How to Make a Kaplan-Meier Curve for Scientific Papers

A complete guide to creating publication-quality Kaplan-Meier survival curves. From survival data to journal-ready export.

What is a Kaplan-Meier curve?

A Kaplan-Meier curve (or survival curve) is a statistical visualization that estimates the probability of survival over time. It is the standard method for displaying survival analysis results in clinical trials, oncology, and epidemiology. The curve shows the proportion of subjects who have not experienced the event (e.g., death, relapse, progression) at each time point.

Key elements:

  • • Survival probability (y-axis) — proportion surviving at each time
  • • Time (x-axis) — days, months, or years
  • • Step function — drops at each event time
  • • Tick marks — censored observations
  • • Log-rank p-value — statistical comparison between groups
  • • Number at risk table — subjects remaining at each time

Step-by-Step Guide

  1. Collect survival data

    For each subject, record: time to event or censoring, event status (1=event, 0=censored), and group (treatment, control, stage, etc.).

  2. Compute survival probabilities

    Apply the Kaplan-Meier estimator: S(t) = ∏ (1 - d_i / n_i) where d_i is events at time t_i and n_i is subjects at risk.

  3. Plot curves

    Y-axis: survival probability (0–1). X-axis: time (days, months, or years). Plot each group as a separate step-function line.

  4. Add tick marks

    Place small vertical tick marks at each censored observation time. This shows where subjects were lost to follow-up.

  5. Add statistics

    Include the log-rank p-value on the plot. Add a table with sample sizes, events, and median survival per group below the plot.

  6. Export for publication

    Set width to journal column or double column. Export at 300 DPI. Ensure labels are readable. Use colorblind-friendly line styles.

Interpreting a Kaplan-Meier Curve

  • ✓Steeper drop — Higher event rate. A curve that drops quickly indicates poor survival.
  • ✓Flat region — No events during that period. Indicates stable survival.
  • ✓Tick marks — Censored subjects. Many tick marks at late times suggest short follow-up.
  • ✓Group separation — Clear separation with p < 0.05 indicates different survival between groups.

Frequently Asked Questions

How do you make a Kaplan-Meier curve for a scientific paper?

To make a Kaplan-Meier curve: (1) collect survival data (time to event or censoring, event status), (2) compute survival probabilities using the Kaplan-Meier estimator, (3) plot survival probability (y-axis) vs time (x-axis), (4) add tick marks for censored observations, (5) include confidence intervals (optional), (6) add a legend with group names and sample sizes, (7) include a p-value from log-rank test, (8) export at 300 DPI at journal width.

What is a Kaplan-Meier curve used for?

A Kaplan-Meier curve is used to estimate survival probability over time from lifetime data. It is the standard visualization for survival analysis in clinical trials, oncology, and epidemiology. The curve shows the proportion of subjects surviving (or remaining event-free) at each time point.

What do the tick marks on a Kaplan-Meier curve mean?

Tick marks (small vertical lines on the survival curve) indicate censored observations — subjects who were lost to follow-up, withdrew, or had not experienced the event by the end of the study. Censored subjects contribute to the survival probability up to their last known time but are not counted as events.

What is the best tool for making Kaplan-Meier curves?

FigureGuild is ideal for publication-ready Kaplan-Meier curves. It auto-computes survival probabilities, adds tick marks, and applies journal formatting. R (survival package + ggplot2) and Python (lifelines) are also used but require coding. GraphPad Prism supports survival analysis with a paid license.

Should Kaplan-Meier curves include confidence intervals?

Confidence intervals are recommended for single-group curves but optional for multi-group comparisons. CIs show the uncertainty around the survival estimate. For multi-group curves, the p-value from the log-rank test is more informative. If showing CIs, use shaded regions or dashed lines.

How do you compare groups in a Kaplan-Meier curve?

Groups are compared using the log-rank test (Mantel-Cox test). The p-value from the log-rank test is displayed on the plot. If p < 0.05, the survival curves are significantly different. The hazard ratio (HR) with 95% CI is also commonly reported.

Related Pages

Kaplan-Meier TemplateAI Figure GeneratorExport at 300 DPI

Create Kaplan-Meier Curves Automatically

FigureGuild auto-computes survival probabilities, adds tick marks, and exports at journal dimensions — all from your clinical data.

Try FigureGuild Free