Objective To evaluate the application of pooling HIV nucleic acid amplification testing (NAAT) among men who had sex with men (MSM) population, and to investigate suitable HIV screening strategy and the feasibility of calculation of HIV incidence using pooling NAAT among MSM population in China.Methods Four thousand eight hundred and fifty-six samples were collected from MSM population from April 2008 to September 2009 among with 4 156 were in Heilongjiang province and 700 were in Beijing in China. After standard testing with an HIV ELISA and WB confirmation testing, HIV antibody-negative samples were pooled and screened for HIV using NAAT.A three-stage pooling strategy was adopted.The HIV positive rate estimated by the four HIV screening strategies was calculated.In addition, 4 156 HIV positive specimens from Heilongjiang province were screened with the BED capture enzyme immunoassay (BED-CEIA).The HIV-1 incidences were estimated by BED-CEIA assay and pooling NAAT individually.ResultsOne hundred and forty-three of 4 856 subjects were HIV infected.130 were 3rd and 4th generation ELISA positive; 13 were antibody-negative but acutely HIV infected.According to the evaluation of four HIV screening strategies, routine HIV screening test together with pooling NAAT was more effective than other strategies for screening out window period generation ELISA+WB+pooling NAAT' were 2.68%(95% confidence interval CI=2.22%-3.14%), 2.82%(95%CI=2.35%-3.29%), 2.94%(95%CI=2.46%-3.42%) and 2.94%(95%CI=2.46%-3.42%), respectively.The differences were not significant (χ2=0.854 3, P=0.836 4).Of the 88 HIV positive samples from Heilongjiang province, 44 participants were tested as recent HIV infections by BED-CEIA assay. The estimated HIV-1 incidence was 2.36% (95%CI=1.63%-3.08%) and 2.92% (95%CI=1.01%-4.83%) based on BED-CEIA assay and pooling NAAT,respectively.Conclusions Pooling NAAT is a effective screening test in HIV negative population to detect window period infection among MSM population in China.