Skip to content

how to calculate odd ratios with R?

5 messages · Luigi Marongiu, Sorkin, John, Michael Dewey +2 more

Luigi Marongiu
# 
Hello,
Is it possible to calculate with a single function the odd ratios?
Now I can use this implement:
```
or <- (De/He)/(Dn/Hn) # Disease exposed, Healthy non-exposed
logo <- log(or)
x <- sqrt(((1/De) + (1/He) + (1/Dn) + (1/Hn)))
lower_ci = exp(logo - 1.96*x)
upper_ci = exp(logo + 1.96*x)
cat("OR:", round(or, 3), "(", round(lower_ci, 3), "-", round(upper_ci, 3), ")",
    spe = "")
```
for instance,
```
De <-6
Dn <-3
He <-4
Hn <-5
or <- (De/He)/(Dn/Hn)
logo <- log(or)
x <- sqrt(((1/De) + (1/He) + (1/Dn) + (1/Hn)))
lower_ci = exp(logo - 1.96*x)
upper_ci = exp(logo + 1.96*x)
cat("OR:", round(or, 3), "(", round(lower_ci, 3), "-", round(upper_ci, 3), ")",
    spe = "")
```
Is there a simple function from some package that can also add a
p-value to this test? Or how can I calculate the p-value on my own?
Sorkin, John
# 
Luigi,
Odds ratios can be produced using a logistic regression, which can be performed using the glm function. The following has a detailed description of how logistic regression can be performed using R:

https://stats.idre.ucla.edu/r/dae/logit-regression/

John


John David Sorkin M.D., Ph.D.
Professor of Medicine
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology and Geriatric Medicine
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)
Michael Dewey
# 
Dear Luigi

You could try the epitools package which gives a large number of ways of 
doing this. I would have thought that using Wald intervals for the log 
odds ration was not optimal with small frequencies.

Michael
On 06/07/2020 14:01, Luigi Marongiu wrote:

  
    
  


  


      

    

    

      
      

        


  

        

      Stefan Evert
    

    

      
      #
    

  


  

      
fisher.test() computes exact confidence intervals for the odds ratio.

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Minato Nakazawa
    

    

      
      #
    

  


  

      
A review (sorry in Japanese) on the calculation of odds ratios 
with confidence intervals using several packages in R is given
by Prof. Okumura, Mie Univ.

https://oku.edu.mie-u.ac.jp/~okumura/stat/2by2.html

x <- matrix(c(6, 3, 4, 5), 2)
# using vcd
library(vcd)
res <- oddsratio(x)
exp(confint(res))
summary(res)
# Note: summary(oddsratio(x, log=FALSE)) gives inappropriate p-value.
# using fmsb
library(fmsb)
oddsratio(x, p.calc.by.independence=FALSE)

Best,
Minato Nakazawa

On Mon, 6 Jul 2020 15:01:34 +0200

        
Luigi Marongiu <marongiu.luigi at gmail.com> wrote: