BEST-CLI Site Performance Comparison Between Multidisciplinary Sites and Non-Multidisciplinary Sites
Maria Fernanda Villarreal, MD1, Susan Assmann2, Matthew Menard, MD3, Kenneth Rosenfield, MD4, Flora S. Siami, MPH2, George Sopko, MD, MPH5, Diane Reid, MD5, Michael Strong3, Alik Farber, MD1.
1Boston Medical Center, Boston, MA, USA, 2New England Research Institutes, Inc. (NERI), Watertown, MA, USA, 3Brigham and Women’s Hospital’s (BWH), Boston, MA, USA, 4Massachusetts General Hospital, Boston, MA, USA, 5National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health (NIH), Bethesda, MD, USA.
INTRODUCTION
There is limited evidence regarding the role of multidisciplinary teams in vascular clinical trials. Best Endovascular versus Best Surgical Therapy for Patients with Critical Limb Ischemia (BEST-CLI), a randomized trial comparing surgical and endovascular revascularization strategies, is structured to foster multidisciplinary collaboration. To evaluate the impact of collaboration, we compared enrollment rates at multidisciplinary and non-multidisciplinary sites.
METHODS
We analyzed 132 BEST-CLI sites open for enrollment. Multidisciplinary sites had investigators belonging to at least 2 different specialties (vascular surgery and interventional cardiology or radiology). Sites having only vascular surgery investigators were considered single-disciplinary. To test our hypothesis, a linear regression model of monthly enrollment rate was constructed considering the following characteristics: multidisciplinary vs. single-disciplinary, and total number of investigators.
RESULTS
Ninety (68%) sites were multidisciplinary and 42 (32%) sites were single-disciplinary. At time of analysis these sites had enrolled 1345 patients. The mean and median monthly enrollment at multidisciplinary sites were 0.30 and 0.24 (range 0 - 1.77). The mean and median monthly enrollment at single-disciplinary sites were 0.27 and 0.24 (range 0 - 1.30). In the regression model, there was no significant difference in monthly enrollment rates between multidisciplinary and single-disciplinary sites; after adjustment for the total number of investigators, multidisciplinary sites averaged 0.02 fewer enrolled
per month (95% CI -0.13 - 0.09, p=0.72). Monthly enrollment rates were significantly higher for sites with more investigators (0.02 higher for each additional investigator, (95% CI 0.00 - 0.03, p=0.0497)
CONCLUSION
Clinical enrollment rates in randomized trials are affected by multiple factors. After adjusting for the total number of investigators, we found no significant relationship between multi-disciplinary status and monthly enrollment rates. Monthly enrollment rate was positively correlated with number of site investigators. Given the importance of enrollment for trial execution, further evaluation of factors that impact enrollment is crucial.
Back to 2019 Abstracts