BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20251121T152639EST-5243DUjtVo@132.216.98.100
DTSTAMP:20251121T202639Z
DESCRIPTION:Title:\n\nMarginal Meta-Analysis for Combining Multiple Randomi
 zed Clinical Trials with Rare Events – Lessons Learned from Avandia Story
 \n\n \n\nAbstract:\n\nMeta analysis (MA) is commonly used in the post-mark
 eting safety studies for FDA regulated medical products\, including drugs\
 , medical device\, and etc. Avandia Studies (Nissen et al\, 2007\, 2010) i
 s a powerful example to show how important MA is in real life for quantify
 ing the safety concerns with policy impacts. However\, the fact that the r
 e-analysis of same Avandia data could reach different conclusions showed c
 learly the statistical challenges and difficulties associated with standar
 d fixed effect and random effect MA methods. Specifically\, the inclusion 
 and exclusion of zero trials\, changing the effect estimand to risk differ
 ence\, and/or using other fixed effect MA methods rather than Peto\, would
  all lead to different results. Lesson learned from Avandia studies inspir
 ed our discovery of the problems associated with “homogeneous effects” or 
 “effect at random” assumption – the validity assumption underlying standar
 d MA approaches\, and led to a set of more relaxed Study at Random assumpt
 ions. Additionally\, two more concerns motivated our research on this marg
 inal meta analysis: (1)\, rare events in safety studies often lead to low 
 power in homogeneity test associated with standard MA approaches. Even tho
 ugh they may bias the results\, various types of add-hoc continuation corr
 ections were proposed and widely used to improve the performance of standa
 rd MA estimators. (2) Non-collapsibility issues associated with odds ratio
  limit the interpretability of many popular MA estimators too. As a result
 \, based on the new flexible study homogeneity assumption\, we proposed a 
 marginal meta analysis approach with natural weights which provided a cons
 istent treatment effect estimate for marginal causal effects combining ran
 domized clinical trials in safety studies. This estimator is particularly 
 useful when the outcome is rare\, and double zero trials are naturally acc
 ounted in the estimation without any add-doc continuity correction. System
 atic simulation studies show that the proposed estimator performs reasonab
 ly well under different rationales. This method is re-applied in Avandia s
 afety evaluation as a real case application. This is a joint work with my 
 students\, Elande Baro\, Yun-Yu Cheng\, and colleague from FDA\, Guoxing S
 oon.\n\nYi Huang1\, Elande Baro2\, Yun-Yu Cheng1\, Guoxing Soon2\n	1: Dept.
  of Mathematics and Statistics\, University of Maryland\, Baltimore County
 \n	2: Office of Biostatistics\, OTS\, CDER\, US FDA\n	Keywords: Meta Analysi
 s\, Rare Events\, Homogeneity Assumptions\, Effect at Random\, Avandia\, Z
 ero trials.\n
DTSTART:20170516T161500Z
DTEND:20170516T171500Z
LOCATION:Room 25\, Purvis Hall\, CA\, QC\, Montreal\, H3A 1A2\, 1020 avenue
  des Pins Ouest
SUMMARY:Dr Yi Huang. University of Maryland\, Baltimore
URL:/mathstat/channels/event/dr-yi-huang-university-ma
 ryland-baltimore-267948
END:VEVENT
END:VCALENDAR