Target Doses {DoseFinding} R Documentation

## Calculate dose estimates for a fitted dose-response model (via fitMod or bFitMod) or a Mods object.

### Description

The TD (target dose) is defined as the dose that achieves a target effect of Delta over placebo (if there are multiple such doses, the smallest is chosen):

TD_\Delta = \min \{x|f(x) > f(0)+\Delta\}

If a decreasing trend is beneficial the definition of the TD is

TD_\Delta = \min \{x|f(x) < f(0)-\Delta\}

When \Delta is the clinical relevance threshold, then the TD is similar to the usual definition of the minimum effective dose (MED).

The ED (effective dose) is defined as the dose that achieves a certain percentage p of the full effect size (within the observed dose-range!) over placebo (if there are multiple such doses, the smallest is chosen).

ED_p=\min\{x|f(x) > f(0) + p(f(dmax)-f(0))

Note that this definition of the EDp is different from traditional definition based on the Emax model, where the EDp is defined relative to the asymptotic maximum effect (rather than the maximum effect in the observed dose-range).

### Usage

TD(object, Delta, TDtype = c("continuous", "discrete"),
direction = c("increasing", "decreasing"), doses)

ED(object, p, EDtype = c("continuous", "discrete"), doses)


### Arguments

 object An object of class c(Mods, fullMod), DRMod or bFitMod Delta, p Delta: The target effect size use for the target dose (TD) (Delta should be > 0). p: The percentage of the dose to use for the effective dose. TDtype, EDtype character that determines, whether the dose should be treated as a continuous variable when calculating the TD/ED or whether the TD/ED should be calculated based on a grid of doses specified in ‘⁠doses⁠’ direction Direction to be used in defining the TD. This depends on whether an increasing or decreasing of the response variable is beneficial. doses Dose levels to be used, this needs to include placebo, ‘⁠TDtype⁠’ or ‘⁠EDtype⁠’ are equal to ‘⁠"discrete"⁠’.

### Value

Returns the dose estimate

### Author(s)

Bjoern Bornkamp

Mods, fitMod, bFitMod, drmodels

### Examples

## example for creating a "full-model" candidate set placebo response
## and maxEff already fixed in Mods call
doses <- c(0, 10, 25, 50, 100, 150)
fmodels <- Mods(linear = NULL, emax = 25,
logistic = c(50, 10.88111), exponential = 85,
betaMod = rbind(c(0.33, 2.31), c(1.39, 1.39)),
linInt = rbind(c(0, 1, 1, 1, 1),
c(0, 0, 1, 1, 0.8)),
doses=doses, placEff = 0, maxEff = 0.4,