Abstract
Participant-centered analysis involves applying the customary methods of statistical decision making at the level of the individual research participant. Consequently, each individual is declared a responder who benefited, a nonresponder, or possibly a responder who was harmed, using intensively collected data that were specific to that individual. There are several implications of the participant-centered approach. More data actually relevant to the important outcome need to be collected on individuals. The study results can be summarized in a simple table of responders/nonresponders by treatment group, and probabilities of true response can be estimated. The actual nature of the data collected and the statistical models used to analyze them drop into the background. Finally, production of individual-level decisions permits standard statistical approaches to be applied to the issue of which modality should be recommended for which person, instead of focusing on average effects and which modality should be recommended for everyone.
Get full access to this article
View all access options for this article.