Premium
Clinical trials based on patients' preferences
Author(s) -
Acland J. D.
Publication year - 1964
Publication title -
clinical pharmacology and therapeutics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.941
H-Index - 188
eISSN - 1532-6535
pISSN - 0009-9236
DOI - 10.1002/cpt196456part1687
Subject(s) - preference , clinical trial , affect (linguistics) , argument (complex analysis) , medicine , psychology , statistical significance , statistics , mathematics , communication
In many clinical trials, the relative efficacy of drugs has to be assessed by means of subjective measures based on the preferences expressed by the subjects. Patients are often unable to distinguish between two or more alternative forms of treatment, and a proportion of “no preference” decisions may consequently be recorded. Contrary to my previous argument,1 the proportion of “no preference” decisions does not affect the significance level attained in statistical tests based on the relative proportions of positive decisions only. It is maintained, however, that the proportion of sUch decisions alters the weight to be attached to a given difference between the numbers of positive choices for alternative treatments. As a consequence, it is argued that, in sequential clinical trials based on patients' preferences, different designs must be chosen according to the different proportions of “no preference” decisions which are expected to occur.