This chapter explores an imputation procedure for multilevel count data within a sequential regressions Multiple Imputations (MI) s framework, using a Bayesian regression generalized linear mixed effects Poisson model to create the multiple imputations. Functions for over-dispersed, zero-inflated and two-level zero-inflated and over-dispersed count data are also available. Multiple imputations is a state-of-the-art procedure to address the missing data problem, as it allows the use of all available information in the dataset to predict missing information. The combined standard error incorporates both between and within imputation variability to address the extra estimation uncertainty due to missing data. The most commonly used frameworks that are being used today to create multiple imputations are the Bayesian joint modeling approach and sequential regressions MI. Future research should also address typical multilevel/growth modeling issues like hetero-scedasticity or auto-correlation in the within-group errors, i.e. future versions of the proposed imputation function should allow to model the issues as well.