Applied Missing Data Analysis in the Health Sciences

Chapter 3Design Considerations in the Presence of Missing Data

It should be pointed out that researchers themselves should be blamed in part for the high prevalence of missing data in biomedical, behavioral, and social research. Many missing data problems arise from a poor study design and lack of careful planning. They can be largely avoided if enough attention is given at the beginning of the study. As Benjamin Franklin put it, “an ounce of prevention is worth a pound of cure.” So a small amount of planning ahead can often lead to greatly reduced bias and improved efficiency, sometimes can even salvage an otherwise wasted effort. In this chapter, we outline some design and conduct strategies to avoid or reduce missing data in biomedical research studies. Most of the advice is based on a recent National Research Council report (National Research Council, 2010) and a special report in the New England Journal of Medicine (Little et al., 2012b). More technical information can be found in Little et al. (2012a).

3.1 Design Factors Related to Missing Data

The best advice regarding missing data is probably the one given by R.A. Fisher, “the best solution to handle missing data is to have none.” This may sound intuitive; however, it is very difficult to achieve in practical settings. In fact, for most clinical studies, more often than not there will be missing data. Some researchers might put little effort in study designs and data collection to prevent missing data, since the missing ...