Flashcards in Classes #8-#10: Measures of Association Deck (13)
In what kind of study is risk ratio/ relative risk (RR) used?
Customarily used in studies where subjects are allocated based on exposure (y/n) and evaluated for disease (outcome).
**Example: cohort studies
How do you calculate risk (incidence risk (IR)) for both the exposed and non-exposed?
Risk of OUTCOME in "exposed" = A / (A+B)
Risk of OUTCOME in "non-exposed" = C / (C+D)
What is the risk ratio (RR) and how do you calculate it?
The RR is the ratio of the risk from 2 different groups.
RR = risk in exposed / risk in non-exposed
What are the 3 major ratios in research?
Risk Ratio (RR)
Odds Ratio (OR)
Hazard Ratio (HR)
What are the 3 things that you should look for when interpreting ratio's (RR/OR/HR)?
1) direction of words (increased or decreased)
3) group comparison
What does the absolute risk reduction (ARR) tell you and how is it calculated?
ARR defines the risk difference in the outcome among "exposed" that can be "attributed" to the actual exposure.
RRR = Risk in "non-exposed" - risk in "exposed"
How do you calculate the relative risk reduction (RRR)?
RRR = ARR / R-unexposed
How do you interpret number needed to treat and how is it calculated?
Interpretation: number of (whole) patients needed to be treated to experience the studied event outcome.
-- outcome event can be a BENEFICIAL or HARMFUL event
NNT or NNH = 1 / ARR
What kind of study is the odds ratio (OR) typically used for?
OR is customarily used in studies where subjects are allocated based on disease presence (y/n) and evaluated for exposure.
--example: case-control studies
What does the calculation: A/C give?
What does the calculation: B / D give?
The odds of exposure vs. odds of NOT being exposed (in cases)
The odds of exposures vs. odds of NOT being exposed (in controls)
How do you calculate the Odds Ratio (OR)?
OR = [(A/C) / (B/D)]
What is the property in the ratio's RR, OR, and HR is the ratio is 1.0?
Then the event (or outcome) is equally likely for both groups.
- if "ratio" is >1.0 then the event (outcome) is more likely to occur in the main study (comparison) group
- if "ratio" is