Throughout the last decades, multiple imputation (MI) has become one of the standard state-of-the-art techniques to handle missing data. MI procedures are nowadays implemented in most data analysis packages like for example SPSS, SAS, STATA, or Mplus. Unfortunately however, currently available commercial statistical software is still highly limited regarding the imputation of incomplete count data, and especially multilevel count data. The distribution of count data is typically not normal and data analysis requires special statistical models like Poisson or negative binomial (NB) models. Analogous, missing data imputation ought to be based on these models, as well. We present a flexible and easy to use solution to create multiple imputations of incomplete overdispersed multilevel count data, based on a generalized linear mixed effects NB model. The procedure works as an add-on for the popular and powerful multiple imputation package mice (van Buuren & Groothuis-Oudshoorn, 2011) for the R language and environment for statistical computing. It ought to be used, when the equidispersion assumption of a multilevel Poisson imputation model (Kleinke & Reinecke, in press) is violated.