A joint latent class analysis for adjusting survival bias with application to a trauma transfusion study

Abstract

There is no clear classification rule to rapidly identify trauma patients who are severely hemorrhaging and may need substantial blood transfusions. Massive transfusion (MT), defined as the transfusion of at least 10 units of red blood cells within 24 h of hospital admission, has served as a conventional surrogate that has been used to develop early predictive algorithms and establish criteria for ordering an MT protocol from the blood bank. However, the conventional MT rule is a poor proxy, because it is likely to misclassify many severely hemorrhaging trauma patients as they could die before receiving the 10th red blood cells transfusion. In this article, we propose to use a latent class model to obtain a more accurate and complete metric in the presence of early death. Our new approach incorporates baseline patient information from the time of hospital admission, by combining respective models for survival time and usage of blood products transfused within the framework of latent class analysis. To account for statistical challenges, caused by induced dependent censoring inherent in 24-h sums of transfusions, we propose to estimate an improved standard via a pseudo-likelihood function using an expectation-maximization algorithm with the inverse weighting principle. We evaluated the performance of our new standard in simulation studies and compared with the conventional MT definition using actual patient data from the Prospective Observational Multicenter Major Trauma Transfusion study.

From Donor to Recipient: Considerations for Blood Transfusion Outcomes Research

“…Transfusion and survival bias. Due to the many factors that impact patient outcomes in the case of trauma and the urgent nature of resuscitation after injury, investigators stress the importance of survival bias in analysis of patient outcomes: Do patients survive because of the treatments they receive, or would they survive if they had received the treatment earlier (Ning et al, 2016)? To account for survival bias, researchers use statistical modeling techniques like the Cox proportional hazards model to provide a fluid comparison of patients based on the treatments they receive over time, including in previous studies focused on the ratios of PRBCs used during massive transfusion in patients with major trauma (Holcomb et al, 2013;Snyder et al, 2009).…”