This was a prospective study in Manitobahu, Canada. A total of 3541 women ≥ 15 years of age received quadrivalent HPV vaccination, in a 1:3 ratio with age-matched controls of 9594 unvaccinated women. Cox regression models were used to estimate HRs for three outcomes: ASCUS, LSIL, and HSIL. As a result, among girls aged 15-17 years, the effectiveness of vaccination was 35% (95% CI -19% – 65%), 21% (-10% – 43%), and -1% (-44% – 29%) for HSIL, LSIL, and ASCUS, respectively. For women who had ≥ 1 Pap smear after recruitment, the corresponding validity estimates were higher, 46% (0% – 71%), 35% (10% – 54%) and 23% (-8% – 45%), respectively. For women ≥ 18 years of age with no previous history of abnormal cytology, quadrivalent vaccine reduced the risk of HSIL by 23% (-17% – 48%), but there was no evidence that the vaccine protected those women with a previous history of abnormal cytology (-8% [59% to 27%]). Clearly, the majority of vaccinated women may not have a protective effect against HSIL or lower grade lesions, especially if the age at vaccination is ≥ 18 years or if there is a history of abnormal cytology. These findings support the importance of this scenario that vaccination must precede significant exposure to HPV and emphasize that cervical cancer screening procedures should cover all sexually active women, even if they have already received vaccination.