Final multivariable model for Herpes Simplex Virus-2
| Level | Type of variable | Characteristic | Males | Females |
| OR* (95% CI) | OR* (95% CI) | |||
| Level 1 (distal variables) | Socio-demographic | Age | 1.04 (1.0 to 1.1) | 1.02 (1.0 to 1.0) |
| Site† | ||||
| Bangalore | 0.7 (0.5 to 1.0) | 0.7 (0.5 to 0.9) | ||
| Hyderabad | 2.1 (1.5 to 3.0) | 1.9 (1.4 to 2.7) | ||
| Chandigarth | 0.7 (0.5 to 1.1) | 0.8 (0.5 to 1.1) | ||
| Religion | ||||
| Hindu | 1.0 | 1.0 | ||
| Christian | 4.4 (2.1 to 9.0) | 2.8 (1.4 to 5.7) | ||
| Sikh | 2.2 (1.0 to 4.8) | 2.2 (1.1 to 4.5) | ||
| Muslim | 1.1 (0.1 to 9.0) | 1.1 (0.2 to 5.5) | ||
| Marital status | ||||
| Never married | 1.0 | |||
| Ever married | 4.1 (1.2 to 13.7) | |||
| Education | ||||
| Graduate | 1.0 | 1.0 | ||
| Primary | 0.8 (0.4 to 1.4) | 1.4 (0.8 to 2.8) | ||
| llliterate | 1.0 (0.5 to 1.9) | 1.5 (0.8 to 2.6) | ||
| Level 2a (proximal variables) | Non-sexual behaviour | Syphilis status | ||
| Negative | 1.0 | 1.0 | ||
| Positive | 6.2 (2.2 to 17.4) | 5.6 (1.7 to 17.8) | ||
| Alcohol consumption | ||||
| None | 1.0 | 1.0 | ||
| 1–2 times per week | 0.5 (0.2 to 1.0) | 1.3 (0.7 to 2.5) | ||
| ≥3 times per week | 1.0 (0.5 to 1.9) | 0.6 (0.2 to 1.7) | ||
| Ever smoking | ||||
| No | 1.0 | |||
| Yes | 1.1 (0.6 to 1.9) | |||
| Level 2b (proximal variables) | Sexual behaviour | Lifetime no of partners | ||
| 1 | 1.0 | |||
| ≥2 | 2.8 (1.1 to 7.4) | |||
| Non-regular partner in last year | ||||
| No | 1.0 | 1.0 | ||
| Yes | 0.3 (0.0 to 1.6) | 4.7 (1.2 to 18.7) | ||
The multivariable logistic regression model was constructed using backward elimination. At each level, we show only the variables that remained in the final model.
↵* ORs are calculated by adjusting each variable to city, site and the variables that remained in the model from the previous and same levels.
↵† The referent group is the entire sex- and site-specific study population.