|Thursday, February 18|
|PS1 Poster Session 1 & Opening Mixer sponsored by SAS||
Thu, Feb 18, 5:30 PM - 7:00 PM
Use of Multivariate Data Analysis in Optimization of Risk-Based Monitoring of Multicenter Trials (303257)*Xiaoqiang Xue, Unaffiliated
Keywords: RIsk based monitoring, advanced and predictive analytics, multivariate data analysis
Following Guideline from FDA and EMEA, Risk Based Monitoring (RBM) aims to enhance human subject protection and clinical trial data quality while reducing full SDV (Source Data Verification). Our approach is based on the multivariate data analysis using Mahalanobis’ distance and mixed effect model (including Gamma-Poisson, Beta-Binomial, Gamma-Exponential or Pareto). The objectives are to find possible data errors such as outliers, values too close or too far away from the means by examining all information accumulated through the selected variables, and derive a decision rule that minimizes the risk function associated with randomized monitoring. Our major target is to monitor the clinical trial execution by both patient clinical data and operational data, including, enrollment, query, action items data, to ensure patient safety and protocol compliance during clinical trials.