Clinical Study

High-Risk Human Papillomavirus Is Associated with HIV Acquisition among South African Female Sex Workers

Table 1

Distribution of detected HIV cases and risk factors of HIV acquisition.

FactorDistributionNumber of HIV infectionsMean survival (days)Univariate HR
(95% CI)
Multivariate aHR*
(95% CI)

Number of high-risk HPV
 029.5% (26/88)51148
 143.2% (38/88)911101.5 (1.1 to 2.1) 1.7 (1.01 to 2.7)
 214.8% (13/88)5934
 3-412.5% (11/88)6775
Number of low-risk HPV
 039.8% (35/88)71134
 135.2% (31/88)119921.1 (0.12 to 0.69) 0.77 (0.49 to 1.2)
 215.9% (14/88)5932
 39.1 (8/88)21136
Age (years)
 Less than 2446.6% (41/88)1893311
 24 or more53.4% (47/88)711920.29 (0.12 to 0.69) 0.28 (0.079 to 0.97)
Intervention
 Placebo46.6% (41/88)9113211
 Nonoxynol-953.4% (47/88)1610131.8 (0.81 to 4.2) 2.0 (0.75 to 5.1)
Education
 Less than 6 years40.9% (36/88)6118411
 6 years or more59.1% (52/88)199902.7 (1.1 to 6.8) 5.5 (1.6 to 19.6)
Condom use
 Never or  sometimes80.7% (71/88)22104411
 Often19.3% (17/88)311750.57 (0.17 to 1.9) 0.29 (0.05 to 1.6)
Anal sex
 No56.8% (50/88)18102911
 Yes43.2% (38/88)711370.53 (0.22 to 1.3) 0.35 (0.12 to 1.0)
Length of sex work (months)
 Less than 2450.0% (44/88)13104511
 24 or more50.0% (44/88)1210910.89 (0.41 to 1.9) 1.5 (0.41 to 5.6)
Number of clients per week
 Less than 2046.6% (41/88)13105611
 20 or more53.4% (47/88)1210760.78 (0.36 to 1.7) 1.4 (0.48 to 3.9)
Trichomonas vaginalis
 Negative64.2% (52/81)17103011
 Positive35.8% (29/81)611090.52 (0.21 to 1.3) 1.4 (0.41 to 4.6)
Candida albicans
 Negative63.0% (51/81)16103811
 Positive37.0% (30/81)711290.67 (0.28 to 1.6) 0.55 (0.18 to 1.7)
Neisseria gonorrhoeae
  Negative91.3% (73/80)20107411
 Positive8.8% (7/80)310651.33 (0.40 to 4.5) 2.3 (0.53 to 9.6)
Chlamydiatrachomatis
 Negative86.5% (64/74)20103911
 Positive13.5% (10/74)711200.77 (0.23 to 2.6) 0.25 (0.06 to 1.1)

Syphilis
  Negative73.2% (60/82)20104211
 Positive26.8% (22/82)411380.50 (0.17 to 1.5) 0.66 (0.17 to 2.6)

(a) HR: (adjusted) hazard ratio.
HPV: human papillomavirus.
The following variables were included in the multivariate analysis: number of high-risk HPV, number of low-risk HPV, age, intervention group, education, condom use, anal sex, length of sex work, number of clients per week, infection with Trichomonas vaginalis, infection with Candida albicans, infection with Neisseria gonorrhoea, infection with Chlamydia trachomatis, and infection with syphilis.
Increase of hazard ratio of HIV acquisition for an increase of one unit of the number of HPV. The P value corresponds to a linear trend.