Table 9: Studies of the association between circumcision status and the prevalence of syphilis.

StudyIntact +ve/−veCircumcised +ve/−veOdds ratio95% confidence intervalExact odds ratioExact confidence interval

Bailey 7/1696/1380.950.31–2.900.95280.2673–3.5173
Buvé: Kismu17/3930/15613.920.83–232.909.48701.1611–[55.8609]
Buvé: Ndola57/4807/480.810.35–1.880.81460.3442–2.2355
Bwayo RPR*14/16242/7291.500.80–2.811.49930.7380–2.8812
Bwayo TPHA*38/75106/3511.681.07–2.621.67611.0403–2.6740
Cook 20/52013/22236.693.27–13.706.57053.0874–14.4743
Dave 10/48233/9790.680.19–2.460.67670.1738–3.8332
Diseker10/21218/6221.620.74–3.691.62900.6608–3.7908
Donovan 1/1142/1830.800.07–8.950.80320.0135–15.5915
Gray et al. [44]*482/411793/8151.030.81–1.301.02600.8091–1.3116
Gray et al. [163]*446/391796/7090.840.67–1.060.84100.6629–1.0750
Hand420/1090108/2190.780.60–1.010.78150.6007–1.0204
Laumann 12/110613/13121.100.50–2.411.09500.4546–2.6147
Lavreys 11/8448/6031.650.82–3.291.64380.7400–3.3705
Lloyd 81/30025/931.000.61–1.661.00440.5937–1.7413
Mor192/13838384/362901.311.10–1.561.31121.0956–1.5651
Newell 125/122945/5971.350.95–1.921.34910.9379–1.9695
Otieno-NyunyaNANA2.21.3–3.72.21.3–3.7
Parker 9/5813/7263.751.01–13.913.74520.9292–216041
Reynolds 128/16399/1511.310.65–2.631.31010.6510–2.9907
Rodriguez-Diaz68/37837/1770.860.56–1.330.86080.5448–1.3756
Schneider107/504925/9860.840.54–1.300.83590.5335–1.3562
Schrek white 10/1421/352.460.31–19.902.45570.3293–109.9779
Schrek black 19/1046/391.190.44–3.191.18630.4155–3.9024
Seed 24/57010/2330.980.46–2.080.98110.4438–2.3363
Talukdar25/3398/940.870.38–1.980.86680.3640–2.2979
Todd 187/35439/1371.861.25–2.761.85411.2310–2.8387
Uganda149/664337/24521.491.03–2.141.48641.0276–2.1990
Urassa 775/3282155/7721.180.97–1.421.17610.9707–1.4311
Vaz74/74829/4331.480.95–2.311.47670.9321–2.3934
Wilson 90/91010/2942.911.49–5.662.90591.4838–6.3494
Random effects summary effect:1.30361.1103–1.5306

Heterogeneity chi-square ( ) was 67.70 ( ).
The RPR data from the Bwayo study and the 2004 Radkai data were used in calculating the summary odds ratios.
Brackets have infinite confidence interval replaced with value equidistant from the odds ratio as the measured confidence interval.