Intention-to-treat analysis

The intention-to -treat analysis is a method of medical statistics that identifies a particular data analysis principle ( often shortened in German medical journals as well as intention-to -treat called ).

It is a principle analysis of data collected in controlled randomized clinical trials. In general, a comparison of efficacy between drug (verum ) and placebo or sham intervention and treatment is carried out in these studies at a particular therapy. In this case, all patients who were included in the active treatment group or the placebo group at the beginning of the study, look at the end and in the evaluation of results. The data of patients who withdraw during the current study - for example, at the request of the patient or because the need for other therapy was given - so do not participate and the study treatment until the scheduled end of the study or who have changed the treatment group for any reason, must the evaluation are recorded and are thereby evaluated in the group that was assigned to them.

This evaluation principle often first appears somewhat illogical, but most closely resembles the conditions in daily practice. It often happens, for example, the accidental or deliberate omission of prescribed medications.

In contrast, should the patient be removed from the data, which do not take medication, for example, because their expectations were not met, or because they are no longer capable of doing, the study results would be falsified. It must in each case the reasons are always checked that have led to the corresponding patient is excreted from the study. In particular, in cases of death, the question arises whether the death could not be associated with the administered therapy. Doubtful cases are to be considered as treatment failures.

Or in another example illustrates: a clinical study will compare two different diets (diet A and diet B). Both diets have a weight loss goal. To this end, the study participants are divided into two groups ( "study group A" and "study group B"), each of which receives diet A or B. A diet is effective, during Diet B is predominantly does not have the desired effect, both of which are not known, but at the beginning of the study. During the study, mainly those study participants to leave the study, in which the desired effect (weight loss) fails. Thus, the results appear better than they actually are really, mainly because the study participants remain, where the diet has worked in both study arms. This distortion applies to a much greater extent for the study arm B because there especially many study participants leave the study. The difference in effectiveness between diet A and diet B is thus appear smaller than it actually is.

Thus, intention-to -treat means that the data of all patients who are previously intended ( intention ) to treat (to treat), afterwards must also be evaluated. This happens regardless of whether the treatment was carried out actually as planned ( per protocol ).

The opposite of this process is the per-protocol analysis, in which only the data of those patients are evaluated, which behaved prüfplankonform ( ie were actually treated in the strictest sense with the proposed intervention or medication in the test report ). A subspecies of the per-protocol analysis is the as treated analysis. Here, the data are analyzed for all those subjects who have received in the course of the study to be tested intervention or medication, regardless of which study group they were originally assigned.