Survival Analysis Flashcards by Olivia Hutt

What is the statistical treatment of survival times called?

Survival analysis

How well did you know this?

Not at all

Perfectly

What is the endpoint in survival analysis?

The terminal event of interest (e.g., death, conception)

How well did you know this?

Not at all

Perfectly

What inherent feature makes survival times unsuitable for conventional statistical methods?

The event of interest is almost never observed in all subjects

How well did you know this?

Not at all

Perfectly

What does it mean when survival times are said to be censored?

The period of observation was stopped before the event of interest occurred

How well did you know this?

Not at all

Perfectly

What is a life table?

A table that estimates the proportion of a population surviving a given length of time

How well did you know this?

Not at all

Perfectly

How is the probability of surviving a given length of time calculated?

By considering time in many small intervals, known as conditional probability

How well did you know this?

Not at all

Perfectly

What does a Kaplan-Meier survival curve represent?

A graph of the cumulative survival probability

How well did you know this?

Not at all

Perfectly

What does the Y-axis of a survival curve indicate?

Proportion of individuals still surviving (or event-free) at each time point

How well did you know this?

Not at all

Perfectly

What is the logrank test used for?

Testing the null hypothesis that groups being compared have the same survival experience

How well did you know this?

Not at all

Perfectly

What does a hazard ratio indicate?

The likelihood of an event occurring, factor increase for each unit increase of the variable

How well did you know this?

Not at all

Perfectly

Whats the hazard ratio name in r?

Exp(B)

How well did you know this?

Not at all

Perfectly

What is the assumption made in Cox regression regarding the effects of different variables?

The effects of the different variables on survival are constant over time and additive

How well did you know this?

Not at all

Perfectly

What is the baseline hazard in Cox regression denoted as?

h0(t)

How well did you know this?

Not at all

Perfectly

What does the coefficient in Cox regression output represent?

Log hazard ratios

How well did you know this?

Not at all

Perfectly

What should be the minimum number of events per predictor variable when using cox regression?

At least 10, preferably 20

How well did you know this?

Not at all

Perfectly

What is a common incorrect approach to analyzing survival data?

Summarizing survival by the proportion of subjects still alive at a certain time

What is the significance of a p-value in the logrank test?

It indicates whether there is evidence to suggest a difference in survival between groups

What is the interpretation of a hazard ratio greater than 1?

The predictor increases risk of event occurring

What is the interpretation of a hazard ratio less than 1?

The predictor reduces risk of event occurring

What statistical method is commonly used to analyze time-to-event data?

Cox regression

What information do survival curves provide regarding treatment groups?

Divergence indicates differing survival

What is the significance of the 95% confidence interval in survival analysis?

It provides a range of values within which the true survival probability is likely to fall

What is the assumption regarding censored observations in survival analyses?

There is no information about the times of censored observations; otherwise, they could be considered failures instead of censored.

Why is it unwise to interpret the flattening of survival curves?

Unless there are many subjects still at risk, it may not indicate meaningful results

What is the purpose of performing a Cox regression for each numerical variable?

To determine which variables are significant ## Footnote This helps in identifying independent predictors of survival.

What test can be used for categorical variables in survival analysis?

Logrank test ## Footnote It is often used in conjunction with Kaplan-Meier estimates.

What should be done if a variable's significance changes when included in a model?

Remove it as it may not be an independent predictor of survival ## Footnote This indicates potential confounding.

What is the independent characteristic of hazard in survival analysis?

Hazard is independent of time ## Footnote It relates only to the likelihood of an event occurring.

What assumptions must be considered in Cox regression?

> effects of variables on survival are constant > effects are additive ## Footnote This is a critical assumption for the validity of the model.

What is a source of bias in studies involving volunteer participants?

Self-selection bias ## Footnote Individuals may be predisposed to perform better than the general population.

When can median survival time not be calculated?

If the survival curve does not drop below 0.5. ## Footnote Median values are often not available in studies.

What is a recommended improvement for study design in treatment arms?

Needs to be randomized ## Footnote Randomization helps eliminate bias and ensures balanced factor levels.

How do you calculate 95% CI for cumulative survival?

estimate +/- std.error*1.96

When reporting CI for survival, what do you do if upper limit exceeds 1?

round to 1; total survival probability cannot exceed 1.

What is the interpretation of a hazard ratio?

i.e. hazard ratio of 2.5 indicates the hazard of vomiting in experiment 2 is almost 2.5 times the hazard in experiment 1 (could be a clinically important difference)

How would you interpret a hazard ratio of Exp(B) = 1.03 for a continuous variable?

For each one unit increase in the predictor, the hazard (risk of death/event) increases by 3%.

In survival analysis, what test can be used to identify association?

a) logrank test b) p < 0.05 = association

Why is a z-test not appropriate to analyse survival data of 2 groups?

a) does not take into account the different lengths of follow-up or the censoring b) would produce different results depending on the time point that the classification was made.

What are the steps for carrying out cox regression?

a) analyse each variable individually using cox regression to determine significance. b) put significant variables in model (in order of significance) to determine if they are confounding or independent predictors.