Skip to main content
. 2017 Apr 11;46(5):1593–1606. doi: 10.1093/ije/dyx045

Table 3.

Multivariable model of predictors of time to HIV transmission among Zambian women and men in HIV discordant relationships

Primary modela,b M+F- couples
F+M- couples
Exposures of interest aHR 95% CI P-value aHR 95% CI P -value
Genital inflammation of woman
 STI 1.36 0.76 2.43 0.30 1.01 0.50 2.05 0.97
 Non-STI 1.55 1.01 2.38 0.04 1.92 1.15 3.22 0.01
 No ref       ref      
BIA of woman                
 Yes 2.33 1.43 3.81 < 0.01 2.76 1.69 4.51 < 0.001
 No ref       ref      
Genital ulcer of woman                
 Yes 2.08 1.27 3.41 < 0.01        
 No ref              
Genital inflammation of man                
 STI 3.33 1.79 6.17 < 0.01 e    
 Non-STI 1.18 0.43 3.21 0.75 1.19 0.51 2.77 0.69
 No ref       ref      
BIA of man                
 Yes 3.35 2.24 5.03 < 0.001 4.11 2.52 6.72 < 0.001
 No ref       ref      
Genital ulcer of man                
 Yes 1.49 1.05 2.11 0.03 3.44 2.20 5.38 < 0.001
 No ref       ref      
Baseline and time-varying variables
Age of woman (per year increase) 0.98 0.95 1.00 0.04 0.99 0.96 1.02 0.42
Woman reads Nyanja                
 Yes, easily ref              
 With difficulty/not at all 1.22 0.82 1.80 0.32        
Circumcised manc                
  Yes         ref      
  No         2.10 1.12 3.95 0.02
VL of positive partner (per log10 copies/ml increase) 1.68 1.35 2.08 < 0.001 1.80 1.44 2.23 < 0.001
Any self-reported unprotected sex with study partner since last visit        
  Yes 1.38 1.01 1.88 0.04 1.91 1.33 2.75 < 0.01
  No ref       ref      
Sperm present on vaginal swab wet mount                
  Yes 1.41 0.88 2.26 0.15 1.73 1.03 2.91 0.04
  No ref       ref      
Current pregnancy                
  Yes         1.35 0.79 2.31 0.27
  No         ref      
Interval since enrolment                
  0–3 months 4.30 2.75 6.72 < 0.001 5.07 2.93 8.78 < 0.001
 < 3 months ref       ref      
Pre-2002 model 1d
Circumcision and smegma status
 Circumcised ref       ref      
 Uncircumcised and no smegma 1.68 0.93 3.04 0.08 4.50 2.36 8.59 < 0.001
 Uncircumcised and smegma 3.16 1.54 6.46 < 0.01 8.59 3.72 19.86 < 0.001
Pre-2002 model 2d    
Circumcision status and smegma by man's BIA        
 Circumcised        
     BIA of man: Yes 0.36 0.06 1.99 0.24
     BIA of man: No         ref      
  Uncircumcised and no smegma                
     BIA of man: Yes         4.95 2.43 10.10 < 0.001
     BIA of man: No         0.85 0.31 2.34 0.75
  Uncircumcised and smegma              
     BIA of man: Yes         13.03 4.97 34.17 < 0.0001
     BIA of man: No           e    

Foreskin smegma recorded from 1994-2002. P-values are two-tailed from Cox models. Adjusted point estimate for genital ulcer of woman in F+M- primary model (not included in final models): aHR = 0.94, 95% CI:0.55-1.62 (P = 0.84).

a

‘Generalized’ R-squared = 0.63 for M+F- couples (primary model).

b

‘Generalized’ R-squared = 0.80 for F+M- couples (primary model).

c

Circumcised at baseline or ever during follow-up.

d

Controlling for the primary model variables.

e

Small sample size; a measure of association could not be estimated.