In this report of a large and representative sample, we report anti-HBc prevalences and sexual factors associated with anti-HBc positivity. These factors included geographic region and younger age at sexual debut. We also demonstrated that consistent condom use is associated with lower anti-HBc prevalences at the population level.
In a previous study of 1,048 pregnant women between 14 and 19 years in Lima, anti-core prevalence was reported to be 3.5% 
. Our study, however, reports anti-HBc prevalence among males and females between 18 and 29 years of age from the general population in 28 cities. Similar to other reports 
, anti-HBc prevalence was higher in the jungle than in the highlands and coastal regions. Overall, the reported prevalence of HBV infection is highest in the western Amazon area including Brazil and adjacent regions in Colombia, Peru, and Venezuela 
. The reason for this finding is not completely clear; however, participants from jungle cities more often reported sexual debut <16 years of age (42.3% for jungle, 19.9% for highlands and 26.2% for coastal cities) and 5 or more lifetime sex partners (25.6% for jungle, 15.8% for highlands and 16.9% for coastal cities).
Lower age at first sexual intercourse was associated with anti-HBc positivity. We found that the odds of anti-HBc positivity increased with younger age at sexual debut, owing to increasing exposure to cumulative risk (10% of protective effect per year of delay of first sexual intercourse according to the multivariable model). Only 4.6% of the participants with anti-HBc positivity also had HBsAg positive results, in accordance with reports of the percentage of HBV infected people having chronic disease 
. Interestingly, increasing anti-HBc prevalence was found when the analysis was performed according to population at sexual risk, being higher among MSMs than women receiving money or goods for sex or general population. This result is in line with STDs trends in Peru 
In the final multivariable model, variables found in previous studies to be associated with HBV infection such as educational level or lifetime number of sex partners were not independently associated with anti-HBc positivity 
. Among those who reported never having had sexual intercourse, prevalence of anti-HBc was 4.0%, suggesting that sexual activity is not the only mode of hepatitis B transmission in this population. Among those without a sexual history who were anti-HBc positive result, about 75% were between 18 and 21 years indicating early infection, either through vertical or early horizontal transmission.
Male condoms are an efficient and inexpensive way of reducing the risk of acquisition of STIs. As has been demonstrated previously, consistent and correct condom use protects against bacterial STIs 
, including gonorrhea 
, and chlamydial infection 
. Evidence also confirms that consistent condom use significantly reduces the risk of sexual transmission of HIV 
, herpes simplex virus 
and human papillomavirus infections 
. However, data on the effectiveness of condoms in reducing HBV infection have been limited.
In this cross sectional study, we adjusted our crude estimates for variables found to be associated with both condom use and HBV infection: gender, geographic region, and age at sexual debut. In addition, the risk estimate for consistent condom use was significant after adding to the model other potential confounders previously described in the literature such as lifetime number of sex partners and education level 
. In addition, we were able to estimate the attributable risk percent; but, the OR used for this calculation may be confounded by unmeasured factors, which might affect the estimation of the AR%. Two previous studies have reported that consistent condom use can reduce acquisition of HBV infection among FSW. The first study of acquisition of HBV infection among FSW in India used HBsAg as a serological marker. The intervention group underwent a 6-month program of educational videos, small group discussions, educational materials, received free condoms, and it found incidences of 0.04 compared to 0.12 per person-year of follow-up among intervention and control women, respectively 
. The second reported that long-term consistent condom use had protective effect for anti-HBc positivity among FSW participating in a control program in Peru 
. However, two additional studies carried out in other populations found conflicting results. The first, a population-based study in India, used a smaller sample size and found protective effect of condoms only among the female subgroup 
. The second, from Brazil, found that non-use of condoms was highly frequent (57%) among truck drivers but was not associated with decreased hepatitis B prevalences 
Our study has several limitations. First, the questionnaire was designed in the context of a STD intervention and did not include information about non-sexual risk factors for HBV. Although prevalence of intravenous drug use among Peruvians has been considered negligible 
, we could not evaluate other important factors such as tattooing, body piercing, blood transfusions, perinatal transmission, vaccination and household contacts. Second, our survey was limited to a population of 18 to 29 years in urban areas. Information regarding other age groups and rural areas are also essential to understand transmission within the overall population. Third, because condom use was based on self-reports, there is the possibility of incorrect recall or social desirability bias. Our focus on younger adults was in part to reduce recall bias. Fourth, only 40% of the total sample responded questions regarding condom use. We compared demographics and sex-risk variables among those who reported data about condom and those who did not: only distributions of marital status and lifetime number of sex partners were different between both groups (See Supporting Information S4
). Thus, inference of results to the general population might not be completely appropriate. Further studies are needed to corroborate our results. Finally, condom use data were based on the participant's last three partners during the last three months and sexual experiences during the past 3 months are not necessarily representative of all sexual experience, raising the possibility of misclassification. However, we believe that any misclassification in the condom use variable would be non-selective with regards to HBV status, and thus would result in attenuation of the true association; as a result, the protective effect of condom could be greater than the association found in this study. The strengths of this study include the random sampling methodologies, the large population-based sample, extensive and explicit data on recent sexual behaviors and sex practices, and the use of computer-assisted self-interview (CASI) techniques to obtain sensitive data.
In conclusion, anti-HBc prevalences are highest in jungle region and among men who have sex with men. There are several risk factors associated with anti-HBc positivity. Consistent condom use was associated with lower prevalences of anti-HBc. Findings from this study emphasize the need for primary prevention programs (vaccination) especially in the jungle population, and imply that condom use promotion could also be a potential strategy to prevent infection with HIV and other sexually transmitted pathogens, but also HBV infection.