Vaccine and Infectious Disease Division

Shedding the bias in HSV-2 studies

Clinical studies of HSV-2 often classify populations of HSV-2 positive individuals by using the term “shedder” for those who release detectable levels of the virus from their genital mucosa  at least once over several samples.  However this classification may be problematic as HSV-2 shedding is a variable event that may not be detected during research studies, potentially leading to the clinical misclassification of shedding individuals as “non-shedders”.   To determine how the designation of “shedder” may impact study interpretation, VIDI joint assistant member Dr. Amalia Magaret, VIDI affiliate investigator Dr. Anna Wald, and a fellow colleague evaluated two different methods used to describe the shedding of HSV from the genital mucosa.  Using data on HSV-2 seropositive persons and mathematical equations, they showed that comparisons using the classification “shedders” do not appear to be appropriate in any context, as the number of participants described as shedders was dependent on the length of the study.  Instead, they suggest using overall shedding rates and making group comparisons using Poisson regression (a statistical analysis that assumed the number of days on which HSV is detected follows the Poisson distribution) to avoid the potential bias associated with summarizing detection over repeated samples.

Use of the designation "shedder" in mucosal detection of herpes simplex virus DNA involving repeated sampling. Magaret AS, Johnston C, Wald A. Sex Transm Infect. 2009 Aug;85(4):270-5.