Dissertation/Thesis Abstract

The randomized controlled trial under investigation: Threats to validity posed by participant preferences
by Floyd, Anna Hillary Landis, Ph.D., State University of New York at Stony Brook, 2009, 190; 3401703
Abstract (Summary)

Although the randomized controlled trial (RCT) represents the gold standard of experimental design, its use with human participants can result in threats to both internal and external validity, compromising the intended rigor of randomization. These threats can result from preferences for a particular trial condition, and are most likely to occur in studies in which participants cannot be blinded to their condition, or when certain treatment options are unavailable outside of trials. Although research designs that take participant preferences into account have been developed (i.e., preference controlled trials, or PCTs), little research has been directed at understanding how preferences affect participant experiences and outcomes in RCTs. I designed a trial using a non-patient, undergraduate student population to directly compare participants in a RCT arm of a study to those in a PCT arm of the same study (a Wennberg design). The trial intervention was presented as a treatment that might improve performance on college examinations. This study investigated: (1) the effects of the prospect of being randomized on study accrual; (2) the interaction of participant preferences and treatment assignment on participant feelings about the research, belief in the effectiveness of treatment, intervention contamination, intervention adherence and engagement, trial attrition, and outcomes (examination scores); and (3) the effects of having a choice of treatment on participant feelings, belief in the effectiveness of treatment, intervention contamination, intervention adherence and engagement, trial attrition, and outcomes. Approximately 25% of eligible participants signed up for the study, and among those, 93% enrolled in the trial. Accrual rates did not differ between the RCT and PCT arms of the study. Participants preferred the treatment group to the control group in a 2:1 ratio, and those who were mismatched to their preferred condition had significantly more negative feelings about participating in the research, including greater feelings of anger about the study, regret in having signed up for the study, and envy towards other participants, as well as less positive feelings about being involved in the study, and were less convinced that the study would provide useful information. These effects were greater among participants holding strong, as opposed to weak, preferences for condition. Being mismatched to preference did not significantly affect belief in the effectiveness of treatment, adherence or engagement in the trial, nor did it affect examination scores. Among participants who were assigned their preferred condition, being matched by choice (as in a PCT) as compared to by chance (as in a RCT) did not affect their feelings about being involved in the trial, belief in the effectiveness of treatment, adherence or engagement, contamination, attrition, or their examination score.

This study provides important information regarding participant feelings about being in a RCT. Negative feelings about being randomized to a non-preferred treatment arose even in this non-patient sample, whereby the outcome of the intervention (examination score) was important to participants, but was not related to their morbidity or mortality. It is quite likely that the effects could be more pronounced in clinical trials with patient samples whereby morbidity or mortality might be affected. Although this study did not find that such negative feelings affected intervention adherence, or outcomes, it is possible that findings would be different in clinical trials of medical interventions, where participants and potential participants may be more invested in having access to a particular treatment condition. This study indicates that, at the very least, participant feelings are affected by the randomization process—which holds potentially important ethical implications for patients participating in clinical trials.

Indexing (document details)
Advisor: Moyer, Anne
School: State University of New York at Stony Brook
School Location: United States -- New York
Source: DAI-B 71/04, Dissertation Abstracts International
Subjects: Social psychology, Experimental psychology
Keywords: Clinical trials, Human subjects, Randomized controlled trials, Research design, Treatment preference
Publication Number: 3401703
ISBN: 978-1-109-69046-0
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy