WIHS women have been followed on a semiannual basis. Mid-interval (i.e., the midpoint calendar data between two consecutive visits) was used to estimate the time of each incident event. Women were censored for missed visits, and we assumed missing at random. Since our statistical models adjusted for CD4+ count and HIV viral load as well as other risk factors the analyses addressed the main factors that might cause informative censoring in the WIHS cohort. Incident detection of HPV incident detection of type-specific HPV was defined as a positive HPV DNA test result for a specific type in a participant who was negative for that HPV type in all earlier visits (prevalent infections are excluded).
