scholarly journals Sample size calculation for stepped-wedge cluster-randomized trials with more than two levels of clustering

2019 ◽  
Vol 16 (3) ◽  
pp. 225-236 ◽  
Author(s):  
Steven Teerenstra ◽  
Monica Taljaard ◽  
Anja Haenen ◽  
Anita Huis ◽  
Femke Atsma ◽  
...  
Keyword(s):  
Hand Hygiene ◽  
Sample Size ◽  
Sample Size Calculation ◽  
Correlation Coefficients ◽  
Stepped Wedge ◽  
Change Trial ◽  
Stepped Wedge Trials ◽  
Hygiene Compliance ◽  
Standard Designs ◽  
Cluster Randomized

Background/Aims: Power and sample size calculation formulas for stepped-wedge trials with two levels (subjects within clusters) are available. However, stepped-wedge trials with more than two levels are possible. An example is the CHANGE trial which randomizes nursing homes (level 4) consisting of nursing home wards (level 3) in which nurses (level 2) are observed with respect to their hand hygiene compliance during hand hygiene opportunities (level 1) in the care of patients. We provide power and sample size methods for such trials and illustrate these in the setting of the CHANGE trial. Methods: We extend the original sample size methodology derived for stepped-wedge trials based on a random intercepts model, to accommodate more than two levels of clustering. We derive expressions that can be used to determine power and sample size for p levels of clustering in terms of the variances at each level or, alternatively, in terms of intracluster correlation coefficients. We consider different scenarios, depending on whether the same units in a particular level are repeatedly measured as a cohort sample or whether different units are measured cross-sectionally. Results: A simple variance inflation factor is obtained that can be used to calculate power and sample size for continuous and by approximation for binary and rate outcomes. It is the product of (1) variance inflation due to the multilevel structure and (2) variance inflation due to the stepped-wedge manner of assigning interventions over time. Standard and non-standard designs (i.e. so-called “hybrid designs” and designs with more, less, or no data collection when the clusters are all in the control or are all in the intervention condition) are covered. Conclusions: The formulas derived enable power and sample size calculations for multilevel stepped-wedge trials. For the two-, three-, and four-level case of the standard stepped wedge, we provide programs to facilitate these calculations.

Sample size determination for stepped wedge cluster randomized trials in pragmatic settings

2021 ◽  
pp. 096228022110223
Author(s):  
Jijia Wang ◽  
Jing Cao ◽  
Song Zhang ◽  
Chul Ahn
Keyword(s):  
Sample Size ◽  
Closed Form ◽  
Sample Size Calculation ◽  
Repeated Measurements ◽  
Sample Size Determination ◽  
Stepped Wedge ◽  
Pragmatic Trials ◽  
Stepped Wedge Trials ◽  
Cluster Randomized ◽  
The Impact

The stepped-wedge cluster randomized design has been increasingly employed by pragmatic trials in health services research. In this study, based on the GEE approach, we present closed-form sample size calculation that is applicable to both closed-cohort and cross-sectional stepped wedge trials. Importantly, the proposed method is flexible to accommodate design issues routinely encountered in pragmatic trials, such as different within- and between-subject correlation structures, irregular crossover schedules for the switch to intervention, and missing data due to repeated measurements over prolonged follow-up. The closed-form formulas allow researchers to analytically assess the impact of different design factors on sample size requirement. We also recognize the potential issue of limited numbers of clusters in pragmatic stepped wedge trials and present an adjustment approach for underestimated variance of the treatment effect. We conduct extensive simulation to assess the performance of the proposed sample size method. An application example to a real clinical trial is presented.

scholarly journals A tutorial on sample size calculation for multiple-period cluster randomized parallel, cross-over and stepped-wedge trials using the Shiny CRT Calculator

2020 ◽  
Vol 49 (3) ◽  
pp. 979-995 ◽  
Author(s):  
Karla Hemming ◽  
Jessica Kasza ◽  
Richard Hooper ◽  
Andrew Forbes ◽  
Monica Taljaard
Keyword(s):  
Sample Size ◽  
Group Treatment ◽  
Single Point ◽  
Sample Size Calculation ◽  
Stepped Wedge ◽  
Cross Sectional ◽  
Multiple Observations ◽  
Sample Size Calculations ◽  
Correlation Structures ◽  
Cluster Randomized

