Survival Analysis in R

Biostatistics

The complete, practical guide to survival analysis in R — censoring and time-to-event data, Kaplan-Meier curves, the log-rank test, and the Cox proportional-hazards model, with publication-ready survminer visuals. Recipe-first, reproducible, clinician-friendly.

Learn › Biostatistics › Survival Analysis

Survival analysis answers a question ordinary statistics can’t: how long until something happens — and what changes that time? Time to death, to relapse, to discharge, to a part failing. The catch is censoring: when the study ends, many patients are still alive, so you know they survived at least this long but not their true event time. A t-test or a linear model throws that information away; survival methods use it. This is the complete, modernized guide to time-to-event analysis in R — every method as a recipe you can run, with publication-ready survminer curves you can paste straight into a paper.

How this guide is organized

A · Fundamentals

Censoring · hazard & survival functions

B · Curves (KM)

Kaplan-Meier · median survival · publication curves

C · Comparing groups

Log-rank · stratified & weighted tests

D · Cox regression

Hazard ratios · PH assumptions · model building

E · Advanced

Competing risks · RMST · parametric & frailty

The guide follows the same journey as a survival-analysis course, in order:

  1. Fundamentals — start with what survival analysis is: what censoring means, why ordinary methods fail on time-to-event data, and how the survival and hazard functions describe time until an event.
  2. Curves & estimation (Kaplan-Meier) — estimate survival probabilities and read the median survival from a Kaplan-Meier curve — the signature ggsurvplot with a risk table, confidence band, and the median line.
  3. Comparing groups — test whether survival differs between groups (treatment vs control, biomarker-high vs -low) with the log-rank test, and plot the curves with the p-value on the panel.
  4. Cox regression — model how covariates change the hazard with the Cox proportional-hazards model: hazard ratios, 95% CIs, and a ggforest plot — the workhorse of clinical survival analysis.
  5. Advanced & applied — extend to PH-assumption diagnostics, competing risks, parametric models, and regulatory time-to-event work as later chapters land.

Which method should I use?

The right tool depends on what you’re asking and what you have measured:

Your question Method Start here
What does my time-to-event data even mean? Concepts (censoring, hazard, survival) What is survival analysis?
What’s the survival probability over time / the median? Kaplan-Meier estimation Kaplan-Meier
Do two or more groups survive differently? Log-rank test Log-rank test
How do covariates (age, sex, dose, biomarker) change the hazard? Cox proportional-hazards Cox regression

Whatever you choose, the workflow is the same: build the Surv(time, status) object → estimate (KM or Cox) → draw the publication ggsurvplot (curves + risk table + CI + p-value) → interpret in plain language (median survival, log-rank p, hazard ratio with its 95% CI) → check assumptions → report.

All lessons

An overall Kaplan-Meier survival curve for the 228 lung-cancer patients. The stepped line starts at a survival probability of 1.0 at day zero and falls steadily, dropping below 0.5 around 310 days (the median survival). A shaded blue 95% confidence band surrounds the curve, small vertical tick marks show censored patients, and a number-at-risk table runs along the bottom showing how many patients remain over time.

What Is Survival Analysis? Censoring, Hazard & Survival

A gentle introduction to survival analysis in R. Understand time-to-event data (time to death, relapse, or discharge), why right-censoring breaks a t-test or linear…

0117 minLive

Two Kaplan-Meier curves overlaid on the same 929 colon-cancer patients. The overall-survival curve (event = death only) sits higher and falls more slowly, crossing 50% at about 2550 days. The recurrence-free survival curve (event = recurrence or death, whichever first) falls earlier and lower, crossing 50% at about 1590 days. The gap between the curves is the burden of recurrences that happen before death — the same cohort, two endpoints, two different answers.

Choosing a Survival Endpoint: OS vs PFS, EFS, TTP & DFS

Overall survival (OS), progression-free survival (PFS), event-free survival (EFS), time-to-progression (TTP), disease-free survival (DFS) and time-to-treatment-failure (TTF)…

0221 minLive

A curve of the required number of events against the target hazard ratio, for 80% power, a two-sided 5% significance level, and 1:1 allocation. The curve is low and flat for strong effects (HR around 0.5 needs about 66 events) and rises steeply as the hazard ratio approaches 1 (HR 0.7 needs 247 events, HR 0.8 needs 631, HR 0.85 needs nearly 1,200). A dashed marker highlights HR 0.7 at 247 events.

Sample Size for Survival Analysis: How Many Events Do You Need?

