Even in well designed and conducted epidemiological studies, data will be missing. This may include missing observations of the exposure and under study, confounders, or the outcome. Possible mechanisms for data being missing will be discussed, as well as their potential impact in terms of bias. Focus will be on methods to handle missing data. Examples and exercises will come from various epidemiological studies, including diagnostic, prognostic, etiologic, and therapeutic studies.
By the end of the course, you should be able to:
- Explain different mechanisms giving rise to missing data
- Recognize missing data as a potential source of bias in epidemiologic research
- Describe key assumptions of methods to handle missing data
- Apply imputation methods to deal with missing data
Missing covariate data in medical research: To impute is better than to ignore. J Clin Epidemiol 2010:63;721-7.
Using the outcome for imputation of missing predictor values was preferred. J Clin Epidemiol 2006:59;1092-101.
Multiple imputation for multilevel data with continuous and binary variables. Stat Sci 2018:33;160-83.