add_covariate_effect {pharmr}

add_covariate_effect

Description

Adds covariate effect to :class:pharmpy.model.

The following effects have templates:

• Linear function for continuous covariates (lin)

• Function:

math::

coveff = 1 + theta * (cov - median)

• Init: 0.001

• Upper:

• If median of covariate equals minimum: :math:⁠100,000⁠

• Otherwise: :math:⁠frac{1}{{median} - {min}}⁠

• Lower:

• If median of covariate equals maximum: :math:⁠-100,000⁠

• Otherwise: :math:⁠frac{1}{{median} - {max}}⁠

• Linear function for categorical covariates (cat)

• Function:

• If covariate is most common category:

math::

coveff = 1

• For each additional category:

math::

coveff = 1 + theta

• Init: :math:0.001

• Upper: :math:⁠100,000⁠

• Lower: :math:⁠-100,000⁠

• Piecewise linear function/"hockey-stick", continuous covariates only (piece_lin)

• Function:

• If cov <= median:

math::

coveff = 1 + theta1 * (cov - median)

• If cov > median:

math::

coveff = 1 + theta2 * (cov - median)

• Init: :math:0.001

• Upper:

• For first state: :math:⁠frac{1}{{median} - {min}}⁠

• Otherwise: :math:⁠100,000⁠

• Lower:

• For first state: :math:⁠-100,000⁠

• Otherwise: :math:⁠frac{1}{{median} - {max}}⁠

• Exponential function, continuous covariates only (exp)

• Function:

math::

coveff = exp(theta * (cov - median))

• Init:

• If lower > 0.001 or upper < 0.001: :math:⁠frac{{upper} - {lower}}{2}⁠

• If estimated init is 0: :math:⁠frac{{upper}}{2}⁠

• Otherwise: :math:0.001

• Upper:

• If min - median = 0 or max - median = 0: :math:100

• Otherwise:

math::

min(fraclog(0.01)min - median, fraclog(100)max - median)

• Lower:

• If min - median = 0 or max - median = 0: :math:0.01

• Otherwise:

math::

max(fraclog(0.01)max - median, fraclog(100)min - median)

• Power function, continuous covariates only (pow)

• Function:

math::

coveff = (fraccovmedian)^theta

• Init: :math:0.001

• Upper: :math:⁠100,000⁠

• Lower: :math:-100

Usage

add_covariate_effect(
  model,
  parameter,
  covariate,
  effect,
  operation = "*",
  allow_nested = FALSE
)

Arguments

model

(Model) Pharmpy model to add covariate effect to.

parameter

(str) Name of parameter to add covariate effect to.

covariate

(str) Name of covariate.

effect

(str) Type of covariate effect. May be abbreviated covariate effect (see above) or custom.

operation

(str) Whether the covariate effect should be added or multiplied (default).

allow_nested

(logical) Whether to allow adding a covariate effect when one already exists for the input parameter-covariate pair.

Value

(Model) Pharmpy model object

Examples

## Not run: 
model <- load_example_model("pheno")
model <- add_covariate_effect(model, "CL", "APGR", "exp")
model$statements$before_odes$full_expression("CL")

## End(Not run)