Risk-Based Monitoring and Fraud Detection in Clinical Trials Using JMP and SAS

Improve efficiency while reducing costs in clinical trials with centralized monitoring techniques using JMP and SAS. International guidelines recommend that clinical trial data should be actively reviewed or monitored; the well-being of trial participants and the validity and integrity of the final analysis results are at stake. Traditional interpretation of this guidance for pharmaceutical trials has led to extensive on-site monitoring, including 100% source data verification. On-site review is time consuming, expensive (estimated at up to a third of the cost of a clinical trial), prone to error, and limited in its ability to provide insight for data trends across time, patients, and clinical sites. In contrast, risk-based monitoring (RBM) makes use of central computerized review of clinical trial data and site metrics to determine if and when clinical sites should receive more extensive quality review or intervention. Risk-Based Monitoring and Fraud Detection in Clinical Trials Using JMP and SAS presents a practical implementation of methodologies within JMP Clinical for the centralized monitoring of clinical trials. Focused on intermediate users, this book describes analyses for RBM that incorporate and extend the recommendations of TransCelerate Biopharm Inc., methods to detect potential patient-or investigator misconduct, snapshot comparisons to more easily identify new or modified data, and other novel visual and analytical techniques to enhance safety and quality reviews. Further discussion highlights recent regulatory guidance documents on risk-based approaches, addresses the requirements for CDISC data, and describes methods to supplement analyses with data captured external to the study database. Given the interactive, dynamic, and graphical nature of JMP Clinical, any individual from the clinical trial team - including clinicians, statisticians, data managers, programmers, regulatory associates, and monitors - can make use of this book and the numerous examples contained within to streamline, accelerate, and enrich their reviews of clinical trial data. The analytical methods described in Risk-Based Monitoring and Fraud Detection in Clinical Trials Using JMP and SAS enable the clinical trial team to take a proactive approach to data quality and safety to streamline clinical development activities and address shortcomings while the study is ongoing. This book is part of the SAS Press Contents About This Book Purpose Is This Book for You? Prerequisites About the Examples Software Used to Develop the Book's Content Example Code and Data Output and Graphics Used in This Book Additional Resources Keep in Touch About the Author Acknowledgments Introduction 1.1 Overview 1.2 Topics Addressed in This Book 1.2.1 Risk-Based Monitoring 1.2.2 Fraud Detection 1.2.3 Snapshot Comparisons 1.3 The Importance of Data Standards 1.4 JMP Clinical 1.5 Clinical Trial Example: Nicardipine 1.6 Organization of This Book References Risk-Based Monitoring: Basic Concepts 2.1 Introduction 2.2 Risk Indicators 2.2.1 Individual Risk Indicators 2.2.2 Overall Risk Indicators 2.2.3 Default Risk Thresholds 2.2.4 Default Actions 2.3 Geocoding Sites 2.4 Reviewing Risk Indicators 2.4.1 Site-Level Risk 2.4.2 Country-Level Risk 2.4.3 Subject-Level Risk 2.5 Final Thoughts References Appendix Walk-through of the RBM Reports Definitions of Risk Indicators and Important Terms in Pseudo-code Risk-Based Monitoring: Customizing the Review Experience 3.1 Introduction 3.2 Defining Alternate Risk Thresholds and Actions 3.2.1 Thresholds for Individual and Overall Risk Indicators 3.2.2 Thresholds for User-Added Risk Indicators 3.2.3 Weights for Overall Risk Indicators 3.2.4 Defining Actions for Elevated Risk 3.3 Performing Additional Statistical and Graphical Analyses 3.3.1 Statistical Analyses 3.3.2 Graphing 3.4 Creating JMP Scripts and Add-Ins 3.4.1 New Discontinuation Variables and Risk Thresholds 3.4.2 New Adverse Event Variables and Figures 3.4.3 Analyses at the Monitor Level 3.5 Final Thoughts References Detecting Fraud at the Clinical Site 4.1 Introduction 4.2 Study Visits 4.2.1 Weekdays and Holidays 4.2.2 Study Scheduling 4.3 Measurements Collected at the Clinical Site 4.3.1 Tests with No Variability 4.3.2 Duplicate Sets of Measurements 4.3.3 Digit Preference 4.3.4 A Brief Interlude from the Fraud Detection Menu 4.4 Multivariate Analyses 4.4.1 Multivariate Inliers and Outliers 4.4.2 Hierarchical Clustering of Subjects Within Clinical Sites 4.5 Final Thoughts References Detecting Patient Fraud 5.1 Introduction 5.2 Initials and Birthdate Matching 5.3 Hierarchical Clustering of Pre-Dosing Covariates Across Clinical Sites 5.4 Review Builder: Quality and Fraud 5.5 Final Thoughts References Snapshot Comparisons 6.1 Introduction 6.2 Domain Keys 6.3 Review Flags 6.3.1 Record-Level 6.3.2 Patient-Level 6.4 Adding and Viewing Notes 6.5 Using Review Flags 6.5.1 The Domain Viewer 6.5.2 Demographic Distribution 6.5.3 Patient Profiles 6.5.4 AE Distribution 6.6 Final Thoughts References Final Thoughts A Work in Progress Stay in Touch Recommended Reading Index This book presents a practical implementation of methodologies within JMP Clinical for the centralized monitoring of clinical trials. Focused on intermediate users, this book describes analyses for RBM that incorporate and extend the recommendations of TransCelerate Biopharm Inc., methods to detect potential patient-or investigator misconduct, snapshot comparisons to more easily identify new or modified data, and other novel visual and analytical techniques to enhance safety and quality reviews. It highlights recent regulatory guidance documents on risk-based approaches, addresses the requirements for CDISC data, and describes methods to supplement analyses with data captured external to the study database. Given the interactive, dynamic, and graphical nature of JMP Clinical, any individual from the clinical trial team--including clinicians, statisticians, data managers, programmers, regulatory associates, and monitors--can make use of this book and the numerous examples contained within to streamline, accelerate, and enrich their reviews of clinical trial data. The analytical methods described in this book enable the clinical trial team to take a proactive approach to data quality and safety to streamline clinical development activities and address shortcomings while the study is ongoing. -- Edited summary from book