NBSN 8004 - Biostatistics: Module 4 (RR/OR; Sensitivity, Specificity; Survival Analysis)
Sensitivity
"We would expect x% of individuals with the disease to have an abnnormal (positive) result"
Specificity interpretation
"We would expect x% of individuals without the disease to have a normal (negative) result"
Sensitivity formula
# of true positives/# of times event occurred
Specificity formula
#of true negatives/# of cases of no disease
Trade-off between sensitivity and specificity
A highly sensitive test runs the risk of telling too many people they have a disease when they do not. A highly specific test has the risk of failing to identify a disease in someone who actually has it. Cut-offs are set based on the condition the assessment is testing for (i.e., are the consequences greater for missing a disease, or telling someone they have it when they dont?)
Odds ratio formula
OR = (a/b)/(c/d) = (a*d)/(c*d)
Survival time response
Usually, a continuous measure that may be incompletely determined for some subjects, which are censored
Survival analysis
a process used to analyze data in which the time until the event is of interest., where the response is often referred to as a failure time, survival time, or event time.; represented by the Kaplan Meier measure
Censoring in survival analysis
a technique that deals with points of incomplete data (e.g., participant drops out or researchers cannot find/get in contact with them)
What does a relative risk analysis tell us?
how many more/less times likely a person exposed to a risk is to develop a condition relative to an unexposed person
Whats does an odds ratio analysis tell us?
how many time greater/lesser the odds are of finding an exposure in someone with the disease compared to someone without the diseas
Kaplan Meier measure
nonparametric methods of representing survival analysis
Relative risk formula
p(occurrence when exposed to risk)/p(occurrence when not exposed) RR = [a/(a+b)]/[c/(c+d)]
Odds ratio
ratio of an outcome (disease, death, etc.) occurring because exposure to a risk factor represents the odds that an outcome will occur given a particular exposure, compared to the odds of the outcome occurring in the absence of that exposure
Relative risk (RR)
ratio of the probability of an event occurring (death, death, etc.) in an exposed group to the probability of the event occurring in a comparison, non-exposed group (risk of an outcome due to a risk factor)
When to use relative risk vs. odds ratio?
relative risk is used for cohort studies
Odds
the frequency of occurrence of one event over the frequency of the other event (no. of events)/(no. of non-events)
Specificity
the proportion of true negatives identified by the test; refers to the test's ability to correctly detect patients without a condition
Sensitivity
the proportion of true positives identified by the test; refers to the test's ability to correctly detect patients who do have the condition