Table 5.
Association of concurrency in past 12 months with HIV seroprevalence, unadjusted and after adjusting for confounders1
| Odds ratio (95% CI) |
p-value | |
|---|---|---|
|
OUTCOME = HIV SEROPREVALENCE EXPOSURE = ANY CONCURRENCY IN PAST 12 MONTHS |
||
| Among men reporting sex in past 12 months (N = 1655) | ||
| Model 1 | ||
| Any concurrency in the past 12 months | 1.33 (0.85-2.08) | P = 0.22 |
| Model 2 | ||
| Any concurrency in the past 12 months | 1.13 (0.67-1.89) | P = 0.64 |
| Adjusted for: age group, marital status and lifetime partners2 | ||
| Model 3 | ||
| Any concurrency in the past 12 months | 1.04 (0.62-1.77) | P = 0.88 |
| Adjusted for: age group, marital status, lifetime partners2 and casual partner in past 12 months | ||
|
OUTCOME = HIV SEROPREVALENCE EXPOSURE = CONCURRENCY WITH POLYGYNOUS PARTNERS ONLY |
||
| Among men reporting sex in past 12 months (N = 1655) | ||
| Model 1 | ||
| Concurrency with polygynous partner in past 12 months | 0.50 (0.18-1.39) | P = 0.14 |
| Model 2 | ||
| Concurrency with polygynous partner in past 12 months | 0.40 (0.14-1.11) | P = 0.07 |
| Adjusted for lifetime partners2 in past 12 months | ||
|
OUTCOME = HIV SEROPREVALENCE EXPOSURE = CONCURRENCY EXCLUDING POLYGYNY |
||
| Among men reporting sex in past 12 months (N = 1655) | ||
| Model 1 | ||
| Concurrency with non-spousal partner in past 12 months | 1.78 (1.11-2.86) | P = 0.02 |
| Model 2 | ||
| Concurrency with non-spousal partner in past 12 months | 1.36 (0.82-2.25) | P = 0.25 |
| Adjusted for lifetime partners2 and casual partner in past 12 months | ||
|
OUTCOME = HIV SEROPREVALENCE IN SPOUSE EXPOSURE = ANY CONCURRENCY IN PAST 12 MONTHS |
||
| Among wives of men reporting sex in past 12 months (N = 1134) | ||
| Model 1 | ||
| Any concurrency by husband in the past 12 months | 1.17 (0.71-1.92) | 0.54 |
| Model 23 | ||
| Any concurrency by husband in the past 12 months | 0.97 (0.59-1.59) | 0.90 |
| Adjusted for husband's lifetime partners2 | ||
1The following potential confounders were considered: age, SES, marital status, age at first sex, lifetime partners, casual partner in past year and condom use at last sex. Those variables which changed the OR for the association with concurrency by > 10% were retained.
2Lifetime partners above the median for males in the cohort (see Table 1)
3Adjusted model considered all potential confounders listed in footnote 1, and reported behaviour of wife (lifetime partners, casual partner in past year, and condom use at last sex).