Design a multi-stage group-sequential multi-arm clinical trial for a normally distributed primary outcome

Source: R/des_gs_norm.R

des_gs_norm() determines multi-stage group-sequential multi-arm clinical trial designs assuming the primary outcome variable is normally distributed. It computes required design components and returns information on key operating characteristics.

des_gs_norm(
  K = 2,
  J = 2,
  alpha = 0.025,
  beta = 0.1,
  delta1 = 0.5,
  delta0 = 0,
  sigma = 1,
  ratio = 1,
  power = "marginal",
  stopping = "simultaneous",
  type = "variable",
  fshape = "pocock",
  eshape = "pocock",
  ffix = -3,
  efix = 3,
  integer = FALSE,
  summary = FALSE
)

Arguments

K

A numeric indicating the chosen value for K, the (initial) number of experimental treatment arms. Defaults to 2.
J

A numeric indicating the chosen value for J, the (maximum) number of allowed stages. Defaults to 2.
alpha

A numeric indicating the chosen value for α, the significance level (family-wise error-rate). Defaults to 0.025.
beta

A numeric indicating the chosen value for β, used in the definition of the desired power. Defaults to 0.1.
delta1

A numeric indicating the chosen value for δ1, the 'interesting' treatment effect. Defaults to 0.5.
delta0

A numeric indicating the chosen value for δ0, the 'uninteresting' treatment effect. Defaults to 0.
sigma

A numeric vector indicating the chosen values for σ0 and σ1, the standard deviations of the responses in the control and experimental arms. Must be of length 1 or 2. If of length 1, it is assumed that σ0 =σ1. Defaults to 1.
ratio

A numeric indicating the chosen value for r, the stage-wise allocation ratio to present experimental arms. Defaults to 1.
power

A character string indicating the chosen type of power to design the trial for. Can be "disjunctive" or "marginal". Defaults to "marginal".
stopping

A character string indicating the chosen type of stopping rule. Can be "separate" or "simultaneous". Defaults to "simultaneous".
type

A character string indicating the choice for the stage-wise sample size. Can be "variable" or "fixed". Defaults to "variable".
fshape

A character string indicating the choice for the futility (lower) stopping boundaries. Can be any of "fixed", "obf", "pocock", and "triangular". Defaults to "pocock".
eshape

A character string indicating the choice for the efficacy (upper) stopping boundaries. Can be any of "fixed", "obf", "pocock", and "triangular". Defaults to "pocock".
ffix

A numeric indicating the chosen value for the fixed interim futility bounds. Only used when fshape = "fixed". Defaults to "-3".
efix

A numeric indicating the chosen value for the fixed interim efficacy bounds. Only used when eshape = "fixed". Defaults to "3".
integer

A logical variable indicating whether the computed possible sample sizes required in each arm in each stage should be forced to be whole numbers. Defaults to FALSE. WARNING: If you set integer = TRUE and ratio != 1, obscure results can occur due to difficulties in identifying a suitable whole number sample size that meets the allocation ratio requirement.
summary

A logical variable indicating whether a summary of the function's progress should be printed to the console. Defaults to FALSE.

Value

A list, with additional class "multiarm_des_gs_norm", containing the following elements:

  • A tibble in the slot $opchar summarising the operating characteristics of the identified design.

  • A tibble in the slot $pmf_N summarising the probability mass function of the random required sample size under key scenarios.

  • A numeric vector in the slot $e specifying e, the trial's efficacy (upper) stopping boundaries.

  • A numeric vector in the slot $f specifying f, the trial's futility (lower) stopping boundaries.

  • A numeric in the slot $maxN specifying max N, the trial's maximum required sample size.

  • A numeric in the slot $n_factor, for internal use in other functions.

  • A numeric in the slot $n1 specifying n1, the total sample size required in stage one of the trial.

  • A numeric in the slot $n10 specifying n10, the sample size required in the control arm in stage one of the trial.

  • Each of the input variables.

See also

build_gs_norm, gui, opchar_gs_norm, plot.multiarm_des_gs_norm, sim_gs_norm.

Examples

# The design for the default parameters
des <- des_gs_norm()