| Variable | End of study | ||
| Intervention n/N (%) | Control n/N (%) | P Valueb | |
| Number of partners in past 3 months | |||
| 0 partners | 36/204 (17.7) | 34/190 (17.9) | |
| 1 partner | 135/204 (66.2) | 119/190 (62.6) | |
| ≥2 partners | 33/204 (16.2) | 37/190 (19.5) | 0.61c |
| Unprotected sex at last sexd,e | |||
| Spouse | 5/55 (9.1) | 55/98 (56.1) | <0.001* |
| Regular partner | 3/59 (5.1) | 33/68 (48.5) | <0.001* |
| Casual partner | 10/65 (15.4) | 15/25 (60.0) | <0.001* |
| Commercial partner | 2/23 (8.7) | 9/19 (47.4) | 0.005* |
| Unprotected sex in past monthe,f | |||
| 0 acts | 150/171 (87.7) | 87/161 (54.0) | |
| 1 acts | 5/171 (2.9) | 11/161 (6.8) | |
| 2-5 acts | 11/171 (6.4) | 36/161 (22.4) | |
| 6 or more acts | 5/171 (2.9) | 27/161 (16.8) | <0.001*c |
| Unsafe sex in past monthe,f | |||
| 0 HIV-negative or unknown-status partner | 158/171 (92.4) | 114/161 (70.8) | |
| 1 HIV-negative or unknown-status partner | 3/171 (1.8) | 6/161 (3.7) | |
| 2-5 HIV-negative or unknown-status partners | 5/171 (3.5) | 22/161 (13.7) | |
| ≥6 HIV-negative or unknown-status partners | 4/171 (2.3) | 19/161 (11.8) | <0.001*c |
| Condom use self-efficacy | |||
| Low self-efficacy (score 15-34) | 2/204 (1.0) | 0/190 (0) | |
| Moderate self-efficacy (score 35-54) | 21/204 (10.3) | 90/190 (47.4) | |
| High self-efficacy (score 55-75) | 181/204 (88.7) | 100/190 (52.6) | <0.001*c |
| HIV status disclosed to partnerd,e | |||
| Spouse | 52/55 (94.5) | 89/98 (90.8) | 0.41 |
| Regular partner(s) | 5/59 (8.5) | 10/68 (14.7) | 0.28 |
| Casual partner(s) | 6/65 (9.2) | 9/25 (36.0) | 0.002* |
| Commercial partner(s) | 5/23 (21.7) | 6/19 (31.6) | 0.47 |
| Currently receiving antiretroviral treatment | 70/204 (34.3) | 24/189 (12.7) | <0.001* |
| Internalized stigma | |||
| Low (score 16-40) | 90/203 (44.3) | 107/190 (56.3) | |
| Moderate (score 41-52) | 113/203 (55.7) | 81/190 (42.6) | |
| High (score 53-64) | 0/203 (0.0) | 2/190 (1.1) | 0.034*c |
↵* P values < 0.05 were considered statistically significant.
↵a The following variables were used in propensity score matching: age, gender, education, religion, and employment.
↵b Pearson’s chi-square test unless otherwise indicated.
↵c Chi-square test for trend.
↵d Multiple-response question.
↵e Respondents were asked to report on all their sexual partners in the past 3 months with a maximum of 6 partners.
f Only among sexually active participants.