clusterPower (version 0.7.0)

cpa.sw.binary: Power simulations for cluster-randomized trials: Stepped Wedge Design, Binary Outcome

Description

This function uses a modified Cox method to determine power for stepped wedge cluster-randomized controlled trials. Users can modify a variety of parameters to suit their desired experimental situation.

Usage

cpa.sw.binary(
  nclusters = NA,
  steps = NA,
  nsubjects = NA,
  alpha = 0.05,
  timeEffect = 0,
  ICC = NA,
  p0 = NA,
  p1 = NA,
  tol = 1e-05,
  GQ = 100,
  quiet = FALSE
)

Arguments

nclusters

Number of clusters; accepts non-negative integer scalar (required).

steps

Number of crossover steps; Accepts positive scalar indicating the total number of steps, NOT including the baseline (required).

nsubjects

Number of subjects per cluster; accepts a scalar. Equal cluster sizes are assumed (required).

alpha

Significance level (default=0.05).

timeEffect

Expected time effect over the entire study period (assumed to be linear across time steps); accepts numeric (required). Default = 0 (no time effects).

ICC

Intracluster correlation coefficient as defined by Hussey and Hughes (2007) for participants at first time step; accepts numeric (required).

p0

Estimated baseline effect; accepts numeric (required).

p1

Estimated treatment effect; accepts numeric (required).

tol

Machine tolerance. Accepts numeric. Default is 1e-5.

GQ

Number of quadriture points used in Gaussian Legendre integration; accepts a scalar. Default is 100.

quiet

Suppresses the progress bar; logical. Default is FALSE.

Value

The estimated power.

Details

The stepped wedge trial design is a type of cross-over design in which clusters change treatments in waves. Initially all the clusters receive the same standard treatment, and at the end of the trial all of the clusters will be receiving the treatment of interest. More than one cluster can change treatments in a wave, but the order in which clusters change treatments is randomly determined. The outcome of interest is assessed in each cluster during each wave.

Users must specify the number of subjects per cluster, number of clusters, the number of time steps, the baseline effect, the expected treatment effect, expected absolute difference between treatment arms, ICC, and time effect.

References

Zhou X, Liao X, Kunz LM, Normand ST, Wang M, Spiegelman D. A maximum likelihood approach to power calculations for stepped wedge designs of binary outcomes. Biostatistics. 2020 Jan 1;21(1):102-121. doi: 10.1093/biostatistics/kxy031

Hussey, MA AND Hughes, JP. (2007). Design and analysis of stepped wedge cluster randomized trials. Contemporary Clinical Trials 28, 182<U+2013>191.

Examples

# NOT RUN {
# Estimate power for a trial with 3 steps and 9 clusters at the 
# initiation of the study. Those 
# clusters have 14 subjects each with no time effects. 
# We estimated arm outcome proportions of 
# 0.2 (pre-treatment) and 0.31 (post-treatment) and intracluster 
# correlation coefficient (ICC) of 0.05. 
# The resulting power should be 0.7992842.

# }
# NOT RUN {
sw.bin <- cpa.sw.binary(nclusters = 9,
  steps = 3,
  nsubjects = 14,
  timeEffect = 0,
  ICC = 0.05,
  p1 = 0.31,
  p0 = 0.2)
# }
# NOT RUN {
# }