Design a time-to-event study before you collect data. Learn the one insight that governs survival power — it is driven by the number of EVENTS, not the number of patients —…

0320 minLive

A follow-up timeline for six patients. Each patient is a horizontal line running from study entry at time zero to their last contact. Patients 1, 3, and 5 end in a solid dot marking the observed event (death). Patients 2, 4, and 6 end in an open right-pointing arrow marking that they were still event-free at last follow-up (right-censored). A dashed vertical line marks the study data-cutoff time, after which no events can be observed.

How to Format Survival Data in R: The Surv() Object, Censoring & Validation

Get your data ready for survival analysis in R. Build the Surv(time, status) object, understand the (time, status) contract and status coding (0/1 vs 1/2 — and the gotcha)…

0420 minLive

A survminer Kaplan-Meier plot of survival by sex on the lung-cancer data. Two step-function survival curves descend from 1.0 over time, the female curve staying consistently above the male curve, each with a shaded 95% confidence band. Dashed median-survival lines mark 270 days for men and 426 days for women. The log-rank p-value (0.0013) is printed on the plot, and a risk table beneath shows the number of patients still at risk over time.

Kaplan-Meier Survival Curve in R: Estimate and Plot, Step by Step

Learn Kaplan-Meier estimation in R with the survival and survminer packages. Build the Surv(time, status) object, estimate survival curves with survfit on the NCCTG…

0516 minLive

A step-function survival plot on the lung-cancer data. The continuous black Kaplan-Meier curve descends in many small steps from 1.0 at time 0 to about 0.13 by 720 days. Orange points mark the actuarial life-table survival at the end of each 90-day interval; every point sits almost exactly on the Kaplan-Meier curve, showing the two estimates agree.

Life Table Analysis in R: The Actuarial Method, Step by Step

Learn the actuarial life table method in R: estimate survival when follow-up data come in grouped time intervals rather than exact event times. Build the table by hand with…

0618 minLive

A survminer Kaplan-Meier plot of survival by sex on the lung-cancer data. Two step-function survival curves descend from 1.0, the female curve staying above the male curve, each with a shaded 95% confidence band. A dashed horizontal line at survival probability 0.5 meets dashed vertical lines dropping to the x-axis at 270 days for men and 426 days for women, marking the median survival of each group.

Median Survival Time and 95% Confidence Interval in R

Compute the median survival time and its 95% confidence interval in R from a Kaplan-Meier fit with survfit(), read them in plain language, handle the case where the median…

0714 minLive

A basic survminer survival curve of survival by sex on the lung-cancer data: two coloured step-function curves descending from 1.0 over time, with censoring tick marks and a top legend, but no confidence band, p-value, or risk table yet.

ggsurvplot in R: Publication-Ready Survival Curves with survminer

A practical survminer reference for ggsurvplot(): start from a basic survival curve and add, one element at a time, the number-at-risk table, 95% confidence bands, the…

0819 minLive

A survminer Kaplan-Meier plot comparing survival of male and female lung-cancer patients. The female curve stays above the male curve throughout follow-up, each with a shaded 95% confidence band; the log-rank test p-value of 0.0013 is printed on the plot, and a risk table beneath shows the number of patients still at risk over time.

Log-Rank Test in R: Compare Survival Curves Between Groups

Learn the log-rank test in R with the survival and survminer packages. Compare survival between groups on the NCCTG lung-cancer data with survdiff(), read the chi-square…

0916 minLive

A survminer Kaplan-Meier plot split into three side-by-side panels, one per ECOG performance score (0, 1, 2), each comparing the survival of male and female lung-cancer patients. The female curve sits above the male curve in every panel; within-stratum log-rank p-values of 0.11, 0.0068, and 0.25 are printed on the panels, with shaded 95% confidence bands.

Stratified Log-Rank Test in R: Compare Survival, Adjust for a Confounder

Learn the stratified log-rank test in R with the survival and survminer packages. Compare survival between groups while controlling for a confounding prognostic factor using…

1019 minLive

A Kaplan-Meier plot comparing survival of male and female lung-cancer patients. The female curve separates from the male curve early — within the first roughly 300 days — and stays above it. A shaded band marks that early period, where an early-weighted Fleming-Harrington test concentrates its weight.

Weighted Log-Rank Tests in R: Fleming-Harrington for Non-Proportional Hazards

Run weighted log-rank tests in R when proportional hazards fail. Use survdiff(rho = 1) for the early-weighted Peto-Peto test and survMisc::comp() for the Fleming-Harrington…

1119 minLive

A survminer Kaplan-Meier plot comparing survival of three colon-cancer treatment arms. The Levamisole-plus-fluorouracil curve sits clearly above the Observation and Levamisole-alone curves, which overlap each other; each curve has a shaded 95% confidence band, the overall log-rank p-value of 0.003 is printed on the plot, and a risk table beneath shows the number of patients still at risk in each arm over time.

