How Cox models react to a study-specific confounder in a patient-level pooled dataset: Random-effects better cope with an imbalanced covariate across trials unless baseline haz

05/08/2018
by   Thomas McAndrew, et al.
0

Combining patient-level data from clinical trials can connect rare phenomena with clinical endpoints, but statistical techniques applied to a single trial may become problematical when trials are pooled. Estimating the hazard of a binary variable unevenly distributed across trials showcases a common pooled database issue. We studied how an unevenly distributed binary variable can compromise the integrity of fixed and random effects Cox proportional hazards models. We compared fixed effect and random effects Cox proportional hazards models on a set of simulated datasets inspired by a 17-trial pooled database of patients presenting with ST-segment elevation myocardial infarction (STEMI) and non-STEMI undergoing percutaneous coronary intervention. An unevenly distributed covariate can bias hazard ratio estimates, inflate standard errors, raise type I error, and reduce power. While uneveness causes problems for all Cox proportional hazards models, random effects suffer least. Compared to fixed effect models, random effects suffer lower bias and trade inflated type I errors for improved power. Contrasting hazard rates between trials prevent accurate estimates from both fixed and random effects models. When modeling a covariate unevenly distributed across pooled trials with similar baseline hazard rates, Cox proportional hazards models with a random trial effect more accurately estimate hazard ratios than fixed effects. Differing between-trial baseline hazard rates bias both random and fixed effect models. With an unevenly-distributed covariate and similar baseline hazard rates across trials, a random effects Cox proportional hazards model outperforms a fixed effect model, but cannot overcome contrasting baseline hazard rates.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/22/2020

Subgroup identification in individual patient data meta-analysis using model-based recursive partitioning

Model-based recursive partitioning (MOB) can be used to identify subgrou...
research
06/24/2022

Accounting for Inconsistent Use of Covariate Adjustment in Group Sequential Trials

Group sequential designs are common in clinical trials, as they allow fo...
research
12/24/2020

Bayesian prognostic covariate adjustment

Historical data about disease outcomes can be integrated into the analys...
research
12/31/2020

CauchyCP: a powerful test under non-proportional hazards using Cauchy combination of change-point Cox regressions

Non-proportional hazards data are routinely encountered in randomized cl...
research
09/20/2023

Confounding by Scarcity: An Overlooked Source of Bias in Pragma@c Trials

Pragmatic trials evaluate the effectiveness of health interventions comp...
research
10/19/2020

Efficient Similarity-Preserving Unsupervised Learning using Modular Sparse Distributed Codes and Novelty-Contingent Noise

There is increasing realization in neuroscience that information is repr...
research
03/09/2023

Likelihood-based Inference for Skewed Responses in a Crossover Trial Setup

This work proposes a statistical model for crossover trials with multipl...

Please sign up or login with your details

Forgot password? Click here to reset