Combining Individual Patient Data and Aggregated Data in Network Meta-Analysis with Correlated Outcomes

• Activity Number: 616
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Chengxing Lu*
• Keywords: Networking meta-analysis ; Individual Patient Data ; Correlated outcomes

Abstract

Networking meta-analysis (NMA) is a useful tool in many aspects such as designing a new study, assisting marketing strategy, demonstrating treatment benefit for reimbursement purpose and so on. While individual patient data (IPD) is considered as the gold standard in such analyses, the practical challenge is usually researchers find not all of the studies have IPD available. Another challenge may rise while using NMA in help with designing a new study is that not all outcomes of interest were reported across the available studies. This paper aims to address the two challenges by building models combining individual patient level information and aggregated information in one-step approach, while borrowing strength from correlated outcomes for the outcomes of interest that may only have information available from limited studies. An application in ophthalmology will be presented to illustrate the modeling along with simulation studies.