Pairwise Log-Rank Test in R: Which Survival Curves Differ?

Compare three or more survival curves in R. After an overall log-rank test says survival differs, run pairwise log-rank tests with survminer’s pairwise_survdiff() and adjust…

1217 minLive

A survminer ggforest forest plot of the multivariate Cox model on the lung data. Sex (female) sits clearly left of the reference line at hazard ratio 1 (protective), ECOG performance score sits right of it (harmful), and age sits on the line. Each row shows the hazard ratio, 95% confidence interval, and p-value.

Cox Proportional Hazards Model in R: Hazard Ratios, Step by Step

Learn the Cox proportional hazards model in R with the survival and survminer packages. Fit univariate and multivariate coxph models on the NCCTG lung-cancer data, interpret…

1317 minLive

A survminer ggcoxzph diagnostic with one panel per covariate (age, sex, ECOG score), each showing scaled Schoenfeld residuals plotted against transformed survival time. In every panel the solid smoothing spline is roughly flat and horizontal, staying within the dashed plus-or-minus two-standard-error band, indicating no systematic time trend and supporting the proportional-hazards assumption.

Testing the Proportional Hazards Assumption in R (cox.zph, Schoenfeld)

Validate a Cox proportional hazards model in R. Test the proportional-hazards (PH) assumption with cox.zph and the scaled Schoenfeld residual plot ggcoxzph, check for…

1419 minLive

A survminer ggforest forest plot of the final five-variable Cox model on the lung data. Sex (female) sits clearly left of the reference line at hazard ratio 1 (protective) and ECOG performance score sits well right of it (harmful); physician Karnofsky, patient Karnofsky, and weight loss sit close to the line. Each row shows the hazard ratio, 95% confidence interval, and p-value, with the global model tests and concordance reported beneath.

Cox Model Variable Selection in R: Build a Multivariable Model

Build and select a multivariable Cox proportional hazards model in R. Screen candidate covariates univariately, compare selection strategies — clinical knowledge, stepwise…

1523 minLive

A forest plot comparing the transplant hazard ratio from two models on the Stanford heart-transplant data. The naive baseline-only model sits far left of the reference line at 1 (hazard ratio about 0.18, a spuriously strong protective effect from immortal-time bias), while the correct time-varying model sits right on the line at hazard ratio about 1.0 (no effect). The contrast shows the bias the time-varying analysis removes.

Time-Varying Covariates in a Cox Model in R (and Immortal-Time Bias)

Model covariates whose value changes during follow-up with a Cox model in R. Build the counting-process (start, stop, event) data layout with tmerge, fit coxph(Surv(start…

1620 minLive

A survminer ggadjustedcurves plot of adjusted survival probability against time in days for three tumor-differentiation strata from a stratified Cox model on the colon-cancer data. The well-differentiated curve stays highest, the moderate curve sits in the middle, and the poorly-differentiated curve drops fastest and lowest, the three coloured curves separating clearly over follow-up.

Stratified Cox Model in R: Fix a Proportional-Hazards Violation with strata()

Learn the stratified Cox model in R with the survival and survminer packages. When the proportional-hazards assumption fails for a categorical covariate, stratify on it with…

1721 minLive

Fitted parametric survival curves for the Weibull, exponential, and log-normal distributions overlaid on the black Kaplan-Meier step function for the lung-cancer data. The Weibull curve tracks the Kaplan-Meier closely across the whole follow-up, the log-normal is close but slightly high in the tail, and the exponential diverges noticeably, sitting above the data early and dropping too slowly.

Parametric Survival Models in R: Weibull, AFT, and Extrapolation

Learn parametric survival models in R with the survival and flexsurv packages. Fit Weibull, exponential, log-normal and log-logistic accelerated failure time (AFT) models…

1823 minLive

Two cumulative incidence curves for progression to plasma-cell malignancy over 24 years on the mgus2 data. The naive 1 minus Kaplan-Meier curve climbs steadily to about 21 percent, while the correct cumulative incidence function curve stays much lower, levelling near 10 percent. The widening gap between them is the bias from treating the competing risk of death as censoring.

Competing Risks Analysis in R: Cumulative Incidence and Fine-Gray

Run a competing risks analysis in R with the survival and tidycmprsk packages. See why 1 - Kaplan-Meier overestimates the cumulative incidence of an event when a competing…

1921 minLive

Two Kaplan-Meier survival curves for a colon-cancer trial, observation in blue and Lev+5FU in gold, with the area under each curve shaded up to a dashed vertical line at five years. The gold treated curve sits above the blue observation curve from about 300 days onward, and the extra gold sliver between the two curves is the gain in restricted mean survival time. The shaded area under each curve up to tau is that arm's RMST.

