Dataset Information

Impact of baseline covariate imbalance on bias in treatment effect estimation in cluster randomized trials: Race as an example.

ABSTRACT: Individual-level baseline covariate imbalance could happen more frequently in cluster randomized trials, and may influence the observed treatment effect. Using computer and real-data simulations, this paper quantifies the extent and impact of covariate imbalance on the estimated treatment effect for both continuous and binary outcomes, and relates it to the degree of imbalance for different numbers of clusters, cluster sizes, and covariate intraclass correlation coefficients. We focused on the impact of race as a covariate, given the emphasis of regulatory and funding bodies on understanding the influence of demographic characteristics on treatment effectiveness. We found that bias in the treatment effect is proportional to both the degree of baseline covariate imbalance and the covariate effect size. Larger numbers of clusters result in lower covariate imbalance, and increasing cluster size is less effective in reducing imbalance compared to increasing the number of clusters. Models adjusted for important baseline confounders are superior to unadjusted models for minimizing bias in both model-based simulations and an innovative simulation based on real clinical trial data. Higher outcome intraclass correlation coefficients did not affect bias but resulted in greater variance in treatment estimates.

SUBMITTER: Yang S

PROVIDER: S-EPMC8337048 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Impact of baseline covariate imbalance on bias in treatment effect estimation in cluster randomized trials: Race as an example.

Yang Siyun S Starks Monique Anderson MA Hernandez Adrian F AF Turner Elizabeth L EL Califf Robert M RM O'Connor Christopher M CM Mentz Robert J RJ Roy Choudhury Kingshuk K

Contemporary clinical trials 20190620

Individual-level baseline covariate imbalance could happen more frequently in cluster randomized trials, and may influence the observed treatment effect. Using computer and real-data simulations, this paper quantifies the extent and impact of covariate imbalance on the estimated treatment effect for both continuous and binary outcomes, and relates it to the degree of imbalance for different numbers of clusters, cluster sizes, and covariate intraclass correlation coefficients. We focused on the i ...[more]

PMID: 31228563

Similar Datasets

Project description:BackgroundIn cluster-randomized controlled trials (C-RCTs), covariate-constrained randomization (CCR) methods efficiently control imbalance in multiple baseline cluster-level variables, but the choice of imbalance metric to define the subset of "adequately balanced" possible allocation schemes for C-RCTs involving more than two arms and continuous variables is unclear. In an ongoing three-armed C-RCT, we chose the min(three Kruskal-Wallis [KW] test P values) > 0.30 as our metric. We use simulation studies to explore the performance of this and other metrics of baseline variable imbalance in CCR.MethodsWe simulated three continuous variables across three arms under varying allocation ratios and assumptions. We compared the performance of min(analysis of variance [ANOVA] P value) > 0.30, min(KW P value) > 0.30, multivariate analysis of variance (MANOVA) P value > 0.30, min(nine possible t test P values) > 0.30, and min(Wilcoxon rank-sum [WRS] P values) > 0.30.ResultsPairwise comparison metrics (t test and WRS) tended to be the most conservative, providing the smallest subset of allocation schemes (10%-13%) meeting criteria for acceptable balance. Sensitivity of the min(t test P values) > 0.30 for detecting non-trivial imbalance was 100% for both hypothetical and resampled simulation scenarios. The KW criterion maintained higher sensitivity than both the MANOVA and ANOVA criteria (89% to over 99%) but was not as sensitive as pairwise criteria.ConclusionsOur criterion, the KW P value > 0.30, to signify "acceptable" balance was not the most conservative, but it appropriately identified imbalance in the majority of simulations. Since all are related, CCR algorithms involving any of these imbalance metrics for continuous baseline variables will ensure robust simultaneous control over multiple continuous baseline variables, but we recommend care in determining the threshold of "acceptable" levels of (im)balance.Trial registrationThis trial is registered on ClinicalTrials.gov (initial post: December 1, 2016; identifier: NCT02979444 ).

Project description:BackgroundTraditionally in acute stroke clinical trials, the primary clinical outcome employed is a dichotomized modified Rankin Scale (mRS). New statistical methods, such as responder analysis, are being used in stroke studies to address the concern that baseline prognostic variables, such as stroke severity, impact the likelihood of a successful outcome. Responder analysis allows the definition of success to vary according to baseline prognostic variables, producing a more clinically relevant insight into the actual effect of investigational treatments. It is unclear whether or not statistical analyses should adjust for prognostic variables when responder analysis is used, as the outcome already takes these prognostic variables into account. This research aims to investigate the effect of covariate adjustment in the responder analysis framework in order to determine the appropriate analytic method.MethodsUsing a current stroke clinical trial and its pilot studies to guide simulation parameters, 1,000 clinical trials were simulated at varying sample sizes under several treatment effects to assess power and type I error. Covariate-adjusted and unadjusted logistic regressions were used to estimate the treatment effect under each scenario. In the case of covariate-adjusted logistic regression, the trichotomized National Institute of Health Stroke Scale (NIHSS) was used in adjustment.ResultsUnder various treatment effect settings, the operating characteristics of the unadjusted and adjusted analyses do not substantially differ. Power and type I error are preserved for both the unadjusted and adjusted analyses.ConclusionsOur results suggest that, under the given treatment effect scenarios, the decision whether or not to adjust for baseline severity when using a responder analysis outcome should be guided by the needs of the study, as type I error rates and power do not appear to vary largely between the methods. These findings are applicable to stroke trials which use the mRS for the primary outcome, but also provide a broader insight into the analysis of binary outcomes that are defined based on baseline prognostic variables.Trial registrationThis research is part of the Stroke Hyperglycemia Insulin Network Effort (SHINE) trial, Identification Number NCT01369069.

Dataset Information

Impact of baseline covariate imbalance on bias in treatment effect estimation in cluster randomized trials: Race as an example.

Publications

Impact of baseline covariate imbalance on bias in treatment effect estimation in cluster randomized trials: Race as an example.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets