HIV incidence rate ratio and BED practical power when using log-linear Poisson regression to analyze simulated samples of young people.

NC = not calculable NA = Not applicable CI = confidence interval.a bWeights k1 and k2 introduced in the Poisson log-linear model for each group a among HIV-positive tested recent seroconverters and b among HIV-positive tested long-term seroconverters.cFactor by which the incidence rate is multiplied in the intervention group in comparison with the control group.dThe BED theoretical power was obtained from Table 1 (0.62 for simulations 1 thru 7, 0.91 for simulations 8 thru 10, and 0.53 for simulation 11).