DATA MISSING: HOW TO SOLVE AND HOW TO ESCAPE THE PROBLEM
The article is devoted to the problem of missing data in clinical trials and clinical studies. The author considered three mechanisms of generating missing data in a collected sample. Each mechanism type is reviewed in details in terms of its effects on the sample representativeness and the magnitude of the result bias. The ways to reduce probability and amount of missing data are pointed in the phase of planning and at the stage of statistical data processing and inference.