Abstract It has long been recognized that sample size calculations for cluster randomized trials require consideration of the correlation between multiple observations within the same cluster. When measurements are taken at anything other than a single point in time, these correlations depend not only on the cluster but also on the time separation between measurements and additionally, on whether different participants (cross-sectional designs) or the same participants (cohort designs) are repeatedly measured. This is particularly relevant in trials with multiple periods of measurement, such as the cluster cross-over and stepped-wedge designs, but also to some degree in parallel designs. Several papers describing sample size methodology for these designs have been published, but this methodology might not be accessible to all researchers. In this article we provide a tutorial on sample size calculation for cluster randomized designs with particular emphasis on designs with multiple periods of measurement and provide a web-based tool, the Shiny CRT Calculator, to allow researchers to easily conduct these sample size calculations. We consider both cross-sectional and cohort designs and allow for a variety of assumed within-cluster correlation structures. We consider cluster heterogeneity in treatment effects (for designs where treatment is crossed with cluster), as well as individually randomized group-treatment trials with differential clustering between arms, for example designs where clustering arises from interventions being delivered in groups. The calculator will compute power or precision, as a function of cluster size or number of clusters, for a wide variety of designs and correlation structures. We illustrate the methodology and the flexibility of the Shiny CRT Calculator using a range of examples.

scholarly journals Improving Hand Hygiene Compliance in Nursing Homes: Protocol for a Cluster Randomized Controlled Trial (HANDSOME Study)

10.2196/17419 ◽  
2020 ◽  
Vol 9 (5) ◽  
pp. e17419 ◽  
Author(s):  
Gwen R Teesing ◽  
Vicki Erasmus ◽  
Mariska Petrignani ◽  
Marion P G Koopmans ◽  
Miranda de Graaf ◽  
...  
Keyword(s):  
Infectious Disease ◽  
Nursing Homes ◽  
Hand Hygiene ◽  
Hand Hygiene Compliance ◽  
Disease Incidence ◽  
Controlled Trial ◽  
World Health ◽  
The World ◽  
Hygiene Compliance ◽  
Cluster Randomized

Background Hand hygiene compliance is considered the most (cost-)effective measure for preventing health care–associated infections. While hand hygiene interventions have frequently been implemented and assessed in hospitals, there is limited knowledge about hand hygiene compliance in other health care settings and which interventions and implementation methods are effective. Objective This study aims to evaluate the effect of a multimodal intervention to increase hand hygiene compliance of nurses in nursing homes through a cluster randomized controlled trial (HANDSOME study). Methods Nursing homes were randomly allocated to 1 of 3 trial arms: receiving the intervention at a predetermined date, receiving the identical intervention after an infectious disease outbreak, or serving as a control arm. Hand hygiene was evaluated in nursing homes by direct observation at 4 timepoints. We documented compliance with the World Health Organization’s 5 moments of hand hygiene, specifically before touching a patient, before a clean/aseptic procedure, after body fluid exposure risk, after touching a patient, and after touching patient surroundings. The primary outcome is hand hygiene compliance of the nurses to the standards of the World Health Organization. The secondary outcome is infectious disease incidence among residents. Infectious disease incidence was documented by a staff member at each nursing home unit. Outcomes will be compared with the presence of norovirus, rhinovirus, and Escherichia coli on surfaces in the nursing homes, as measured using quantitative polymerase chain reaction. Results The study was funded in September 2015. Data collection started in October 2016 and was completed in October 2017. Data analysis will be completed in 2020. Conclusions HANDSOME studies the effectiveness of a hand hygiene intervention specifically for the nursing home environment. Nurses were taught the World Health Organization’s 5 moments of hand hygiene guidelines using the slogan “Room In, Room Out, Before Clean, After Dirty,” which was developed for nursing staff to better understand and remember the hygiene guidelines. HANDSOME should contribute to improved hand hygiene practice and a reduction in infectious disease rates and related mortality. Trial Registration Netherlands Trial Register (NTR6188) NL6049; https://www.trialregister.nl/trial/6049 International Registered Report Identifier (IRRID) DERR1-10.2196/17419

scholarly journals Is the R coefficient of interest in cluster randomized trials with a binary outcome?

