Statistical Methods for Clinical Trials with Twin Births

Pennsylvania State Univ Hershey Med Ctr, Hershey PA

Investigators

Linked publications & trials

Abstract

[unreadable] DESCRIPTION (provided by applicant): Analysis of neonatal clinical trials involving twin births is complicated by the correlation that may arise from related individuals. Although the multiple-gestation rate may be cited as part of study entry characteristics, the handling of twin births for randomization or analysis is not commonly discussed. Methods currently exist for analyzing correlated data, but optimal methods for studies that may have as little as 10% of the individuals related have yet to be established. [unreadable] [unreadable] The proposed research aims to raise awareness of twin considerations in neonatal trial design and analysis and provide guidance on the selection of appropriate randomization schemes and statistical methods. The methods will be applied to clinical data from the NICHD Neonatal Research Network Randomized Clinical Trial of Intravenous Immune Globulin to Prevent Neonatal Infection in Very-Low-Birth-Weight (501-1500 g) Infants (IVIG trial). The research is significant, as failing to account for the hierarchical structure within complete twin births may impact the estimate of target sample size and the precision of treatment effect estimates and/or decisions regarding treatment efficacy. Thus, the effects of trial design and analysis need to be quantified and taken into consideration for studies that involve singletons and twin births. [unreadable] [unreadable] [unreadable] [unreadable]