Prevalence and Factors Associated with Patient-Reported Outcomes in Pragmatic Randomized Controlled Trials

SHELLEY VANDERHOUT, PhD, RD
University of Ottawa

Patient-reported outcomes (PROs) are subjective measures of health and well-being that come directly from patients and commonly used to measure patient experience, quality of life, and symptoms. Given that pragmatic trials aim to provide evidence to inform clinical care decision making, PROs seem well suited to pragmatic trials; however, their use and reporting in pragmatic trials have not been described. We sought to review pragmatic trials to describe (1) the prevalence and types of PROs used; (2) whether the use of PROs varied across trial characteristics; and (3) how sample sizes and target differences were determined for trials with PROs.

An electronic search filter in MEDLINE was used to identify primary reports of pragmatic randomized controlled trials in health research published 2014-2019, that were registered in ClinicalTrials.gov and self-identified as pragmatic. Trial descriptors were downloaded from ClinicalTrials.gov; information about PROs and sample size calculations were extracted from each report. Data were summarized descriptively. Chi-squared, Cochran-Armitage and Wilcoxon rank-sum tests were used to examine associations between trial characteristics and use of PROs.

Of 415 trials which met inclusion criteria, 235 (57%) measured PROs (144 (35%) at least as primary/co-primary and 91 (22%) as only secondary outcomes). Primary PROs were symptoms (64; 44%), health behaviours (36; 25%), quality of life (8; 13%), functional status (16; 11%), and patient experience (10; 7%). Studies published in higher impact journals or funded by industry were less likely to use PROs as primary/co-primary outcomes, whereas individually (vs. cluster) randomized studies, those conducted in Europe, and those which tested dietary or behavioural interventions were more likely to use PROs as primary/co-primary outcomes. Patient engagement was not associated with use PROs as primary/co-primary outcomes. For the 144 trials with a PRO as primary or co-primary outcome, 126 (88%) reported a sample size calculation for that outcome. No justification was provided for the target difference in 53 (42%); patient or stakeholder opinion was rarely used to justify the target difference (8, 6%).

PROs are not routinely selected as outcomes in pragmatic trials, and patient and stakeholder engagement in determining target differences and sample sizes is rare. Institutions, funding bodies, and scientific journals can encourage the use of PROs in pragmatic trials by creating incentives, providing methodological support, and establishing policies for pragmatic trialists.