In summary, while the statistical methods discussed in this paper are well established and highly efficient, their application to HPV research has been limited(29). To our knowledge, this is the first formal discussion of these methods in the context of HPV natural history data. However, even if the greater use of these methods were to represent an advance, the analysis of the natural history of HPV will continue to be a challenge. It is hoped, therefore, that this paper will not only spur interest in current marginal and mixed effects models, but will additionally interest biostatisticians in the development of new and better methods for the analysis of HPV natural history data.
