An evaluation of constrained randomization for the design and analysis of group‐randomized trials with binary outcomes

Li, Fan; Turner, Elizabeth L.; Heagerty, Patrick J.; Murray, David M.; Vollmer, William M.; DeLong, Elizabeth R.

doi:10.1002/sim.7410

Cited by 75 publications

(149 citation statements)

References 56 publications

(110 reference statements)

Supporting

Mentioning

143

Contrasting

Order By: Relevance

“…We consider two choices of degrees of freedom (DoF). The first DoF dates back to the work of Mancl and DeRouen 29 and equal to the number of clusters minus the number of regression parameters; this DoF has been previously used in the GEE analyses of parallel CRTs, 32 three-level CRTs, 33 crossover CRTs, 6 stepped wedge CRTs, 5,34 and shown to have test size not exceeding the nominal level. The second DoF was suggested in the PhD dissertation of Ford 35 and specifies DoF = I − 2.…”

Section: Calculations 41 | Closed-form Variance Of the Intervention mentioning

confidence: 99%

Design and analysis considerations for cohort stepped wedge cluster randomized trials with a decay correlation structure

2019

Statistics in Medicine

Self Cite

108

View full text Add to dashboard Cite

A stepped wedge cluster randomized trial is a type of longitudinal cluster design that sequentially switches clusters to intervention over time until all clusters are treated. While the traditional posttest-only parallel design requires adjustment for a single intraclass correlation coefficient, the stepped wedge design allows multiple outcome measurements from the same cluster and so additional correlation parameters are necessary to characterize the within-cluster correlation structure. Although a number of studies have differentiated between the concepts of within-period and between-period correlations, only a few studies have allowed the between-period correlation to decay over time. In this article, we consider the proportional decay correlation structure for a cohort stepped wedge design, and provide a matrix-adjusted quasi-least squares approach to accurately estimate the correlation parameters along with the marginal intervention effect. We further develop the sample size and power procedures accounting for the correlation decay, and investigate the accuracy of the power procedure with continuous outcomes in a simulation study. We show that the empirical power agrees well with the prediction even with as few as nine clusters, when data are analyzed with matrix-adjusted quasi-least squares concurrently with a suitable bias-corrected sandwich variance. Two trial examples are provided to illustrate the new sample size procedure.

show abstract

Section: Calculations 41 | Closed-form Variance Of the Intervention mentioning

confidence: 99%

Design and analysis considerations for cohort stepped wedge cluster randomized trials with a decay correlation structure

2019

Statistics in Medicine

Self Cite

108

View full text Add to dashboard Cite

show abstract

“…It is well known that randomization in individual randomized controlled trials (RCTs) balances both observed and unobserved covariates, thereby ensuring the precision and internal validity of the treatment estimate, even without accounting for important covariates. This implies that, for an RCT, a mean difference (for continuous Previous studies sought to investigate imbalance in cluster-level covariates using a continuous outcome, but few previous studies focused on individual-level covariates or evaluated binary clinical outcomes [10][11][12]. Assuming no presence of heterogeneity of treatment effect (HTE), we sought to quantify the impact of imbalance in individual-level covariates (eg, race) on the average treatment effect in CRTs for both continuous and binary outcomes.…”

Section: Introductionmentioning

confidence: 99%

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

Yang

Starks

Hernandez

et al. 2020

Contemporary Clinical Trials

Self Cite

View full text Add to dashboard Cite

A R T I C L E I N F O Keywords: (MeSH terms): Bias Cluster analysis Multivariable analysis Randomized controlled trials as topic Research design A B S T R A C TIndividual-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. outcomes) or odds ratio (for binary outcomes) suffices as a measure of treatment effect. In practice, results for many CRTs have also been reported using similar (unadjusted) estimators [4,5]. However, the internal validity of CRTs is challenged when treatment arms are not comparable. Brierley et al. [6] found that about 30% of CRTs may suffer from selection bias, resulting in substantial covariate imbalance. Previous literature recommended that researchers could control selection bias in CRTs via cautious design, such as identifying patients prior to randomization and blinding [7][8][9]. Aside from selection bias, another important source of covariate imbalance is random chance, and that is the focus of the current paper. We show that baseline covariate imbalance may happen more frequently in CRTs compared to RCTs. If the covariates are predictive of the clinical outcomes, they act as confounders and may cause a spurious association between intervention and clinical outcome when not accounted for properly. As a result, it is important to understand how to accurately account for important individual-level covariates in the design and analysis of CRTs where confounders typically can impact the outcome.

