Table 3.
Associations With Seroconversion Among Incident α9 Human Papillomavirus Infections in a Male Cohort (n = 176)
| Characteristics | No. among seroconversions (n = 10)/ No. among no seroconversion (n = 166) | Univariate HRa (95% CI) | Multivariate HRa (95% CI) |
| Ageb | Continuous | 0.7 (0.3, 1.6) | |
| Race | |||
| White | 9/129 | 1 | |
| Nonwhite | 1/37 | 0.4 (0.04, 3.2) | |
| Circumcised | |||
| No | 2/50 | 1 | |
| Yes | 8/116 | 2.5 (0.3, 19.3) | |
| Ever smoked at enrollment | |||
| No | 7/142 | 1 | 1 |
| Yes | 3/24 | 3.1 (0.8, 11.6) | 3.4 (1.0, 11.5) |
| Time since first vaginal intercourse at incident infection, yc | |||
| 0 to 2 | 2/56 | 1 | |
| 3 to 5 | 7/89 | 2.0 (0.4, 10.2) | |
| 6 to 9 | 1/21 | 5.5 (0.6, 53.7) | |
| Lifetime number of vaginal sex partners at incident infectiond | |||
| 1 to 2 | 2/34 | 1 | |
| 3 to 9 | 7/95 | 1.0 (0.2, 4.9) | |
| 10 or greater | 1/37 | 0.7 (0.1, 6.1) | |
| Number of new partners in 4 mo before incident infectionde | |||
| 0 | 3/95 | 1 | |
| 1 or greater | 7/68 | 2.5 (0.7, 9.0) | |
| Number of times had vaginal intercourse in 4 mo before incident infectionde | |||
| 0 to 3 | 1/34 | 1 | |
| 4 to 19 | 5/69 | 2.0 (0.2, 20.6) | |
| 20 or greater | 4/60 | 2.0 (0.2, 20.3) | |
| Condom use in 4 mo before incident infection among those who had vaginal intercoursee | |||
| None | 1/53 | 1 | |
| Any | 9/86 | 3.5 (0.4, 31.5) | |
| Incident HPV infection site(s) | |||
| Glans/urine only f | 0/38 | 1 | 1 |
| Shaft/scrotum only f | 5/73 | ||
| Glans/urine & shaft/scrotum | 5/34 | 4.0 (1.1, 14.1) | 4.4 (1.2, 16.4) |
| Number of visits incident HPV type detectedg | |||
| 1 | 7/109 | 1 | |
| 2 or greater | 3/57 | 1.2 (0.2, 7.5) | |
NOTE. HR, Hazards ratio; CI, confidence interval; HPV, human papillomavirus.
Hazard ratios and 95% CIs calculated using Cox proportional hazards models stratified according to HPV type within the model, with correlated data accounted for using generalized estimating equations
Actual age at incident infection, rounded to the tenth (range, 18.8–23.2 y)
P values for test for trend by category and continuously were .19 and .72 in the univariate model, respectively
P values for test for trend by category and continuously were greater than .30 in the univariate model
Missing information for 3 infections that did not seroconvert
Glans/urine only and shaft/scrotum only combined for the analysis
Time-varying covariate