The overall purpose of this methodological study was to investigate the strengths and limitations of the randomized clinical trial design in examining the influence of treatment preferences on outcomes. The study was a secondary analysis of data obtained in two randomized clinical trials that evaluated behavioral therapies for insomnia. In both trials, the same design and methods were used to assess participants' treatment preferences and outcomes, however, the treatments differed. The results illustrated the challenges encountered in using the randomized clinical trial design. The challenges were related to the unbalanced distribution of participants with preferences for the study treatments, non-comparability of the subgroups with treatments matched or mismatched to their preferences, differential attrition, which compromised the sample size and composition of the subgroups and limited the use of the planned statistical analyses. Whether these challenges occur in trials of other types of treatments and target populations should be explored in future research. Some strategies were proposed and should be evaluated for their utility in addressing these challenges.
Keywords: Treatment preferences; attrition; challenges; randomized clinical trial; sample characteristics; treatment match–mismatch.