Time-to-event analysis, or who gets better sooner? An emerging concept in headache study methodology.

Survival analysis, or, more generally, time-to-event analysis, is of interest when the data represent the time to a defined event. While well established in oncology, it has not been widely applied to migraine research, possibly because the data are usually collected intermittently, rather than continuously, and because of the awkwardness of interpreting treatment effect in survival terms. However, it represents an interesting approach for the analysis of time-to-headache relief, which addresses the clinically relevant question of who gets better sooner. The analysis uses data from all time-points to define the likelihood of headache relief following treatment throughout the entire assessment period. These data can then be used to quantify and test the difference between two therapies.