Aims: Intention-to-treat (ITT) analyses are the gold standard in endpoint analyses of randomized clinical trials. Valuable information, however, may be lost in this approach as the actual time on trial medication is not accounted for in patients who withdraw early. Since compliance per se can be a prognostic factor and the actual treatment time is a variable likely to influence clinical outcome, this information should be added to an ITT analysis concept. Thus, the aim is to elucidate the influence of actual treatment time on the results of ITT analysis using available data from a well-conducted, large-scale clinical trial. Methods and results: The ANZ trial, a carvedilol study in heart failure, is characterized by a considerable number of early withdrawals in the carvedilol group. Using the ANZ trial database, the percentage of withdrawals was calculated as well as the number of days on treatment. Fourteen out of 21 deaths (67%) in the carvedilol group occurred after discontinuation of carvedilol. At the time of death, 14 out of 29 (48%) patients in the placebo group had withdrawn from treatment. Mean time of patients on medication who withdrew and died later were 301 days for placebo, but only 204 days for carvedilol. We performed a conventional ITT analysis, and an ITT-based Cox-proportional hazards model (modified ITT) in which the actual time on trial medication was entered into the model as an explanatory variable. The risk ratio in conventional ITT analysis was 0.71 (95% confidence interval (CI) 0.41 to 1.24, p = 0.228) and 0.50 (95% Cl 0.28-0.90, p < 0.001) in modified ITT analysis. Conclusions: Incorporation of the actual treatment time substantially influences the result of this exemplary ITT analysis. The resulting risk ratio is clearly in accordance with clinical implications and in close agreement with those of other P-blocker trials in heart failure. Entering the actual treatment time into a Cox-proportional hazards model as an explanatory variable is reasonable if not necessary from a clinical point of view and thus may strengthen the validity of ITT analysis.
