PZ: Calculate promising zone

In IDDI-BE/CPPZ: Calculate conditional power and promising zone at interim analysis

Description

This function calculates the promising zone boundaries at an interim analysis of a
clinical trial

Usage

PZ(
  r,
  n1 = NULL,
  n2 = NULL,
  alpha_1s,
  eff_null = NULL,
  SE = NULL,
  type,
  pow = NULL,
  f = NULL,
  max,
  plot_effect = TRUE
)

Arguments

r	Proportion of subjects assigned to active group
n1	Number of patients / events at first interim analysis (NULL if parameter f provided)
n2	Number of patients / events at final analysis
alpha_1s	one-sided alpha
eff_null	effect corresponding with null hypothesis (e.g. 1 for hazard rates, 0 for difference) This accomodates for non-inferiority analyses
SE	standard error: has to be provided if type not equal to "HR". If for instance SE =1 then eff_est and eff_planned correspond with z-scores
type	if type="HR", then SE is calculated, if type="general", then SE's have to be provided by user
pow	Power, needed to calculate sample size in second part provided to obtain given power
f	n1/n2 at interim analysis. Only to be provided if n1 and n2 not provided
max	(Maximum sample size)/(original sample size), could be for instance 1.5, 2 or 3
plot_effect	TRUE if plotting boundaries with corresponding effect scale

Value

a list of three vectors

• CP_ll lower boundary of promising zone (z-score, b-value, conditional power, effect scale)

• CP_ul upper boundary of promising zone (z-score, b-value, conditional power, effect scale)

References

Lan and Wittes. The B-Value: A Tool for Monitoring Data. Biometrics 1988;44:579-585
Mehta CR, Pocock SJ. Adaptive increase in sample size when interim results are promising: A practical
guide with examples. Statist. Med. 2011;30:3267–3284

Examples

PZ(r=0.5,n2=180,alpha_1s=0.025,type="HR",max=1.5,pow=0.8,f=0.25)
PZ(r=0.5,n2=180,alpha_1s=0.025,type="HR",max=1.5,pow=0.9,f=0.25)

IDDI-BE/CPPZ documentation built on Oct. 19, 2020, 3:56 a.m.