This is a clinical significance curve for the WHO study (the curve is thick to project better, however obviously thinner curves are required for accurate calculations)..

A clinician can select a benefit which they would accept before implementing treatment, and can see the level of confidence associated with it.

For example if a clinician is only interested in a survival benefit of 15%, you can see that this is associated with a confidence level of only 80%.

This type of curve can be constructed for any comparison, whether it is survival, local control or toxicity, and whether the results are absolute or relative. Alternatively, results can also be tabulated.