Censored time-to-event data, where not all subjects experience the event of interest, are common in biomedical research. This module introduces some essential statistical tools in the so-called “survival analysis” of censored time-to-event data that are frequently encountered in biomedical research. The module will:

• Introduce important functions, including the survival function, the hazard function, and the median survival time, in analysis of time-to-event data;
• Review life-table analysis, and introduce Kaplan-Meier estimates;
• Introduce log-rank tests, and alternative testing procedures that weight group comparisons differently over the follow-up time interval;
• Introduce the Cox proportional hazards model for regression analysis of censored time-to-event outcomes;
• Cover power and sample size calculation for the design of a clinical study with censored time-to-even outcomes;
• Introduce other topics, such as competing risks and biased sampling, arising from observational studies, if time permits.

The course will focus on understanding foundational concepts; mathematical details will be kept to a minimum.  Examples of real biomedical studies will be used to demonstrate how survival analysis is performed, reported, and interpreted.   Working knowledge of basic probability and statistical concepts will be assumed.