Assess survival differences using practical comparison tools. Review log-rank strength, event timing, and median survival. Built for clear reporting, exports, and practical study checks.
| Group | Sample Size | Events | Censored | Median Survival | Total Follow-Up | Expected Events A | Variance |
|---|---|---|---|---|---|---|---|
| Treatment | 120 | 34 | 26 | 18.4 | 1450 | 30.8 | 11.6 |
| Control | 118 | 47 | 19 | 14.2 | 1325 | leave blank if using approximation | leave blank if using approximation |
This calculator compares two survival groups with a log-rank style approach.
Total events: D = Events A + Events B
Total sample: N = nA + nB
Expected events for Group A: EA ≈ D × (nA / N)
Variance: Var ≈ D × (nA / N) × (nB / N) × ((N - D) / (N - 1))
Z score: Z = (OA - EA) / √Var
Chi-square: χ² = Z²
P-value: p ≈ erfc(|Z| / √2)
If you already know expected events and variance from a formal analysis, the tool uses those values directly.
Event rate: events / sample size
Approximate hazard rate: events / total follow-up
Approximate hazard ratio: hazard rate A / hazard rate B
Median survival difference: median A - median B
Median survival ratio: median A / median B
This page is best for structured screening. Use full Kaplan-Meier or Cox methods for final reporting.
A statistical comparison of survival curves calculator helps analysts compare time-to-event outcomes across two groups. It is useful in clinical research, epidemiology, reliability studies, and policy evaluation. Survival analysis often includes censored records. That makes ordinary averages less helpful. This tool focuses on event counts, censoring, follow-up time, and median survival. It then estimates a log-rank style comparison from summary inputs.
The calculator is designed for fast screening. It supports two modes. You can rely on an aggregate approximation. You can also enter custom expected events and variance when those values come from a formal survival model. That flexibility is useful when you review published studies, internal trial summaries, or registry reports.
The main output is a log-rank z score, chi-square value, and p-value. These measures test whether the two survival curves differ beyond random variation. The tool also reports event rates, median survival difference, median survival ratio, and an approximate hazard rate ratio based on follow-up time. These extra outputs help users move from simple significance testing to practical interpretation.
A lower event rate may suggest better survival performance. A longer median survival may also indicate a more durable outcome. Still, these summaries should be read with context. Different censoring patterns, nonproportional hazards, and unequal follow-up can change interpretation. Use this calculator as a structured comparison aid, not as a replacement for full Kaplan-Meier estimation.
This page works well for quick statistical review, feasibility work, and educational survival analysis tasks. It is especially helpful when raw patient-level data is unavailable. Researchers can compare treatment and control groups, benchmark cohorts, or review operational durability curves. Because the interface is simple, teams can also export a clean summary for reports, audits, and presentations.
Before acting on the output, confirm that the event definition is consistent in both groups. Check whether follow-up windows are comparable. Review censoring intensity. If the survival curves cross, the classic log-rank framework may miss important timing effects. In that situation, consider weighted tests, restricted mean survival time, or a Cox model with diagnostics. Good survival interpretation always combines statistics, study design, and domain knowledge.
It compares two survival groups using a log-rank style test from summary inputs. It also reports event rates, approximate hazard rates, and median survival differences.
No. This tool is a structured screening calculator. Full Kaplan-Meier analysis needs patient-level or item-level time-to-event records and usually dedicated statistical software.
Use custom values when a formal survival analysis already produced log-rank components. That keeps the p-value and chi-square closer to your official study output.
Censoring means the event was not observed before the follow-up ended or before a subject left the study. It affects how survival comparisons are interpreted.
Yes, for fast review and teaching. Still, final medical conclusions should rely on full validated survival analysis, study protocol checks, and clinical oversight.
Those measures add practical context. Statistical significance alone does not describe the size or direction of the survival difference between groups.
The standard log-rank framework may become less informative. Crossing curves can suggest time-varying effects, which often need more advanced survival methods.
Yes. After calculation, use the CSV button for spreadsheet output or the PDF button for a printable report summary.