Scribd
Upload
249 views18 pages

Survival Analysis Presentation

This document discusses Kaplan-Meier survival analysis. It introduces survival analysis concepts like estimating time-to-event, comparing times between groups, and assessing covariate relationships. The Kaplan-Meier method is used to analyze time-to-event data when there is censoring. An example analyzes lung cancer patients' survival times after two treatments using Kaplan-Meier curves and log-rank tests in SPSS. The analysis found no significant difference in time to death between the two treatments.

Uploaded by

Azizah Azhar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
249 views18 pages

Survival Analysis Presentation

This document discusses Kaplan-Meier survival analysis. It introduces survival analysis concepts like estimating time-to-event, comparing times between groups, and assessing covariate relationships. The Kaplan-Meier method is used to analyze time-to-event data when there is censoring. An example analyzes lung cancer patients' survival times after two treatments using Kaplan-Meier curves and log-rank tests in SPSS. The analysis found no significant difference in time to death between the two treatments.

Uploaded by

Azizah Azhar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

Survival Analysis

Kaplan-Meier

Table Of Content
Introduction to Survival Analysis Basic Concept of Kaplan Meier Survival Analysis Data Requirement Analysis Using SPSS Interpretation

Introduction
Survival analysis
analyzing longitudinal data on the occurrence of events
death, injury, onset of illness, recovery from illness (binary variables) or transition above or below the clinical threshold of a meaningful continuous variable (e.g. CD4 counts).

randomized clinical trial or cohort study design.

Introduction (2)
Objectives
Estimate time-to-event for a group of individuals (KM) To compare time-to-event between two or more groups (LR) To assess the relationship of co-variables to time-to-event (Cox)

Regression vs. Survival Analysis


Technique Linear Regression Logistic Regression Survival Analyses Mathematical model
Y=B1X + Bo (linear)
Ln(P/1-P)=B1X+Bo (sigmoidal prob.)

Yields
Linear changes Odds ratios Hazard rates

h(t) = ho(t)exp(B1X+Bo)

Gage, 2004
5

Basic Concepts
Why do we use survival analysis
Why not compare mean time-to-event between your groups using a t-test or linear regression?
Ignores censoring

Why not compare proportion of events in your groups using risk/odds ratios or logistic regression?
Ignores time

Time-to-event: The time from entry into a study until a subject has a particular outcome
6

Basic Concepts (2)


Censored
the critical event has not yet occurred lost to follow-up other interventions offered event occurred but unrelated cause

http://faculty.chass.ncsu.edu
7

Basic Concepts (3)


For categorical predictor variables eg. drug vs. placebo, drug doses 0, 20, 50, and 100 mg/day. Assumptions
Independence within the sample Independence of censoring Uniformity within time interval Less censored values Enough sample size No pattern seen in plots of data

(PROPHET StatGuide)
8

Example
We have 25 patient with lung cancer and we want to know the survival time after treatment X. Eg censored case in this study : One patient recently enrolled 1 month ago, one patient did not return to our study after 5 months, one patient died due to other cause after 13 months, one patient still alive after finish study at 44 months. Only 10 subjects died from lung cancer in our study. In dataset 0= censored, 1= died
9

Example
Dataset
SN 1 2 3 4 5 6 7 8 9 10 11 Survival Time 1 5 6 6 9 10 10 10 12 12 12 Status 0 0 1 1 0 1 0 1 1 1 1 SN 13 14 15 16 17 18 19 20 21 22 23 24 25 Survival Time 12 13 15 16 20 24 24 27 32 34 36 36 44 Status 1 0 0 0 0 1 0 0 1 1 1 1 0
10

12

12

SPSS Output

11

SPSS Output

12

Interpretation
25 subjects with lung cancer enrolled in the study and was treated with treatment X. The follow up ranged from 1 to 44 months. The censored rate was 44%. The median survival for the subjects who had this treatment was 32 months (CI = 2.78 61.22)
Kaplan Meier Technique Survival Time Mean Median 24.55 32.00 Standard Error 2.92 14.91 95% CI 18.83,30.27 2.78, 61.22

13

Next Example
Is there any difference between treatment X and Y in term of shorter time of death? We have data for treatment Y

14

SPSS Output

15

Interpretation
50 subjects with lung cancer enrolled in this study. 25 of them had treatment X and another half had treatment Y. The censored rate was 36%. The follow up range from 1 to 44 months.

Kaplan Meier Technique Treatment X Mean Median Survival Time 24.55 32.00 Standard Error 2.92 14.91 95% CI 18.83,30.27 2.78, 61.22

16

Interpretation
Kaplan Meier Technique Treatment Y Mean Median Survival Time 23.92 20.00 Standard Error 3.07 4.70 95% CI 17.92,29.93 10.78, 29.22

Log rank test was used to compare the two groups. There is no significant difference on having shorter time to event(death) between treatment X and treatment Y groups ( = 0.032, df = 1, p > 0.05)
*One common misconception of survival analysis is that some researchers interpret the result as one group being more likely to have deaths (this should be given by logistic regression!).

17

References
PROPHET StatGuide: Do your data violate Kaplan-Meier assumptions? Retrieved November 20th, http://www.basic.northwestern.edu/statguide files/kaplan_ass_viol.html Chan, Y. H. (2004). Biostatistics 203. Survival analysis. Singapore Med J, Vol 45(6), 249

18

You might also like