Sensitivity to censored-at-random assumption in the analysis of time-to-event endpoints.

Sensitivity to censored-at-random assumption in the analysis of time-to-event endpoints.