Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

SURVIVAL

ANALYSIS:
KAPLAN-MEIER
METHOD AND COX
REGRESSION
STRATONE
✓Survival analysis, also known as time-to-event analysis,
is a branch of statistics that studies the amount of time it
takes before a particular event of interest occurs
(https://www.investopedia.com/terms/s/survival-

What is analysis.asp). Say the time a certain disease will be


cured.

Survival ✓It was first developed as a tool in biomedical sciences to


study the onset of certain ailment or disease. However,
Analysis? this method is now used in business, engineering,
education, etc.

✓The goal of survival analysis is to “predict” the time or a


particular individual or groups of individual to experience
a specific event of interest.
1.KM Survival Analysis cannot use
multiple predictors.
2.KM Survival Analysis can run only on a
single binary predictor
KAPLAN- 3.KM is a non-parametric procedure.
MEIER
4.KM is used to compare survival
SURVIVAL function of two groups (based on the
ANALYSIS predictor variable)
5.KM involves estimating the probability
of survival in a specified length of time.
1: The event status should consist of two mutually
exclusive and collectively exhaustive states: "censored"
or "event" (where the "event" can also be referred to
as "failure"). The event status is mutually
exclusive because the outcome for a case can either be
KAPLAN- censored or the event has occurred.

MEIER 2: The time to an event or censorship (known as


the "survival time") should be clearly defined and
Assumptions precisely measured. The Kaplan-Meier method, requires
the survival time to be recorded precisely (i.e., exactly
https://statistics.laerd.com/spss-
tutorials/kaplan-meier-using-spss- when the event or censorship occurred) rather than
statistics.php
simply recording whether the event occurred within
some predefined interval. In addition, the survival time
should be clearly defined, whether this is measured in
days, weeks, months, years, or some other time-based
measurement.
3: Where possible, left-censoring should be minimized
or avoided. Left-censoring occurs when the starting point
of an experiment is not easily identifiable.
4: There should be independence of censoring and the
event. This means that the reason why cases are censored
KAPLAN- does not relate to the event. For the assumption of
independent censoring to be met, we need to be confident
MEIER that when we record that a participant is "censored", this is

Assumptions not because they were at greater risk of the event occurring
(i.e., "death" being the "event" in this case). Instead, there
may be many other reasons why a participant is
"legitimately censored“.
5: There should be no secular trends (also known as secular
changes). A characteristic of many studies that involve
survival analysis is that: (a) there is often a long time period
between the start and end of the experiment; and (b) not all
cases (e.g., participants) tend to start the experiment at the
same time.

KAPLAN- 6: There should be a similar amount and pattern of censorship


per group. One of the assumptions of the Kaplan-Meier

MEIER
method and the statistical tests for differences between group
survival distributions is that censoring is similar in all groups
tested. This includes a similar "amount" of censorship per
Assumptions group and similar "patterns" of censorship per group. Failure to
meet the assumption can lead to false conclusions being drawn
about differences in group survival distributions (i.e., rejection
or not of the null hypothesis), based on these statistical tests
"confusing" differences in censoring patterns with actual group
differences in survival distributions (Bland & Altman, 2004;
Hosmer et al., 2008; Norušis, 2012).
✓Survival Time - is defined as the time starting from a
predefined point to the occurrence of the event of
interest. (Bewick V, et.al. 2004)

✓Survival Function defines the probability that the

Basic event of interest has not occurred at time t, S(t) =


P(T > t). It can also be interpreted as the probability of

Concepts survival after time t.


(https://lifelines.readthedocs.io/en/latest/Survival%20Analysis%20intro.html)

✓The Hazard Function also called the intensity function,


is defined as the probability that the subject will
experience an event of interest within a smalltime
interval, provided that the individual has survived until
the beginning of that interval. (www.whatisseries.co.uk)

✓Cumulative Hazard represents the accumulation of


hazard (or risk) for the terminating event over the course
of the observation period (Luke & Homan, 1998).
✓The Kaplan-Meier survival curve is defined as the
probability of surviving in each length of time
while considering time in many small intervals.
KM - ✓There are three assumptions used in this analysis.
SURVIVAL ➢Firstly, we assume that at any time patients who are

