CODEWITHSUNDEEP
r
Piotr Wera
Piotr Wera

Reputation: 49

How to calculate relative risk in R?

How to calculate relative risk in R?

In input i have 

names(DS)
Tab<-table(DS[,5],DS[,11],DS[,3])

      No Yes
  No   4  16
  Yes  40 168

I am new in R programming language...

Upvotes: 3

Views: 25515

Answers (1)

Ben Bolker
Ben Bolker

Reputation: 226332

This is a fairly straightforward calculation; the relative risk is just (pos1/total1)/(pos2/total2) where pos1 is the number of cases in the first group, pos2 in the second group, and the total variables are the group totals

However, you're probably interested in the epitab function of the epitools package:

## maybe:
##   install.packages("epitools")
library("epitools")

See ?epicalc for input formats, but this should work for your example:

tab <- matrix(c(4,16,40,168),byrow=TRUE,nrow=2)
epitab(tab,method="riskratio")
## $tab
##           Outcome
## Predictor  Disease1        p0 Disease2        p1 riskratio     lower    upper
##   Exposed1        4 0.2000000       16 0.8000000  1.000000        NA       NA
##   Exposed2       40 0.1923077      168 0.8076923  1.009615 0.8030206 1.269361
##           Outcome
## Predictor  p.value
##   Exposed1      NA
##   Exposed2       1

You should of course double-check that the results make sense; e.g. p0 is 4/20=0.2 for the first group, 40/208=0.192 for the second group. The risk ratio is ((16/20)/(168/208))=0.8/0.8076=1.009.

Upvotes: 26

Related Questions