Confidence interval

Measured variable that indicates the range in which the true value of a measurement (e.g. therapy effect) is assumed with a specified probability (e.g. 95%). The width of the confidence interval is called the precision of the estimate. It depends on the desired certainty of the estimate (e.g. 95%), the "sample size" and the "standard error" of the sample statistics, i.e. it depends, among other things, on the number of patients included and the result rate. The confidence interval indicates at a glance whether a result is significant (example: relative risk: 0.85; 95% confidence interval: 0.7-0.9; statistically significant, since the entire interval lies below the number 1 (line of no effect). The confidence interval is also a measure of the accuracy of the result: the narrower the confidence interval, the more accurate the estimate.