A Data-Adaptive Targeted Learning Approach of Evaluating Viscoelastic Assay Driven Trauma Treatment Protocols

• Activity Number: 187 - Contributed Poster Presentations: Section on Nonparametric Statistics
• Type: Contributed
• Date/Time: Monday, July 29, 2019 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Linqing Wei* and Alan Hubbard and Lucy Zumwinkle Kornblith and Mitchell Jay Cohen
• Companies: Univ of California - Berkeley, Biostatistics Department and University of California, Berkeley and University of California,San Francisco and University of Colorado School of Medicine
• Keywords: machine learning; semiparametric; causal inference; high-dimensional; limited sample; data-adaptive

Abstract

Trauma is a a leading cause of death and poses clinical and statistical challenges. Using highly predictive measures to optimize treatment assignment is of great current interest.However, the impact of treatment assigned based on viscoelastic assays TEG/ROTEM has not been closely examined.The goal of our study was to employ robust, semiparametric data-adaptive modeling procedures to estimate the potential impact of protocols for achieving hemostasis. Given the common problem of high dimension and small sample,standard doubly-robust estimators can break down due to positivity violations,caused by high correlation between covariates and exposures in trauma data.We develop estimators of our estimands of interest within the collaborative targeted minimum loss-based estimation framework, that optimizes the variance-bias trade-off, not with respect to prediction intervention, but to the estimand of interest.This allows for more automated estimation using machine learning in situations with limited data.The results showed significant improvement in trauma patient’s hemostasis status and a decrease in mortality rate at 6h, but no significant change in mortality rate at 24h after arrival.