Confidence intervals and p-values in clinical decision making

Abstract

Clinical trials are usually performed on a sample of people drawn from the population of interest. The results of a trial are, therefore, estimates of what might happen if the treatment were to be given to the entire population of interest. Confidence intervals (CIs) provide a range of plausible values for a population parameter and give an idea about how precise the measured treatment effect is. CIs may also provide some useful information on the clinical importance of results and, like p-values, may also be used to assess 'statistical significance'. Although other CIs can be calculated, the 95% CI is usually reported in the medical literature. In the long run, the 95% CI of an estimate is the range within which we are 95% certain that the true population parameter will lie. Despite the usefulness of the CI approach, hypothesis testing and the generation of p-values are common in the medical literature. The p-value is often used to express the probability that the observed differences between study groups are due to chance. p-values provide no information on the clinical importance of results. © 2008 Foundation Acta Pædiatrica.