Kaplan-Meier Survival Analysis
Enter follow-up time and outcome events to draw Kaplan-Meier survival curves and compute survival probabilities at each time point and the median survival time; with groups, it automatically draws multiple curves and runs the log-rank test to compare between-group survival. A core tool for oncology prognosis research.
① Input data
One case per row: follow-up time + event (1=endpoint occurred, e.g. death/relapse; 0=censored/lost) + group (optional). Space/Tab/comma-separated; you can paste from Excel.
How to use & methodology
What is Kaplan-Meier analysis used for?
To analyse time-to-event data 'from a start point to an outcome event', such as overall survival or progression-free survival in cancer patients. It handles censored data (lost to follow-up, still alive at study end) and is the most common survival-analysis method in prognosis research.
What is censoring (event=0)?
Censoring means an individual who did not experience the endpoint during observation, e.g. still alive at follow-up's end, or lost mid-study. These individuals provide the information 'survived at least until some time', which KM correctly uses rather than discarding. Enter 0 in the event column.
How do I understand median survival time?
The time at which the survival probability drops to 50%, i.e. the point where exactly half the individuals have had the endpoint. If the curve's tail stays above 50% (most individuals alive or censored), the median survival is 'not reached', which is common in cohorts with good prognosis.
What does the log-rank test tell me?
It compares the overall difference between two survival curves. p<0.05 suggests a statistically significant survival difference. Note it only answers 'is there a difference', not its size; to quantify the hazard ratio (HR), use Cox regression.