|
0.28(0.07)
|
(0.15, 0.41)
|
0.18***
|
1.09 (0.43)
|
(0.24, 1.93)
|
0.71*
|
Preventive behaviors x Educational level
|
--
|
--
|
--
|
--
|
--
|
--
|
-0.30 (0.11)
|
(-0.51, -0.09)
|
-1.30**
|
Preventive behaviors x Professional status
|
--
|
--
|
--
|
--
|
--
|
--
|
0.16 (0.22)
|
(-0.27, 0.58)
|
0.32
|
Preventive behaviors x Economic situation
|
--
|
--
|
--
|
--
|
--
|
--
|
-0.23 (0.10)
|
(-0.43, -0.03)
|
-0.97*
|
Note: B = Unstandardized coefficient; SE = Standard error; β = Standardized coefficient. -- = Data do not include in this model. *: p < .05; **: p < .01; ***: p < .001, 2 tailed. Model 1 R2 = 0.22; Model 2 R2 = 0.28; Model 3 R2 = 0.33.
To further analyze the significant interaction effects, simple slopes analyses was conducted. In terms of the moderating effects of education level on the relationship between preventive behaviors and QoL, Figure 1 shows that preventive behaviors were significantly associated with QoL at primary school or below of educational level (β = 0.78, t510 = 3.86, 95%CI = 0.38 to 1.18, p < 0.001) and at the middle or high school educational level (β = 0.34, t510 = 3.79, 95%CI = 0.16 to 0.52, p < 0.001), while the association was not significant at college or above for educational level (β = 0.09, t510 = 0.82, 95%CI = -0.13 to 0.32, p = 0.411). In terms of the moderating effects of economic situation on the relationship between preventive behaviors and QoL, Figure 2 shows that preventive behaviors were significantly associated with QoL at the below average level for economic situation (β = 0.58, t510 = 4.46, 95%CI = 0.33 to 0.84, p < 0.001) and at the average economic situation (β = 0.18, t510 = 2.06, 95%CI = 0.01 to 0.36, p = 0.040, while the association was not significant at the above average level for economic situation (β = 0.11, t510 = 0.73, 95%CI = -0.18 to 0.41, p = 0.464).