BACKGROUND: Infection with Human Papillomavirus (HPV) is a necessary event in the multi-step process of cervical carcinogenesis. Little is known about the natural history of HPV infection among unscreened young adults. As prophylactic vaccines are being developed to prevent specifically HPV 16 and 18 infections, shifts in prevalence in the post vaccine era may be expected. This study provides a unique opportunity to gather baseline data before changes by nationwide vaccination occur. METHODS AND PRINCIPAL FINDINGS: This cross-sectional study is part of a large prospective epidemiologic study performed among 2065 unscreened women aged 18 to 29 years. Women returned a self-collected cervico-vaginal specimen and filled out a questionnaire. All HPV DNA-positive samples (by SPF(10) DEIA) were genotyped using the INNO-LiPA HPV genotyping assay. HPV point prevalence in this sample was 19%. Low and high risk HPV prevalence was 9.1% and 11.8%, respectively. A single HPV-type was detected in 14.9% of all women, while multiple types were found in 4.1%. HPV-types 16 (2.8%) and 18 (1.4%) were found concomitantly in only 3 women (0.1%). There was an increase in HPV prevalence till 22 years. Multivariate analysis showed that number of lifetime sexual partners was the most powerful predictor of HPV positivity, followed by type of relationship, frequency of sexual contact, age, and number of sexual partners over the past 6 months. CONCLUSIONS AND SIGNIFICANCE: This study shows that factors independently associated with HPV prevalence are mainly related to sexual behaviour. Combination of these results with the relative low prevalence of HPV 16 and/or 18 may be promising for expanding the future target group for catch up vaccination. Furthermore, these results provide a basis for research on possible future shifts in HPV genotype prevalence, and enable a better estimate of the effect of HPV 16-18 vaccination on cervical cancer incidence.