1/17
Looks like no tags are added yet.
absolute risk reduction (ARR)
X-Y
relative risk or risk ratio (RR)
Y / X
relative risk reduction (RRR)
1- RR
number needed to treat (NNT)
1 / ARR (taken out of %)
primary composite outcome
use absolute risk, and absolute risk reduction
relative risk
can be in decimal or in percentage
but make sure if using percentage to move the decimal
RR and OR
the outcome must occur infrequently (less than 10% event rates) in both the treatmen and control group for them tho be though of similarly
small
for low event rates, the OR and RR will be close. the RR and OR will be closer together when the magnitude of the treatment effect is small
risk
the probability of an outcome of interest/ all possible outcomes
odds
the probability of an event/ probability of event not occuring
RR and OR commonalitiies
A RR or OR of 1 indicates there is no difference in risk between groups being compared
a R or OR of more than 1 indicated an increase in risk or odds among the exposed compared to the unexposed
a RR or OR < 1 indicates a decrease in risk (for odds) in the exposed group
confidence intervals can be calculated for both RR and OR
retrospective study
use odds
allows for measurement of the strength of the association between exposure and outcome
we don’t know the total number of exposed people so we cannot use risk
prospective study
you can use risk because the total number of people exposed in available or you can use odds because you have the information needed to calculate it
odds
what do you use for retrospective study
risk or odds
what do you use for prospective study
number needed to treat (NNT)
1 / ARR
# needed to treat for a certain time to prevent 1 event
round up
do not calculate for non-significant findings
number needed to harm (NNH)
1/ ARI
# needed to treat for a certain time to cause 1 adverse event or harm
round down because it is more conservative
do not calculate for non-significant findings
forest plot
a graphical display of results that summarizes individual populations (subgroups) or individual studies (in the case of a meta-analysis)