BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250508T120658EDT-0759iudUh7@132.216.98.100
DTSTAMP:20250508T160658Z
DESCRIPTION:Shu Yang\, PhD\n\nAssociate Professor\n	Department of Statistics
  |\n	NC State University\n\nWHEN: Wednesday\, November 6\, 2024\, from 3:30
  to 4:30 p.m.\n	WHERE: Hybrid | 2001 ºÃÉ«TVl College Avenue\, Room 1201\; Zo
 om\n	NOTE: Shu Yang will be presenting from North Carolina\n\nAbstract\n\nR
 andomized controlled trials (RCTs) are the gold standard for causal infere
 nce on treatment effects\, but they can be limited by small sample sizes d
 ue to the indications associated with rare diseases and small patient popu
 lations\, where ethical concerns or patient reluctance may limit control g
 roup assignment. Hybrid controlled trials use external controls (ECs) from
  historical studies or large observational databases to enhance statistica
 l efficiency. However\, non-randomized ECs can introduce biases that compr
 omise validity and inflate Type I errors for treatment discovery\, particu
 larly in small samples. To address this\, we extend the Fisher randomizati
 on test to hybrid controlled trials. Our approach involves a test statisti
 c combining RCT and EC data and is based solely on randomization in the RC
 T. This method strictly controls the Type I error rate\, even with biased 
 ECs\, and improves power by incorporating unbiased ECs.\n	\n	To mitigate the
  power loss caused by biased ECs\, we introduce Conformal Selective Borrow
 ing\, which uses individual conformal p-values to selectively incorporate 
 unbiased ECs\, offering the flexibility to use either computationally effi
 cient parametric models or off-the-shelf machine learning models to constr
 uct the score function\, along with model-agnostic reliability. We identif
 y a risk-benefit trade-off in the power of FRT\, associated with different
  selection thresholds for conformal p-values\, analogous to the mean squar
 ed error trade-offs observed in the data integrative estimators. We propos
 e a data-driven selection of the threshold value to achieve robust perform
 ance across different levels of hidden bias. The advantages of our method 
 are demonstrated through simulations and an application to a small-sized l
 ung cancer trial with ECs from the National Cancer Database.\n\nSpeaker bi
 o\n\nShu Yang is Associate Professor of Statistics at North Carolina State
  University. She received her Ph.D. in Applied Mathematics and Statistics 
 from Iowa State University and postdoctoral training at Harvard T.H. Chan 
 School of Public Health. Her primary research interest is causal inference
  and data integration\, particularly with applications to comparative effe
 ctiveness research in health studies. She also works extensively on method
 s for missing data and spatial statistics. She has been Principal Investig
 ator for several U.S. NSF\, NIH\, and FDA research projects. For more info
 rmation\, please visit https://shuyang.wordpress.ncsu.edu/.\n
DTSTART:20241106T203000Z
DTEND:20241106T213000Z
SUMMARY:Enhancing Statistical Validity and Power in Hybrid Controlled Trial
 s: A Randomization Inference Approach with Conformal Selective Borrowing
URL:/epi-biostat-occh/channels/event/enhancing-statist
 ical-validity-and-power-hybrid-controlled-trials-randomization-inference-a
 pproach-360430
END:VEVENT
END:VCALENDAR