Describes a method for computing a condence interval for the odds ratio. What might be a concern about this method?
Assuming the table is:
Disease | No Disease | |
Case | a | c |
Control | b | d |
The Odds ratio is calculated as:
OR = (a/b) / (c/d)
To calculate the confidence interval, the formula is:
Here, 1.96 is the z-critical value at alpha = 0.05 significance level
The standard error is calculated as follows:
Hence, we now know how to calculate the confidence interval of an odds ratio, by first calculating the standard error and then calculating the confidence interval.
The concern about the calculations using this method is that when any of the values a or b or c or d is zero, this method might not work as we might not be able to calculate the standard error.
Get Answers For Free
Most questions answered within 1 hours.