The threat to validity from missing data is potentially greater for observational studies (including studies utilizing electronic health records and administrative datasets) due to the fact that data can be missing for exposures, confounders, and outcomes as compared to only outcomes for well-conducted randomized clinical trials. Statistical techniques that account for the missing data rely on the validity of assumptions concerning the factors leading to the missing values and how they relate to the study outcomes. Missing data are traditionally categorized as missing completely at random, missing at random (MAR) and missing not at random. This presentation will cover the approaches for evaluating whether the MAR assumption may be violated for a given set of data and the model-based analytical approach of mixed-effect model for repeated measures and inverse probability weighting. The aforementioned topics will be explained with examples. The underlying assumptions as well as advantages and disadvantages of the model-based analytical approach will be discussed.