Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials

Abstract Background The hazard ratio of the Cox proportional hazards model is widely used in randomized controlled trials to assess treatment effects. However, two properties of the hazard ratio including the non-collapsibility and built-in selection bias need to be further investigated. Methods We...

Full description

Saved in:
Bibliographic Details
Main Authors: Helen Bian, Menglan Pang, Guanbo Wang, Zihang Lu
Format: Article
Language:English
Published: BMC 2024-11-01
Series:BMC Medical Research Methodology
Subjects:
Online Access:https://doi.org/10.1186/s12874-024-02402-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846147714876702720
author Helen Bian
Menglan Pang
Guanbo Wang
Zihang Lu
author_facet Helen Bian
Menglan Pang
Guanbo Wang
Zihang Lu
author_sort Helen Bian
collection DOAJ
description Abstract Background The hazard ratio of the Cox proportional hazards model is widely used in randomized controlled trials to assess treatment effects. However, two properties of the hazard ratio including the non-collapsibility and built-in selection bias need to be further investigated. Methods We conduct simulations to differentiate the non-collapsibility effect and built-in selection bias from the difference between the marginal and the conditional hazard ratio. Meanwhile, we explore the performance of the Cox model with inverse probability of treatment weighting for covariate adjustment when estimating the marginal hazard ratio. The built-in selection bias is further assessed in the period-specific hazard ratio. Results The conditional hazard ratio is a biased estimate of the marginal effect due to the non-collapsibility property. In contrast, the hazard ratio estimated from the inverse probability of treatment weighting Cox model provides an unbiased estimate of the true marginal hazard ratio. The built-in selection bias only manifests in the period-specific hazard ratios even when the proportional hazards assumption is satisfied. The Cox model with inverse probability of treatment weighting can be used to account for confounding bias and provide an unbiased effect under the randomized controlled trials setting when the parameter of interest is the marginal effect. Conclusions We propose that the period-specific hazard ratios should always be avoided due to the profound effects of built-in selection bias.
format Article
id doaj-art-f7e6de9e36a44ea0af8b9b8da9718614
institution Kabale University
issn 1471-2288
language English
publishDate 2024-11-01
publisher BMC
record_format Article
series BMC Medical Research Methodology
spelling doaj-art-f7e6de9e36a44ea0af8b9b8da97186142024-12-01T12:32:15ZengBMCBMC Medical Research Methodology1471-22882024-11-0124111610.1186/s12874-024-02402-3Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trialsHelen Bian0Menglan PangGuanbo Wang1Zihang Lu2Department of Epidemiology, Biostatistics and Occupational Health, McGill University, 2001, McGill College CAUSALab, Harvard T.H. Chan School of Public HealthDepartment of Public Health Sciences & Department of Mathematics and Statistics, Queen’s UniversityAbstract Background The hazard ratio of the Cox proportional hazards model is widely used in randomized controlled trials to assess treatment effects. However, two properties of the hazard ratio including the non-collapsibility and built-in selection bias need to be further investigated. Methods We conduct simulations to differentiate the non-collapsibility effect and built-in selection bias from the difference between the marginal and the conditional hazard ratio. Meanwhile, we explore the performance of the Cox model with inverse probability of treatment weighting for covariate adjustment when estimating the marginal hazard ratio. The built-in selection bias is further assessed in the period-specific hazard ratio. Results The conditional hazard ratio is a biased estimate of the marginal effect due to the non-collapsibility property. In contrast, the hazard ratio estimated from the inverse probability of treatment weighting Cox model provides an unbiased estimate of the true marginal hazard ratio. The built-in selection bias only manifests in the period-specific hazard ratios even when the proportional hazards assumption is satisfied. The Cox model with inverse probability of treatment weighting can be used to account for confounding bias and provide an unbiased effect under the randomized controlled trials setting when the parameter of interest is the marginal effect. Conclusions We propose that the period-specific hazard ratios should always be avoided due to the profound effects of built-in selection bias.https://doi.org/10.1186/s12874-024-02402-3Non-collapsibilityBuilt-in selection biasCox proportional hazards modelInverse probability of treatment weightingRandomized controlled trials
spellingShingle Helen Bian
Menglan Pang
Guanbo Wang
Zihang Lu
Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
BMC Medical Research Methodology
Non-collapsibility
Built-in selection bias
Cox proportional hazards model
Inverse probability of treatment weighting
Randomized controlled trials
title Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
title_full Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
title_fullStr Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
title_full_unstemmed Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
title_short Non-collapsibility and built-in selection bias of period-specific and conventional hazard ratio in randomized controlled trials
title_sort non collapsibility and built in selection bias of period specific and conventional hazard ratio in randomized controlled trials
topic Non-collapsibility
Built-in selection bias
Cox proportional hazards model
Inverse probability of treatment weighting
Randomized controlled trials
url https://doi.org/10.1186/s12874-024-02402-3
work_keys_str_mv AT helenbian noncollapsibilityandbuiltinselectionbiasofperiodspecificandconventionalhazardratioinrandomizedcontrolledtrials
AT menglanpang noncollapsibilityandbuiltinselectionbiasofperiodspecificandconventionalhazardratioinrandomizedcontrolledtrials
AT guanbowang noncollapsibilityandbuiltinselectionbiasofperiodspecificandconventionalhazardratioinrandomizedcontrolledtrials
AT zihanglu noncollapsibilityandbuiltinselectionbiasofperiodspecificandconventionalhazardratioinrandomizedcontrolledtrials