Question 2: What is meant by "P=0.15" and what can we conclude from it?
This is the result of a significance test, testing the null hypothesis that the odds ratio in the population is one, i.e. that there is no difference in the odds of an adverse outcome between patients on statins and patients who were not on statins.
The P value is the probability of getting an odds ratio as far or further from what we would expect as is the one we have observed, if the null hypothesis were true.
As this P value is greater than 0.05, we would conclude that the data are not consistent with the null hypothesis. There is little or no evidence for us to conclude that the odds of a bad outcome are different in patients on statins and patients not on statins. Because P > 0.10, we do not even have very weak evidence.
Incidently, you may wonder why I keep asking the same questions about confidence intervals and P values. It is because I want you to become so familiar with these concepts that you don't have to think about them!
Back to Exercise: Statins in pneumonia.
To Applied Biostatistics index.
To Martin Bland's M.Sc. index.
To Martin Bland's home page.
This page maintained by Martin Bland.
Last updated: 16 November, 2006.
Back to top.