Do Assigned Roles in Clinical Trials Affect Patient-Reported Outcomes?

Image Credit: © Andrii Yalanskyi - stock.adobe.com

A study recently published on JAMA Network Open examined a single center, double-blind, three-group randomized clinical trial for migraine to investigate differences in patient-reported outcomes (PROs) based on role expectancy.¹

“It is usually assumed that an individual’s classification as a patient or a healthy person is determined by the presence or absence of disease, but little is known about whether the mere awareness of being a patient or a healthy control can play an important role for reporting outcomes,” the study authors wrote.

The trial highlighted in this study included consecutive patients from a tertiary headache clinic based at a single center in Germany. They were invited to participate between October 2019 and June 2023. Patients in the trial with migraines were randomized into two groups—the first was told that the trial was centered on migraine symptoms, while the second was told that healthy controls were needed for a study about patients with vertigo. Participants viewed standardized roller coaster videos and were asked to report their levels of motion sickness and dizziness.

The final sample included 366 participants—122 with migraine assigned the role of patient (MP), 122 with migraine assigned the role of healthy participant (MH), and 122 headache-free controls (HC).

The assigned role of the participants with migraine had a significant effect on self-disclosure of (1) estimation that symptoms (dizziness) will occur under specific conditions (self-reported vestibular symptoms: [79 MP, 29 MH, 9 HC]), (2) the increase of such symptoms (dizziness) after viewing the roller coaster video, and (3) the reported frequency (median [IQR] self-reported monthly headache days (7 MP [4-15] days; 5 MH [2-10] days)) and severity (median [IQR] migraine disability assessment score (35 MP [20-64] points; 25 MH [11-47] points)) of migraine symptoms.

“The striking feature here is that patients were asked for migraine severity and frequency only after unblinding,” the authors wrote of the results. “This means that patients of both groups were loyal to the adopted role as a patient or a healthy control despite debriefing. This reporting bias could well affect clinical outcome measures where, for example, the difference between medications and placebo effects is small.”

According to the authors, the results of this study presumably mirror a general human behavior that is independent of the definitions of healthy and/or diseased. The findings could also apply to other disease areas where reporting is high subjective. Symptoms including dizziness, pain, nausea, and sensory disturbance all pose challenges to reporting as patients could have varying definitions for each, according to the study. The authors also stress the importance of the clinician-patient relationship, meaning that clinicians must effectively communicate the effects patients’ reports have on future research and how they should behave in a group-compliant manner.

“This study’s findings highlight the importance of being vigilant to the effect of suggestion on study outcomes and should be subject to further discussion in the scientific community as part of the discussion about suboptimal experimental methodology, publication bias, and the need for replication,” the authors concluded. “If the findings are also consistent in other diseases, they probably cannot be avoided in clinical practice but should be known and investigated and also taken into account when, for example, invasive treatments are discussed on the ground of such measures.”

Reference

1. May A, Carvalho GF, Schwarz A, Basedau H. Influence of Role Expectancy on Patient-Reported Outcomes Among Patients With Migraine: A Randomized Clinical Trial. JAMA Netw Open. 2024;7(4):e243223. doi:10.1001/jamanetworkopen.2024.3223