Coenrollment in a critical care trial: Characteristics and consequences

Abstract

Background

Objectives of this study were to determine the characteristics of patients, centers and studies involved in coenrollment, and the association of coenrollment with trial metrics and patient outcomes.

Methods

This pre-planned study within a stress ulcer prophylaxis trial testing pantopazole used descriptive analyses and multilevel regression analysis to examine patterns and predictors of coenrollment among patients in an intensive care unit (ICU).

Results

Among 4821 trial participants, 1719 (35.7 %) were coenrolled in at least one of 145 unique studies. There were 2167 coenrollment events. The most common design of coenrolled studies were individual-patient randomized trials, followed by cluster randomized trials and platform trials. Most coenrollment involved investigator-initiated studies (1924, 88.8 %). Patients with SARS-CoV-2 infection were more likely to be coenrolled than others (odds ratio 1.85 (95 % confidence interval, 1.50, 2.29), p < 0.001). Research coordinators with mid-senior trial experience were more likely to coenrol than others. Coenrolled patients were more likely to miss study drug (median 1 dose, IQR 1–2 doses) compared to others (202 (11.8 %) versus 221 (7.1 %), p < 0.001). Coenrollment did not influence the effect of pantoprazole on gastrointestinal bleeding or 90-day mortality.

Conclusions

In the REVISE trial, one-third of participants were coenrolled, primarily into another academic randomized trial. Patients with SARS-CoV-2 were more likely to be coenrolled than other patients. Experienced research coordinators were more likely to coenrol than other personnel. Coenrollment did not modify the treatment effect of pantoprazole on the primary trial outcomes.

Clinical trial registration

www.clinicaltrials.gov NCT03374800