Sensitivity Analyses for Means or Proportions with Missing Outcome Data

We describe an approach to sensitivity analysis introduced by Robins et al (1999), for the setting where the outcome is missing for some observations. This flexible approach focuses on the relationship between the outcomes and missingness, where data can be missing completely at random, missing at random given observed data, or missing not at random. We provide examples from HIV that include the sensitivity of the estimation of a mean and proportion under different missingness mechanisms. The approach illustrated provides a method for examining how the results of epidemiologic studies might shift as a function of bias due to missing data.

Automated summary

We describe an approach to sensitivity analysis for missing data, focusing on the relationship between outcomes and missingness. The approach can handle completely random missingness, missingness dependent on observed data, and missingness not dependent on observed data. Examples from HIV are provided to illustrate the sensitivity of estimation under different missingness mechanisms. This approach allows for examination of the impact of missing data bias on the results of epidemiologic studies.