Censoring Rules in Time to Event Calculations: Understanding Left, Interval, and Right Censoring in Clinical Trials
INTRODUCTION
In clinical trials, time to event (TTE) is an essential endpoint that measures the duration between the start date of observation and the first date of documentation of an event. Censoring rules are used when the event of interest does not occur within the duration of the study. In this article, we will discuss the different types of censoring, including left censoring, interval censoring, and right censoring, and their applications in TTE calculations.
LEFT CENSORING
Left censoring occurs when a subject leaves the study before an event occurs or the study ends before the event has occurred. In this case, the time to event is undecided as the subject had already experienced the event before the study started. Left censoring is not commonly observed in clinical trials. An example of left censoring is when a patient with lung cancer is diagnosed with the disease before the start of the study.
INTERVAL CENSORING
Interval censoring occurs when the onset of an event is known to have occurred between a time interval, and the exact visit of onset is unknown. For instance, in the example of a patient with lung cancer, if the patient had their first visit at 3 months, was found to have no recurrence of the disease, and missed the 6-month review but was later found to have tumor recurrence at 9 months, the exact date when the tumor occurred again is not known. In this case, there are two ways of dealing with interval censoring:
- Midpoint imputation: The midpoint of the censoring interval is taken as the censor date. In the example above, the midpoint of the interval (6 months) is taken as the censor date.
- Right imputation: The date at which the recurrence is first noted is taken as the censor date. In the example above, the date when the tumor recurrence was first noted (9 months) is taken as the censor date.
RIGHT CENSORING
Right censoring occurs when the event of interest did not occur within the defined study period. For example, if a patient with lung cancer does not experience a recurrence of the disease throughout the 1-year duration of the study, the patient will be censored at the last available visit (1 year). Right censoring is the most common type of censoring observed in clinical trials.
CONCLUSION
In conclusion, censoring rules are an essential aspect of TTE calculations in clinical trials. Left censoring, interval censoring, and right censoring are different types of censoring that are used when the event of interest does not occur within the observation period. Understanding the different types of censoring and their applications in TTE calculations is crucial for accurate and reliable results in survival analysis. Censoring rules should be included in the statistical analysis plan, and they should detail how to handle incomplete data in survival analysis.