2020 ◽  
Vol 29 (9) ◽  
pp. 2470-2480
Author(s):  
Ariane M Mbekwe Yepnang ◽  
Agnès Caille ◽  
Sandra M Eldridge ◽  
Bruno Giraudeau
Keyword(s):  
Sample Size ◽  
Simulation Study ◽  
Randomized Trials ◽  
Intraclass Correlation ◽  
Sample Size Calculation ◽  
Sample Size Determination ◽  
Binary Outcomes ◽  
Cluster Randomized Trials ◽  
Empirical Power ◽  
Cluster Randomized

In cluster randomized trials, the intraclass correlation coefficient (ICC) is classically used to measure clustering. When the outcome is binary, the ICC is known to be associated with the prevalence of the outcome. This association challenges its interpretation and can be problematic for sample size calculation. To overcome these situations, Crespi et al. extended a coefficient named R, initially proposed by Rosner for ophthalmologic data, to cluster randomized trials. Crespi et al. asserted that R may be less influenced by the outcome prevalence than is the ICC, although the authors provided only empirical data to support their assertion. They also asserted that “the traditional ICC approach to sample size determination tends to overpower studies under many scenarios, calling for more clusters than truly required”, although they did not consider empirical power. The aim of this study was to investigate whether R could indeed be considered independent of the outcome prevalence. We also considered whether sample size calculation should be better based on the R coefficient or the ICC. Considering the particular case of 2 individuals per cluster, we theoretically demonstrated that R is not symmetrical around the 0.5 prevalence value. This in itself demonstrates the dependence of R on prevalence. We also conducted a simulation study to explore the case of both fixed and variable cluster sizes greater than 2. This simulation study demonstrated that R decreases when prevalence increases from 0 to 1. Both the analytical and simulation results demonstrate that R depends on the outcome prevalence. In terms of sample size calculation, we showed that an approach based on the ICC is preferable to an approach based on the R coefficient because with the former, the empirical power is closer to the nominal one. Hence, the R coefficient does not outperform the ICC for binary outcomes because it does not offer any advantage over the ICC.

scholarly journals Increased hand hygiene compliance in nursing homes after a multimodal intervention: A cluster randomized controlled trial (HANDSOME)

2020 ◽  
Vol 41 (10) ◽  
pp. 1169-1177
Author(s):  
Gwen R. Teesing ◽  
Vicki Erasmus ◽  
Daan Nieboer ◽  
Mariska Petrignani ◽  
Marion P.G Koopmans ◽  
...  
Keyword(s):  
Randomized Controlled Trial ◽  
Nursing Home ◽  
Nursing Homes ◽  
Hand Hygiene ◽  
Hand Hygiene Compliance ◽  
Controlled Trial ◽  
Multimodal Intervention ◽  
Randomized Controlled ◽  
Hygiene Compliance ◽  
Cluster Randomized

AbstractObjective:To assess the effect of a multimodal intervention on hand hygiene compliance (HHC) in nursing homes.Design, setting, and participants:HHC was evaluated using direct, unobtrusive observation in a cluster randomized controlled trial at publicly funded nursing homes in the Netherlands. In total, 103 nursing home organizations were invited to participate; 18 organizations comprising 33 nursing homes (n = 66 nursing home units) participated in the study. Nursing homes were randomized into a control group (no intervention, n = 30) or an intervention group (multimodal intervention, n = 36). The primary outcome measure was HHC of nurses. HHC was appraised at baseline and at 4, 7, and 12 months after baseline. Observers and nurses were blinded.Intervention:Audits regarding hand hygiene (HH) materials and personal hygiene rules, 3 live lessons, an e-learning program, posters, and a photo contest. We used a new method to teach the nurses the WHO-defined 5 moments of HH: Room In, Room Out, Before Clean, and After Dirty.Results:HHC increased in both arms. The increase after 12 months was larger for units in the intervention arm (from 12% to 36%) than for control units (from 13% to 21%) (odds ratio [OR], 2.10; confidence interval [CI], 1.35–3.28). The intervention arm exhibited a statistically significant increase in HHC at 4 of the 5 WHO-defined HH moments. At follow-up, HHC in the intervention arm remained statistically significantly higher (OR, 1.93; 95% CI, 1.59–2.34) for indications after an activity (from 37% to 39%) than for indications before an activity (from 14% to 27%).Conclusions:The HANDSOME intervention is successful in improving HHC in nursing homes.

Sign in / Sign up

Export Citation Format

Share Document