Description Usage Details Value Note Author(s) References Examples
View source: R/DesignHelpers.R
Returns the known study designs for which power and sample size can be calculated within this package.
1 
This function is for informal purposes and used internally for obtaining characteristics of the designs used in calculation formulas.
Returns a data.frame with

number of the design 

character string for identifying the design 

degrees of freedom of the design 

‘robust’ degrees of freedom of the design 

step width in the iterative sample size estimation 

socalled design constant in terms of total n 

design constant in terms of number of subjects in (sequence) groups 
The design character string has to be used in the functions calls for power and sample size.
The design string for higher order crossover designs is named as:
treatments x sequences x periods
in case of replicate designs and
treatments x periods
in case of crossover designs for more then 2 treatments
with number of sequences equal number of treatments.
The df for the replicate crossover designs are those without carryover in the model.
Chen et al. used models with carryover, i.e., one df lower than here.
The design constant bk in case of design 2x2x4 is here bk=1.
Chen et al. used bk=1.1 due to carryover in the model.
n is the total number of subjects for all designs implemented.
df2 = degrees of freedom for the socalled ‘robust’ analysis (aka Senn’s basic estimator).
These degrees of freedom are often also more appropriate in case of evaluation via a ‘true’
mixed model (e.g. the FDA’ for replicate designs).
The design 2x2x2r
is the 2treatment2sequence2period design with
2 repeated targets determined in each period (sequences TTRR or RRTT) described
by Liu. Implemented are the characteristics of this design for the evaluation
via assuming no S×F interaction and equal variabilities of Test and Reference.
D. Labes
Chen KW, Chow SC, Liu G. A Note on Sample Size Determination for Bioequivalence Studies with Higherorder Crossover Designs. J Pharmacokin Biopharm. 1997;25(6):753–65.
Senn S. Crossover Trials in Clinical Research. Chichester: John Wiley & Sons; 2^{nd} edition 2002.
U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Guidance for Industry. Statistical Approaches to Establishing Bioequivalence. January 2001. download
Liu Jp. Use of the Repeated Crossover design in Assessing Bioequivalence. Stat Med. 1995;14(910):1067–78.
1 
no design df df2 steps bk bknif bkni
1 0 parallel n2 n2 2 4.0 1/1 1.00000000
2 1 2x2 n2 n2 2 2.0 1/2 0.50000000
3 1 2x2x2 n2 n2 2 2.0 1/2 0.50000000
4 2 3x3 2*n4 n3 3 2.0 2/9 0.22222222
5 3 3x6x3 2*n4 n6 6 2.0 1/18 0.05555556
6 4 4x4 3*n6 n4 4 2.0 1/8 0.12500000
7 5 2x2x3 2*n3 n2 2 1.5 3/8 0.37500000
8 6 2x2x4 3*n4 n2 2 1.0 1/4 0.25000000
9 7 2x4x4 3*n4 n4 4 1.0 1/16 0.06250000
10 9 2x3x3 2*n3 n3 3 1.5 1/6 0.16666667
11 10 2x4x2 n2 n2 4 8.0 1/2 0.50000000
12 11 2x2x2r 3*n2 n2 2 1.0 1/4 0.25000000
13 100 paired n1 n1 1 2.0 2/1 2.00000000
name
1 2 parallel groups
2 2x2 crossover
3 2x2x2 crossover
4 3x3 crossover
5 3x6x3 crossover
6 4x4 crossover
7 2x2x3 replicate crossover
8 2x2x4 replicate crossover
9 2x4x4 replicate crossover
10 partial replicate (2x3x3)
11 Balaam's (2x4x2)
12 Liu's 2x2x2 repeated xover
13 paired means
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.