show abstract

“…The design is based on adjustment of baseline covariates to avoid chance imbalance. Covariate-constrained randomization [ 15 , 16 ] is the method which is used to balance for many covariates between the intervention and standard treatment arms. This method limits the risk of selection bias.…”

Section: Methodsmentioning

confidence: 99%

m-Power Heart Project - a nurse care coordinator led, mHealth enabled intervention to improve the management of hypertension in India: study protocol for a cluster randomized trial

Venkateshmurthy

Ajay

Mohan

et al. 2018

Trials

View full text Add to dashboard Cite

BackgroundThe proportion of patients with controlled hypertension (< 140/90 mmHg) is very low in India. Thus, there is a need to improve blood pressure management among patients with uncontrolled hypertension through innovative strategies directed at health system strengthening.MethodsWe designed an intervention consisting of two important components – an electronic decision support system (EDSS) used by a trained nurse care coordinator (NCC). Based on preliminary data, we hypothesized that this intervention will be able to reduce mean systolic blood pressure by 6.5 mmHg among those with uncontrolled blood pressure in the intervention arm compared to the standard treatment arm (paper-based hypertension treatment guidelines). The study will adopt a cluster randomized trial design with the community health center (CHC) as the unit of randomization. The trial will be conducted in Visakhapatnam district (southern India). A total of 1876 participants aged ≥30 years with high blood pressure – systolic blood pressure (SBP) ≥ 160 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg will be enrolled from 12 CHCs. The intervention consists of trained NCCs equipped with an evidence-based hypertension treatment algorithm in the form of the EDSS with regular SMSs to patients with hypertension to promote hypertension treatment and blood pressure control for 12 months. The primary outcome will be difference in the mean change of SBP, from baseline to 12 months, between the intervention and the standard treatment arm. The secondary outcomes are the difference in mean change of DBP; difference in the proportion of patients with controlled blood pressure (< 140/90 mmHg); difference in mean change of fasting blood sugar, HbA1C, eGFR, and albumin to creatinine ratio; difference in the proportion of patients visiting the CHC regularly (number of actual visits to the CHC/number of visits suggested by the EDSS > 80%); difference in proportion of patients compliant to anti-hypertensive medication/s; cost-effectiveness of intervention versus enhanced care. All the outcomes will be assessed at 12 months.DiscussionThe study is expected to provide evidence on the effectiveness of NCC-led, EDSS-based hypertension management in India and can likely offer an exemplar for improving cardiovascular disease (CVD) management in India within the resource-constrained public healthcare system.Trial registrationClinicalTrials.gov, ID: NCT03164317). Registered retrospectively on 23 May 2017 (first patient enrolled on 6 April 2017) because the authors did not receive a response to their original registration submission (5 January 2017) to the Clinical Trial Registry – India (CTRI).Electronic supplementary materialThe online version of this article (10.1186/s13063-018-2813-2) contains supplementary material, which is available to authorized users.

show abstract

An evaluation of constrained randomization for the design and analysis of group‐randomized trials with binary outcomes

Cited by 75 publications

References 56 publications

Design and analysis considerations for cohort stepped wedge cluster randomized trials with a decay correlation structure

Design and analysis considerations for cohort stepped wedge cluster randomized trials with a decay correlation structure

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

m-Power Heart Project - a nurse care coordinator led, mHealth enabled intervention to improve the management of hypertension in India: study protocol for a cluster randomized trial

Contact Info

Product

Resources

About