Predictor
|
Number (%) of respondents with a summed average behavior score > 3
|
Unadjusted
|
Adjusted
|
---|
OR
|
p
|
OR
|
95% Confidence Interval
|
p
|
---|
Gender
|
Male (Reference)
|
802 (67.5)
| | | | | | |
Female
|
1219 (73.6)
|
1.34
|
< 0.001
|
1.21
|
1.00
|
1.44
|
0.040
|
Age (years)
| | |
0.001
| | | |
0.009
|
< 25 (Reference)
|
811 (67.4)
| | | | | | |
25–40
|
901 (73.1)
|
1.32
|
0.002
|
1.31
|
1.01
|
1.69
|
0.040
|
> 40
|
30 (75.7)
|
1.51
|
0.002
|
1.79
|
1.25
|
2.58
|
0.002
|
Education
| | | | | |
< University (Reference)
|
1424 (72.4)
| | | | | | |
University
|
597 (67.9)
|
1.24
|
0.014
|
1.11
|
0.91
|
1.35
|
0.298
|
Marital status
|
Not married (Reference)
|
1043 (67.9)
| | | | | | |
Married
|
978 (74.7)
|
1.39
|
< 0.001
|
1.03
|
0.79
|
1.34
|
0.832
|
Area
|
Other region (Reference)
|
1368 (70.9)
| | | | | | |
High-risk area
|
653 (71.4)
|
1.02
|
0.790
|
1.03
|
0.85
|
1.25
|
0.756
|
Knowledge
|
Lower level (Reference)
|
1232 (66.1)
| | | | | | |
High level
|
789 (80.3)
|
2.09
|
< 0.001
|
1.77
|
1.46
|
2.15
|
< 0.001
|
Perceived susceptibility
|
≤ 3 (Reference)
|
1417 (70.3)
| | | | | | |
> 3 (High)
|
604 (72.8)
|
1.13
|
0.191
|
1.01
|
0.73
|
1.38
|
0.966
|
Perceived severity
|
≤ 3 (Reference)
|
78 (48.8)
| | | | | | |
> 3 (High)
|
1943 (72.4)
|
2.75
|
< 0.001
|
1.90
|
1.54
|
2.34
|
< 0.001
|
Perceived controllability
|
≤ 3 (Reference)
|
851 (71.5)
| | | | | | |
> 3 (High)
|
1170 (70.7)
|
0.96
|
0.635
|
1.03
|
0.85
|
1.24
|
0.769
|
Negative emotion
|
≤ 3 (Reference)
|
379 (62.2)
| | | | | | |
> 3 (High)
|
1642 (73.4)
|
1.68
|
< 0.001
|
1.36
|
1.10
|
1.68
|
0.005
|
Self-reported health
|
≤ 3 (Reference)
|
451 (58.6)
| | | | | | |
> 3 (Good)
|
1570 (75.6)
|
2.19
|
< 0.001
|
1.94
|
1.61
|
2.34
|
< 0.001
|
Information attention
|
≤ 3 (Reference)
|
32 (20.5)
| | | | | | |
> 3 (High)
|
1989 (74.0)
|
3.69
|
< 0.001
|
4.16
|
2.74
|
6.32
|
< 0.001
|
Trust in official media
|
≤ 3 (Reference)
|
162 (49.7)
| | | | | | |
> 3 (High)
|
1859 (73.8)
|
2.85
|
< 0.001
|
1.97
|
1.52
|
2.56
|
< 0.001
|
- Note:Unadjusted OR was based on unary logistic regression, then incorporate all variables into the model to produce adjusted results