oddsratio {fmsb} | R Documentation |
Calculate odds ratio and its confidence intervals based on approximation, followed by null-hypothesis (odds ratio equals to 1) testing.
oddsratio(a, b, c, d, conf.level=0.95, p.calc.by.independence=TRUE)
a |
A scalar or a matrix. If matrix, it has to be 2 by 2, which contains the number of individuals who both suffer from exposure and disease as [1, 1], the number of individuals who suffer from disesase but not exposed as [2, 1], the number of individuals who suffer from exposure but are healthy as [1, 2] and the number of individuals who neither suffered from exposure nor disease as [2, 2]. |
b |
If a is a scalar, this has to be given as the number of individuals who suffer from disesase but not exposed. Otherwise, ignored. |
c |
If a is a scalar, this has to be given as the number of individuals who suffer from exposure but are healthy. Otherwise, ignored. |
d |
If a is a scalar, this has to be given as the number of individuals who neither suffered from exposure nor disease. Otherwise, ignored. |
conf.level |
Probability for confidence intervals. Default is 0.95. |
p.calc.by.independence |
Logical. If TRUE, calculating p-value by testing the null-hypothesis of independence between exposure and disease. Otherwise, calculating p-value by inverse-function of confidence intervals calculation (the result becomes the same as the vcd package). Default TRUE. |
estimate |
Calculated point estimate of odds ratio. |
conf.int |
A numeric vector of length 2 to give upper/lower limit of confidence intervals. |
p.value |
The significant probability as the result of null-hypothesis testing. |
This function can also accept a matrix as argument, as suggested by Dr. Toshiaki Ara (toshiaki.ara@gmail.com). Thanks for a good suggestion.
Minato Nakazawa minato-nakazawa@people.kobe-u.ac.jp https://minato.sip21c.org/
Rothman KJ (2012) Epidemiology: An Introduction. 2nd Ed., Oxford University Press, Oxford.
res <- oddsratio(5, 10, 85, 80) str(res) print(res) oddsratio(12, 5, 6, 12) oddsratio(12, 5, 6, 12, p.calc.by.independence=FALSE) DH <- sample(c("Disease", "Health"), 100, replace=TRUE) EN <- sample(c("Exposed", "Nonexposed"), 100, replace=TRUE) x <- table(EN, DH) oddsratio(x) # same as oddsratio(x[1,1], x[2,1], x[1,2], x[2,2])