Restricted Mean Survival Time (RMST) in R: An Interpretable Alternative to the Hazard Ratio

Compute the restricted mean survival time (RMST) in R — the area under the survival curve up to a horizon tau, the expected survival time within the first tau units.…

2021 minLive

A caterpillar plot of the estimated per-litter random effects from a coxme shared-frailty model on the rats data. Each of the 100 litters is a point at its predicted log-hazard random effect, sorted from the lowest (most protective) frailty on the left to the highest (most at-risk) on the right, with a horizontal dashed line at zero marking the average litter. The points fan out well above and below zero, showing that litters differ substantially in baseline tumor risk.

Frailty Models in R: Mixed-Effects Cox for Clustered Survival Data (coxme)

Fit a frailty (mixed-effects / random-effects Cox) model in R with the coxme package. When survival times are clustered or correlated — patients within hospitals, repeated…

2122 minLive

A submission-style Kaplan-Meier plot of overall survival by treatment arm from the CDISC ADTTE dataset. Two step-function survival curves descend from 1.0 over time; the Lev+5FU curve stays consistently above the Observation curve, each with a shaded 95% confidence band. A dashed median-survival line marks 2083 days for the Observation arm (the Lev+5FU median is not reached). The log-rank p-value (0.0017) is printed on the plot, and a number-at-risk table beneath shows subjects still at risk over time.

CDISC ADTTE in R: Submission-Grade Survival Analysis, Step by Step

Analyze a clinical-trial time-to-event endpoint the regulatory way in R. Map a CDISC ADaM ADTTE dataset to a Surv() object — handling the CNSR convention where 1 means…

2224 min

A publication-ready survminer Kaplan-Meier plot of survival by sex on the lung-cancer data. Two step-function survival curves descend from 1.0; the female curve stays consistently above the male curve, each with a shaded 95% confidence band. Dashed median-survival lines mark 270 days for men and 426 days for women, the log-rank p-value of 0.0013 is printed on the plot, and a risk table beneath shows the number of patients still at risk over time.

Complete Survival Analysis Example in R: Kaplan-Meier, Log-Rank, Cox — Start to Finish

A complete, worked survival analysis in R from start to finish on the NCCTG lung-cancer dataset. Frame the question and build the Surv() response, estimate Kaplan-Meier…

2321 minLive

A submission-style Kaplan-Meier plot of overall survival by treatment arm from the CDISC ADTTE dataset. Two step-function survival curves descend from 1.0 over time; the Lev+5FU curve stays consistently above the Observation curve, each with a shaded 95% confidence band. A dashed median-survival line marks 2083 days for the Observation arm (the Lev+5FU median is not reached). The log-rank p-value is printed on the plot, and a number-at-risk table beneath shows subjects still at risk over time.

ADTTE Analysis Example in R: A Complete Regulatory Survival Package

A complete, worked regulatory time-to-event analysis package on a CDISC ADTTE dataset in R. Build the analysis-ready Surv() object from the CNSR flag, run the pre-specified…

2428 min
No matching items
Note

This guide is filling in wave by wave — the high-demand free hooks first: what survival analysis is, Kaplan-Meier estimation, the log-rank test, and the Cox proportional-hazards model. Coming next: PH-assumption diagnostics, median survival and life tables, the survminer publication studio, stratified and parametric Cox models, and advanced topics (competing risks, restricted mean survival time, regulatory time-to-event).

🟢 With an AI agent

Ask Prova “compare survival between my two groups and fit a Cox model on my data” — it answers with survival + survminer code you can run on your own data, then helps you read the Kaplan-Meier curve, the log-rank p-value, and the hazard ratios. The runtime is the judge. Ask Prova →

Was this page helpful?

Prove you can do it. Master the whole Survival Analysis in R series — track your path, build projects, and earn a certificate.

Start free →

Go Pro — unlimited Prova on your own data and a verifiable certificate that proves the skill.

from $15/mo billed yearly

Go Pro →

✓ You're Pro — keep going. The runtime is the judge.

Ready to level up?

Try this on your own data with Prova — the runtime is the judge.Open Prova →

Get new R & Python lessons by email

Practical, reproducible, no spam. Unsubscribe anytime.

Double opt-in. We never share your email.

Share this pageXLinkedInRedditHN

Citation

BibTeX citation:
@online{untitled,
  author = {},
  title = {Survival {Analysis} in {R}},
  url = {https://www.datanovia.com/learn/biostatistics/survival-analysis/},
  langid = {en}
}
For attribution, please cite this work as:
“Survival Analysis in R.” n.d. https://www.datanovia.com/learn/biostatistics/survival-analysis/.