Plot survival curves, compute median survival, and run log-rank tests. Handles censored observations with Greenwood confidence intervals. Data never leaves your browser.
Try it out
Load example survival data to see the full workflow
Columns: Time, Event (1/0), Group (optional). Tab or comma separated.
When to use
Do not use for
KM assumes censoring is non-informative — i.e., censored subjects have the same future survival as those remaining. If subjects drop out because they are sicker, KM overestimates survival.
The log-rank test is most powerful when hazards are proportional (curves don't cross). If groups have similar early survival but diverge later, consider the Wilcoxon or Fleming-Harrington test instead.
With fewer than ~20 subjects per group, confidence intervals are wide and the log-rank test has limited power. Report exact p-values rather than significance thresholds.
If fewer than 50% of subjects experience the event, median survival is undefined. Report "not reached" rather than extrapolating or using the largest observed time.
Kaplan-Meier product-limit estimator with Greenwood variance formula for 95% CIs. Log-rank (Mantel-Cox) chi-square test for group comparisons with chi-square CDF via series expansion.
Last validated 2026-04-05. Calculations are designed for planning and documentation support; verify procurement decisions against manufacturer specifications or institutional SOPs.
ConductScience Kaplan-Meier Survival Analyzer (v1.0). ConductScience, Inc. 2026. Available at: https://conductscience.com/tools/kaplan-meier-analyzer
Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53(282):457-481. doi:10.1080/01621459.1958.10501452
Survival analysis studies the time until an event of interest occurs. Unlike standard statistical methods, it handles censoring — the incomplete observation of event times.
Key concepts: • Survival function S(t): Probability of surviving beyond time t • Hazard function h(t): Instantaneous event rate at time t, given survival to t • Censoring: Right-censoring (most common) occurs when follow-up ends before the event • At-risk set: Subjects still being observed at each time point
Kaplan-Meier analysis is used in:
• Clinical trials: Primary endpoint for overall survival, progression-free survival, disease-free survival • Preclinical studies: Tumor growth delay, time-to-event in animal models • Device reliability: Time-to-failure analysis for medical devices • Behavioral research: Latency to first response, time to criterion performance
The log-rank test is the standard method for comparing treatment vs. control survival curves in randomized trials.