CURVE
censored have the same survival prospects as those who
continue to be followed.
➢Secondly, we assume that the survival probabilities are
the same for subjects recruited early and late in the
study.
➢Thirdly, we assume that the event happens at the time
specified. This creates problem in some conditions when
the event would be detected at a regular examination.
(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3059453/)
SURVIVAL CURVE
INTERPRETATION
The horizontal axis represents time in years, and the vertical axis
shows the probability of surviving or the proportion of people
surviving.
➢ At time zero, the survival probability is 1.0 (or 100% of the
participants are alive).
➢ At 2 years, the probability of survival is approximately 0.83 or
83%.
➢ At 10 years, the probability of survival is approximately 0.55 or
55%.
➢ The median survival is approximately 11 years.

A flat survival curve (i.e. one that stays close to 1.0) suggests very
good survival, whereas a survival curve that drops sharply toward 0
suggests poor survival. (Sullivan, Lisa,
https://sphweb.bumc.bu.edu/otlt/mph-
modules/bs/bs704_survival/BS704_Survival_print.html)
1. The log rank test is a popular test to test the null
hypothesis of no difference in survival between two
or more independent groups. The test compares
COMPARING the entire survival experience between groups and

SURVIVAL
can be thought of as a test of whether the survival
curves are identical (overlapping) or not.
TIMES https://sphweb.bumc.bu.edu/otlt/mph-
modules/bs/bs704_survival/BS704_Survival5.html)
BETWEEN 2. Peto and Peto's version of the log-rank test (after
GROUPS Richard and Julian Peto)
3. Gehan's generalized Wilcoxon test or Gehan–
Breslow–Wilcoxon test (after Edmund Alpheus
Gehan, Norman Edward Breslow and Frank
Wilcoxon)
4. Tarone–Ware test (after Robert E. Tarone and
James Ware)
✓A regression techniques for survival
analysis, which is used to relate several risk
factors or exposures, considered
COX simultaneously, to survival time.

REGRESSION ✓It is a semi-parametric test.


✓Compared to KM survival analysis, it can be
used for multiple predictors.
✓Assumptions. Observations should be
independent, and the hazard ratio should be
constant across time; that is, the
proportionality of hazards from one case to
another should not vary over time. The latter
assumption is known as the proportional
hazard's assumption.
➢Log likelihood ratio used to test the significance of the
regression model (synonymous to chi-square statistic).

➢Beta coefficients (slope) – determine the effect of


predictor to survival probability. It can be positive or

Important
negative. Positive slope indicates more likely of larger
hazard rate, while negative means less likely hazard rate.

Statistics ➢Hazard ratio (Exp(B)) - reflects the multiplicative change


in the hazard for the terminal event per unit increase on
a predictor. Hazard ratios less than 1 are associated with
negative regression slopes, whereas values greater than
1 are associated with positive slopes (Tabachnick &
Fidell, 2013). A hazard ratio of 1, indicates there is no
change in the hazard per unit change on the covariate.
Biostatistics Series Module 9: Survival Analysis, Avijit
Hazra and Nithya Gogtay.
www.ncbi.nlm.nih.gov/pmc/articles/PMC5448258/
Introduction to survival analysis,
www.lifelines.readthedocs.io/en/latest/Survival%20Analysis%20i
ntro.html

References Statistics review 12: survival analysis. Bewick V, Cheek L, Ball J.


Crit Care. 2004 Oct; 8(5):389–94)
What is a Cox model?, Stephen J
Walters. www.whatisseries.co.uk
Understanding survival analysis: Kaplan-Meier estimate, Manish
Kumar Goel, Pardeep Khanna, and Jugal Kishore.
www.ncbi.nlm.nih.gov/pmc/articles/PMC3059453/
Luke, D.A., & Homan, S.M. (1998). Time and change: Using survival
analysis in clinical assessment and treatment evaluation.
Psychological Assessment, 10, 360-378.
Tabachnick, B.G., & Fidell, L.S. (2013). Using multivariate statistics
(6th ed). Upper Saddle River, NJ: Pearson

You might also like