Table 5. Summary of stepwise multivariate logistic regression model predicting “probable secondary infection” within the household levela.
Risk factor | Coefficient | SE | Odds ratio (95% CI) | p value |
---|---|---|---|---|
Type of Index Person (IP) |
|
|
|
|
Healthcare worker |
|
|
1.00 |
|
Amoy Gardens Block E residents |
3.074 |
0.487 |
21.62 (8.33 to 56.10) |
<0.001 |
Amoy Gardens other Block residents |
1.901 |
0.425 |
6.69 (2.91 to 15.39) |
<0.001 |
Other community member |
1.705 |
0.354 |
5.50 (2.75 to 11.01) |
<0.001 |
Date of IP’s fever onset |
|
|
|
|
Before March 25 |
|
|
1.00 |
|
On or after March 25 |
–0.696 |
0.235 |
0.50 (0.32 to 0.79) |
<0.001 |
Duration IP stayed home between fever onset and hospitalization (d) |
|
|
|
|
<2 |
|
|
1.00 |
|
3–5 |
0.283 |
0.258 |
1.33 (0.80 to 2.20) |
0.274 |
>6 |
1.045 |
0.265 |
2.84 (1.69 to 4.78) |
<0.001 |
IP visited by any household member when hospitalized? |
|
|
|
|
No |
|
|
1.00 |
|
Yes | 0.483 | 0.242 | 1.62 (1.01 to 2.60) | 0.046 |
aAge was not significant in the multivariable analysis.