Abstract Number:
1427
Submission Type:
Invited Paper Session
Participants:
Jiwei Zhao (1), Joe Ibrahim (2), Jiwei Zhao (1), Shu Yang (3), Ram Tiwari (4), Ilya Lipkovich (5)
Institutions:
(1) University of Wisconsin-Madison, N/A, (2) UNC Chapel Hill, N/A, (3) North Carolina State University, Department of Statistics, N/A, (4) Bristol Myers Squibb, N/A, (5) Eli Lilly and Company, N/A
Chair:
Discussant:
Session Organizer:
Speaker(s):
Shu Yang
North Carolina State University, Department of Statistics
Session Description:
Randomized controlled trials (RCTs) have long been considered the gold standard in clinical research for substantiating the safety and efficacy of treatments or interventions. However, RCTs are marked by their high cost, extended recruitment periods, and ethical and practical constraints, particularly in cases involving rare or life-threatening diseases. In response to the 21st Century Cures Act, the accessibility of real-world data (RWD) from disease registries, electronic health records, and external/historical controls has surged. These sources can now effectively generate fit-for-purpose real-world evidence (RWE) that holds great potential in guiding healthcare and regulatory decision-making.
In this session, we will present the latest research findings regarding the integration of ECs within the context of RCTs. Our primary focus will be on the analysis of the advantages achieved through the judicious utilization of external control data, along with an examination of possible associated risks.
There are potential risks associated with the use of ECs in RCTs, stemming from the possibility of incomparability between ECs and the concurrently recruited controls in the RCT. In statistical terms, these two control groups may originate from significantly different populations. As a result, conducting analyses without due scrutiny could introduce substantial bias into the estimation of treatment effects, leading to misleading conclusions. On the flip side, the advantages of incorporating external controls can be realized and effectively harnessed through careful and accurate utilization of external data information. This can be achieved through a range of statistical and machine learning methods, which include, but are not limited to, pre-testing, the definition of bias functions to quantify differences between these two control groups, and selective information borrowing.
This research holds both significance and timeliness, aligning seamlessly with the overarching theme of JSM 2024: "Statistics and Data Science: Informing Policy and Combatting Misinformation".
This session will feature three distinguished speakers and one accomplished discussant, all of whom are highly active researchers in this field, renowned for their outstanding contributions and exceptional presentation skills. The availability of each participant has been thoroughly confirmed by the organizer. We anticipate that this session will draw a diverse audience with a wide range of backgrounds and extensive expertise, spanning data integration, model transportability, the estimation of causal estimands, as well as semiparametric modeling, across academia, industry, and regulatory agencies.
This session actively promotes diversity among the speakers and the discussant, striving for a well-rounded representation in terms of their affiliations (ensuring a healthy mix of academia and industry), gender (including both female and male perspectives), ethnicity, and their seniority, as defined by their years of experience.
Speaker 1 Title: Leveraging external data to augment a single arm study or a control arm of an RCT using a propensity-score based method with illustrations
Speaker 2 Title: Mitigating Bias in Treatment Effect Estimation: Strategies for Utilizing External Controls in Randomized Trials
Speaker 3: Title: Sensitivity Analysis Framework for Unmeasured Confounding when Borrowing from External Controls in Clinical Trials
Sponsors:
Biometrics Section 3
Biopharmaceutical Section 2
ENAR 1
Theme:
Statistics and Data Science: Informing Policy and Countering Misinformation
Yes
Applied
Yes
Estimated Audience Size
Medium (80-150)
I have read and understand that JSM participants must abide by the Participant Guidelines.
Yes
I understand and have communicated to my proposed speakers that JSM participants must register and pay the appropriate registration fee by June 1, 2024. The registration fee is nonrefundable.
I understand