The prevalence of HIV among MSM, China had been increasing rapidly in the past few years in Harbin, P. R. China. The surveillance data among MSM indicated that the prevalence of HIV had increased dramatically from 1% in 2006 to 4.4% in 2008 in Harbin 
. The prevalence of HIV was 9.5% in this survey, suggesting a marked increase in the HIV prevalence among MSM in Harbin in recent years.
In our survey, 86.8% of participants from MSM gave the 6 or over correct answers of AIDS knowledge and 58.3% of participants received intervention services. There was 57.6% of participants who never used or seldom used condoms with male sexual partners in the past 6 months, which was much higher than those in previous studies in Qingdao and Beijing 
. The high AIDS knowledge awareness, high intervention coverage and the low proportion of protective sexual behavior happened simultaneously and presented the separation of knowledge and behavior 
. It may be why a significant increase of HIV prevalence happened among MSM in recent years in Harbin. In addition, more sexual partners were common in this population, the average of male sexual partners was 3, the highest was 100 in this survey. More sexual partners would significantly increase the risk for HIV infection and accelerate the spread of HIV in MSM. Participants with 5 to 10 or ≥10 male sex partners in the past 6 months had higher risk for HIV infection than those with 0 to 1 male sex partners (OR 3.3, 95% CI 1.3 to 8.6, P
0.025; OR 5.6, 95%CI 1.9 to 17.1, P
In our survey, with the growth of age, the risk of HIV infection increased gradually, the participants aged ≥35 had significantly increased the risk of HIV infection than those in aged ≤20 (OR 4.7, 95%CI 1.1 to 20.8, P
0.026). At the same time, compared to the participants who had one year or less in duration of homosexual behavior, the MSM with over 10 years had a high risk for HIV infection (OR 5.4, 95% CI 1.9 to 13.6, P
0.001) (aOR 1.1, 95%CI 1.04 to 1.13, P
0.0002). Thus, our results indicated that along with the increase in duration of homosexual behavior, the risk of HIV infection also correspondingly increased in MSM.
Syphilis infection was also very popular in MSM population. According to our results of testing for syphilis, 14.3% of participants were infected with syphilis and 1.9% for co-infected with HIV and Syphilis in this survey. Syphilis infections significantly increased the risk of HIV infection. Some studies found that, after controlling for other risk factors, participants who had been infected with syphilis were more likely to be infected with HIV 
. Thus, the strength of syphilis diagnosis and treatment should be effectively increased in MSM population.
Participants whose marital status was cohabitation were more likely to be infected with HIV, the risk of HIV infection was 6.9 times higher than that married (OR
6.9, 95%CI 1.7 to 28.1, P
0.011). We could speculate that participants whose marital status was cohabitation had not received the bondage of marriage and they were not very single-minded to his partner, and therefore they maybe had more risk behavior and more sexual partners than those married.
Commercial sexual behavior may be also an important risk factor. 5.8% of participants had commercial sexual behavior in this survey. The result of multivariate analysis showed that participants were more likely to be infected with HIV than those without commercial sexual behavior (aOR 4.0, 95%CI 1.2–13.7, P
0.025). Because of making money in the nature of commercial sexual behavior, participants would provide sexual services to more commercial partners, and provid sexual service to not only man customers but also man. Thus the MSW (male sex workers, MSW) played a ”bridge„ and ”amplifier„ role in HIV/AIDS transmission between MSM and the general population 
. Due to this kind of crowd were hidden and floated, it was hard to find them. Implementation of intervention in this population was more difficult than those in without commercial sexual behavior. Therefore, commercial sexual behavior may be also an important risk factor.
A small proportion (30.9%) of this survey population had been engaged in heterosexual sex, which was less than other research results. A previous survey from Shenzhen (China) showed that 68.5% of MSM had sex with the female sex partners in 2004 
. Another survey from Petersburg (Russia) showed that 79% of MSM had sex with women 
. In this survey, 16% of MSM was married, these MSM had sex with not only men but also their spouses. Due to both social and secular pressure, marriage made it possible to maintain the relationships with men. The high risk of behaviors among this population imply that MSM might be bridge population for HIV transmission 
. In the AIDS/HIV intervention, it was a noticeable problem how intervention was done in bisexual MSM.
It was worth being advertent that safety of sex activities in the past 6 months was a protective factor for HIV infection in this survey. Our results revealed that participants who had a protected sex in the past 6 months was lower at a risk of HIV infection than those in an unprotected sex (OR 0.4, 95%CI 0.2 to 0.8, P
0.01) (aOR 0.4, 95%CI 0.17 to 0.99, P
0.044). By now, it is widely believed that using condoms in a correct way and the most effective way of preventing AIDS in the world, and it was also proved in this survey. In our study, 57.6% of MSM population never used or seldom used condoms, the 30.9% of MSM having sex with females and 62.0% of those having unprotected sex in the past 6 months. It accelerated HIV transmitting from MSM to general population in Harbin, P. R. China. Thus, the propaganda about correct condom-using in MSM should be strengthened and increased in order to prevent HIV infection themselves in the future. Considering the risk factors of HIV infection among MSM population, comprehensive intervention are needed urgently, including not only education and promotion of condom-use, but also reducing the number of sexual partners and creating a single, committed relationships.
This survey is subject to several limitations. The data was examined retrospectively from a cross-sectional study and cannot prove exact causality. The snowball sampling was used in this survey, some selected biases may have influenced the results of this study. Owing to the subjects of survey came from a convenient sample of MSM who live in Harbin, P. R. China, the results may not be generalized to the greater population of MSM. The participates were approached and invited to in this study by outreach staff from non-governmental organizations which specifically serve the MSM community. Some MSM were unwilling to expose their identities to the publicity, which was called recessive crowd. Thus, they did not participate in our survey. As it involves personal privacy in the questionnaire, the participants may not have answered accurately. There would be some recall bias on some problems related to behaviors in the last 6 months or in the past year.
The prevalence of HIV among MSM was rapidly increased in the past few years in Harbin, P. R. China. Our findings showed the high AIDS knowledge awareness, high intervention coverage and the low proportion of protective sexual behavior happened simultaneously in MSM community of Harbin. They presented the separation of knowledge and behavior. Considering the ”separation of knowledge and behavior„ in MSM in Harbin, we might further study to understand the reasons of such a separation, and take corresponding intervention measures. Cohabitation, duration of homosexual behavior and homosexual partners in past 6 months were important risk factors which associated with the HIV infection and commercial sexual behavior may be also an important risk factor. Participants who had protected sex in the past 6 months were at a lower risk of the HIV infection. Because it is widely believed that using condoms in a correct way is the most effective way of preventing AIDS in the world, we should strengthen propaganda and expand the usage of the condom by multi-sectoral collaboration in MSM in Harbin. Syphilis infection was also very popular in this population, and significantly increased the risk for being infected with HIV. Targeted, tailored, and comprehensive interventions are needed urgently to prevent HIV infection amongst MSM.