It depends what you mean by "handle". If you mean, Can you include subjects who have incomplete data (i.e., dataset is unbalanced), then the answer is that it's not a problem. If you mean, How can we adjust for the fact that data are missing in a not-at-random way, then the answer is that there are no well established ways to do this.
-------------------------------------------
Jeremy Weedon
Asst. Clinical Professor (biostatistician)
SUNY/Downstate Medical Center
-------------------------------------------