Skip to main content

Table 2 Change in sexual behaviour between Sero4 and Sero5 associated with VCT use at Sero4, HIV-negative individuals

From: The impact of voluntary counselling and testing services on sexual behaviour change and HIV incidence: observations from a cohort study in rural Tanzania

  N with outcome (%)   
  No VCT Had VCT Crude risk ratio (95% CI) Adjusted risk ratio (95% CI)
Number of sex partners in last year    
  3173 % 301 %   
No change 2060 64.9 169 56.1 1  
Increase 306 9.6 25 8.3 1 (0.64-1.54) 0.82 (0.52-1.30)
Decrease 807 25.4 107 35.5 1.62 (1.25-2.09)*** 1.42 (1.07-1.88)**
Acquired or lost a spouse α     
  3136 % 297 %   
No change 2701 86.1 233 78.5 1 1
Acquired 255 8.1 43 14.5 1.95 (1.38-2.77)*** 1.88 (1.29-2.74)***
Lost 180 5.7 21 7.1 1.35 (0.84-2.17) 1.56 (0.95-2.55)*
Acquired or lost a regular non-cohabiting partner   
  3168 % 301 %   
No change 2601 82.1 226 75.1 1 1
Acquired 251 7.9 27 9.0 1.24 (0.81-1.88) 0.88 (0.55-1.43)
Lost 316 10.0 48 15.9 1.75 (1.25-2.44)*** 1.57 (1.10-2.25)**
Acquired or lost a high risk partner    
  3156 % 299 %   
No change 2879 91.2 267 89.3 1 1
Acquired 193 6.1 19 6.4 1.06 (0.65-1.73) 0.97 (0.58-1.62)
Lost 84 2.7 13 4.3 1.67 (0.92-3.03) 1.41 (0.75-2.65)
Condom use at last sex with spouse #    
  2000 % 190 %   
No change 1951 97.6 184 96.8 1 1
Stopped using 31 1.6 2 1.1 0.68 (0.16-2.88) 0.45 (0.10-1.93)
Started using 18 0.9 4 2.1 2.36 (0.79-7.04) 1.79 (0.56-5.67)
Condom use at last sex with regular non-cohabiting partner #  
  179 % 31 %   
No change 147 82.1 17 54.8 1 1
Stopped using 12 6.7 8 25.8 5.76 (2.07-16.08)*** 4.88 (1.39-17.16)**
Started using 20 11.2 6 19.4 2.59 (0.92-7.35) 2.21 (0.60-8.14)
Condom use at last sex with high risk partner #   
  12 % 5 %   
No change 8 66.7 3 60.0 1 -
Stopped using 1 8.3 1 20.0 2.67 (0.12-57.62) -
Started using 3 25.0 1 20.0 0.89 (0.06-12.25) -
  1. All outcomes adjusted for age, sex, marital status, area of residence, level of education and previous VCT use as reported at the later sero-survey round.
  2. *p ≤ 0.1, **p ≤ 0.05, ***p ≤ 0.001.
  3. α‘Acquired or lost spouse’ was not adjusted for marital status (as this was the outcome).
  4. #Analyses conduced among those with relevant partner type at both rounds only.