To the Editor Drs Newgard and Lewis1 provided an overview of many important issues to consider when dealing with missing data in clinical research. In particular, they summarized the following limitations of traditional imputation methods: failure to account for uncertainty in imputed values, failure to make full use of observed values, possibilities for bias, and artificially low variance.
Stevens JR, Suyundikov A, Slattery ML. Accounting for Missing Data in Clinical Research. JAMA. 2016;315(5):517-518. doi:10.1001/jama.2015.16461