Table 4. Multivariate analysis using PGG1 vs. PGG2/3 as the dependent variables.
| PGG1 vs. PGG2/3 * | OR | Standard error | Z score | P-value | [95% CI] |
|---|---|---|---|---|---|
| French | 0.60 | 0.14 | -2.21 | 0.027 | [0.39; 0.95] |
| Year ** | 1.07 | 0.04 | 2.06 | 0.039 | [1.003; 1.14] |
| Male Gender | 1.21 | 0.26 | 1.11 | 0.359 | [0.81; 1.81] |
| Age ** | 0.98 | 0.005 | -3.88 | <0.001 | [0.97; 0.99] |
| Pulmonary-TB | 0.54 | 0.11 | -2.95 | 0.003 | [0.36; 0.81] |
* Cases belonging to PGG1 group were coded 1, and cases belonging to PGG2/3 group were coded 0; Reference categories considered were foreign-born, female gender, and extrapulmonary cases. Note that data without information on origin, sex, and localization of disease was excluded from this analysis (further details are provided in S1 File).
** Note that variables "Year" and "Age" were taken as continuous variables. This means that they have not to be taken into account in the reference categories list (foreign born, female gender, and extrapulmonary cases). Reference categories are used only for categorical variables.