Differentiate interval censoring from right censoring and its analysis implications.

• A. Interval censoring occurs when the exact event time is known; right censoring is unknown
• B. Interval censoring is when the exact event time is before a fixed date; right censoring is after
• C. Interval censoring occurs when the exact event time lies between observation times and requires specialized methods; right censoring is common and handled by standard survival analysis
• D. Interval censoring cannot be analyzed in standard software

Answer: (C)

The important idea is how censoring changes how we use time-to-event data. Interval censoring means we only know the event happened somewhere between two observation times, so the exact event time is unknown but lies within a known interval. Because of that uncertainty about the precise moment, you need specialized methods that model the likelihood or distribution over the interval (for example, Turnbull’s nonparametric estimator or interval-censored parametric models). Right censoring, by contrast, is when the event has not occurred by the end of follow-up (or the person is lost to follow-up), so we know the event time is at least a certain value but not when it occurs; this type of censoring is exactly what standard survival analysis methods—like Kaplan-Meier estimators or Cox proportional hazards models—are designed to handle. Therefore, the best characterization is that interval censoring involves the exact event time lying between observation times and requires specialized methods, while right censoring is common and handled by standard survival analysis.