Introduction to Survival Analysis Flashcards
what is survival analysis a subset of?
cohort studies
how long have we been doing survival analysis?
as early as 1669 (how many out of 100 people lived until 86 years old)
what are 3 applications of survival analysis in randomized clinical trials?
- disease or diseae free after an intervention
- cured or not cured after a treatment
- dead or alive at the end of a treatment/trial
give an example of survival analysis in medicine?
a retrospective cohort study of the relationship between aspirin, ibuprofen, and mortality after myocardial infarction; found that adding other drugs in combination with aspirin increased survival rate
what are the 3 objectives of survival analysis?
- estimate time-to-event for a group of individuals
- to compare time-to-event between two or more groups
- to assess the relationship of co-variables to time-to-event
give an example of the “estimate time to event for a group of individuals” objective of survival analysis
such as time until second heart attack for a group of MI patients
give an example of the “compare time to event between to or mroe groups” objective of survival analysis
such as treated vs. placebo MI patients in a randomized control trial
give an example of the “assess the relationship of co-variables to time to event” objective of survival analysis
such as does weight, insulin resistance, or cholesterol influence survival time of MI patients?
how is expected time to event calculated?
1/incidence rate
what is time to event?
the time from entry into a study until a subject has a particular outcome
what is censoring?
subjects are said to be censored if they are lost to follow up or drop out of the study, or if the study ends before they die or have an outcome of interest
what are censored subjects counted as?
censored subjects are counted as alive or disease-free in the time they were enrolled in the study
if the dropout of a censored subject is related to both the outcome and the treatment, what could that dropout do?
that dropout may bias the results
what is the data structure of a two-variable outcome survival analysis?
time variable: ti represents the time at last disease-free observation or time at event
censoring variable: if ci=1, subject had the event; if ci=0 then there was no event by time ti
what kind of statistical tests can be used with two-variable survival analysis?
binary statistics like T test and chi squared
