Table 4. Logistic regression analysis of factors affecting HIV seropositive status of female sex workers in a South African township, with both weighted and unweighted multivariate analysis displayed.
| Variable (n = 502) | Unweighted AOR (95% CI) |
P-value |
Weighted AOR (95% CI) |
P-value |
|---|---|---|---|---|
| Age (categorized along IQR) | ||||
| 18–25 | (ref) | - | (ref) | - |
| 26–30 | 2.2 (1.3–3.9) | 0.005 | 2.9 (1.3–6.3) | 0.008 |
| 31+ | 4.2 (2.6–6.8) | 0.0001 | 4.9 (2.6–9.3) | 0.0001 |
| Place of birth | ||||
| Gauteng | (ref) | - | (ref) | - |
| Other (includes cross-border migrants) | 1.7 (1.1–2.6) | 0.018 | 2.3 (1.3–4.0) | 0.003 |
| Education | ||||
| Incomplete secondary schooling | 2.7 (1.7–4.3) | 0.0001 | 2.8 (1.6–5.0) | 0.0001 |
| Secondary completed /some tertiary | (ref) | - | (ref) | - |
| Partner knows she sells sex | ||||
| Partner does not know | (ref) | - | (ref) | - |
| Partner knows | 1.7 (1.0–2.7) | 0.031 | 2.2 (1.3–3.8) | 0.005 |
| No. of clients in the past day | ||||
| 0–4 | (ref) | - | (ref) | - |
| 5–19 | 1.7 (1.2–2.6) | 0.005 | 1.9 (1.1–3.2) | 0.02 |