sim_test {SimDissolution}

Bootstrap test for the assessment of similarity of drug dissolutions profiles via maximum deviation

Description

Function for testing whether two dissolution profiles are similar concerning the hypotheses H_0: \max_{t\in\mathcal{T}} |m_1(t,\beta_1)-m_2(t,\beta_2)|\geq \epsilon\ vs.\ H_1: \max_{t\in\mathcal{T}} |m_1(t,\beta_1)-m_2(t,\beta_2)|< \epsilon.

$m_1$ and $m_2$ are pharmacokinetic models chosen from a candidate set containing a First order, Hixson-Crowell,Higuchi, Weibull and a logistic model.

See Moellenhoff et al. (2018) <doi:10.1002/sim.7689> for details.

Usage

sim_test(time1, time2 = time1, conc1, conc2, m1, m2, epsilon = 10,
  B = 1000, plot = FALSE)

Arguments

time1, time2

vectors containing the time points of measurements for each of the two formulations; if not further specified the time points are identical in both groups

conc1, conc2

data frames or matrices containing the concentrations obtained for each of the two formulations (see the example)

m1, m2

model types. Built-in models are "firstorder", "hixson", "higuchi", "weibull" and "logistic"

epsilon

positive argument specifying the equivalence threshold (in %), default is 10% corresponding to an f2 of 50 according to current guidelines

B

number of bootstrap replications. If missing, default value of B is 1000

plot

if TRUE, a plot of the absolute difference curve of the two estimated models will be given. The default is FALSE.

Value

A list containing the p.value, the types of models, the f2, the maximum absolute difference of the models, the estimated model parameters, the number of bootstrap replications and a summary of the bootstrap test statistic. Furthermore plots of the two models are given.

References

Moellenhoff et al. (2018) <doi:10.1002/sim.7689>

EMA (2010) <https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-investigation-bioequivalence-rev1_en.pdf>

Examples

data(example_data)
conc1 <- select(filter(example_data,Group=="1"),-Tablet,-Group)
conc2 <- select(filter(example_data,Group=="2"),-Tablet,-Group)
time <- c(10,15,20,30,45,60)
sim_test(time1=time,time2=time,conc1=conc1,conc2=conc2,m1="logistic",m2="logistic",B=500,plot=TRUE)