Dropout rate for intent-to-treat and per protocol analye

Purpoe To decribe dropout rate for the intent-to-treat and per protocol analye from propective clinical trial. Method Review of propective multi-center parallel tudie of 100 patient or more from 1996 onward. Reult We identified 33 article (70 treatment arm) that fit the criteria for thi tudy. No tatitical difference in dropout rate were oberved among drug clae for either the intent-to-treat (P = .075) or per protocol analye (P = .40). A difference wa oberved in the percent dropout rate for the intent-to-treat analye decreaing with the length of the tudy (P < .0001). Thi finding wa not oberved by the number of tudy viit (P = .44). However, a tatitically greater percent dropout rate wa oberved for the per protocol analye increaing with the length of the tudy (P = .034) and number of tudy viit (P = .01). No tatitical difference were oberved or with increaing ample ize of the tudy for either the intent-to-treat or per protocol analye (P > .05). Concluion Known dicontinuation rate for per protocol and intent-to-treat analye may help in planning ample ize for future clinical trial.