Kimberly Shoenbill, a Clinical Investigation PhD student, used statistical and machine learning methods to identify factors that affect Institutional Review Board (IRB) review times. Shoenbill was mentored by Eneida Mendonça, a key member of the UW ICTR Biomedical Informatics group, and collaborated with the UW Health Sciences (HS) IRBs office for her study. Their work was recently published in the Journal of Clinical and Translational Science and has the potential to be widely applicable across the CTSA consortium and beyond. Mendonça notes,
Kim’s study uses machine learning and statistical methods to analyze the root causes of variability in the IRB process. IRB approval delays and obstructions in the initiation of research studies are a major source of frustration for investigators and are likewise a major target of federal government efforts to improve research efficiency.
IRB processing times decreased at UW from an initial 2011-2012 interval to a subsequent 2013-2014 interval. This trend allowed Shoenbill to identify and analyze data key factors that impact IRB protocol processing times. Key factors that influence time to approval include whether the study underwent full board vs. expedited IRB review, if the study fell under VA purview, whether the study underwent scientific review, and month submission sent to the IRB, among others. While the specific factors in this study may vary over time or be unique to a specific institution’s IRB, the study’s methods are applicable to analysis of other institutional review workflows.
Nichelle Cobb, director of the HS-IRBs office and a co-author on the study, notes,
The information we obtained from these novel approaches provided empiric support for our ongoing quality improvement activities and confirmed many of the assumptions we had about the factors that influence turnaround time and how to address these.
Shoenbill is now an assistant professor in the Department of Family Medicine and the Program on Health and Clinical Informatics at the University of North Carolina – Chapel Hill. She was a post-doctoral fellow in the Computation and Informatics in Biology and Medicine (CIBM) training program from 2012-2015.