Intention-to-Treat Analysis but for Treatment Intention:How should Consumer Product Randomized Controlled Trials be Analyzed?
Pages 90-98
Rolf Weitkunat, Gizelle Baker and Frank Lüdicke
DOI: http://dx.doi.org/10.6000/1929-6029.2016.05.02.3
Published: 02 June 2016

Abstract: Background: Experimental study design, randomization, blinding, control, and the analysis of such data according to the intention-to-treat (ITT) principle are de-facto “gold standards” in pharmacotherapy research. While external treatment allocation under conditions of medical practice is conceptually reflected by in-study randomization in randomized controlled trials (RCTs) of therapeutic drugs, actual product use is based on self-selection in a consumer product setting.

Discussion: With in-market product allocation being consumer-internal, there is no standard against which protocol adherence can be attuned, and the question arises, as to whether compliance-based analysis concepts reflect the real-world effects of consumer products.

Summary: The lack of correspondence between RCTs and consumer market conditions becomes evident by the fact that even if, theoretically, all data would be available from all members of the real-world target population, it would be impossible to calculate either an ITT or a per-protocol effect. This renders the calculation of such estimates meaningless in consumer product research contexts.