Contents
When to use PROC FREQ to estimate relative risk?
Note that PROC FREQ can be used to estimate the relative risk only when the row variable has two levels. Since the log odds (also called the logit) is the response function in a logistic model, such models enable you to estimate the log odds for populations in the data.
The relative risk (or risk ratio) is an intuitive way to compare the risks for the two groups. Simply divide the cumulative incidence in exposed group by the cumulative incidence in the unexposed group:
What is the risk ratio when RR < 1?
When RR > 1, % increase = (RR – 1) x 100, e.g. (4.2 – 1) x 100 = 320% increase in risk. Interpretation: Those who had the incidental appendectomy had a 320% increase in risk of getting a post-operative wound infection. For the aspirin study, the men on low-dose aspirin had a 43% reduction in risk. When RR < 1,
How to calculate the relative risk of a predictor?
When assessing the effect of a particular predictor in a model, it is of interest to estimate the relative risk for that predictor adjusted for the effects of the other predictors. For a continuous predictor, the relative risk, px+1 / px, is interpreted as the change in event probability for a unit increase in the predictor.
How can I estimate relative risk in SAS using proc genmod?
We could use either proc logistic or proc genmod to calculate the OR. Since proc genmod will be used to calculate the RR, it will also be used to calculate the OR for comparison purposes (and it gives the same results as proc logistic ). Here is the logistic regression with just carrot as the predictor:
How to write a ratio of relative risk?
The function LOGISTIC (x) = [1 + e- (x)]-1 makes it easy to write the relative risk as a ratio of probabilities. As discussed above, to test that the relative risk equals 1, specify null=1 (requires version 1.8 or later of the NLEST macro).