Added the possibility of using visit dimension instead of observations facts for construction of right-censoring event.
It is useful if we consider that a fact can generate after a subject leaves a study: it can be a new bio bank entry related to a previous sample. I this case, a fact will be added but the last visit related to the subject is ancient.
Right censor built on observation fact may be biased to longer durations.
Added the possibility of using visit dimension instead of observations facts for construction of right-censoring event.
It is useful if we consider that a fact can generate after a subject leaves a study: it can be a new bio bank entry related to a previous sample. I this case, a fact will be added but the last visit related to the subject is ancient.
Right censor built on observation fact may be biased to longer durations.