Quiz 3. Introduction to popPK

What are “mixed effects” in population PK?

• A combination of drug effects and side effects in different individuals

• A combination of both fixed effects (variability) and random effects (typical values)

• A combination of both fixed effects (typical values) and random effects (variability)

• A combination of data from different studies

Answer explanation

Mixed effects models combine fixed effects (typical PK parameters for the population) with random effects (variability). This allows the model to describe both the average trend and individual differences in drug behaviour.

What are the core components of a typical popPK model?

• Dose, concentration, and half-life

• Structural model, covariate model, and statistical model

• Fixed effects, random effects, and unexplained effects

• Software platform, dataset, and objective function value

Answer explanation

A typical popPK model is made up the structural model, the covariate model, and the statistical model.

It is part of the fixed effects:

• Structural model and covariate model

• Statistical model

• Structural model and statistical model

• Statistical model and covariate model

Answer explanation

The fixed effects in a popPK model describe the typical values of PK parameters in the population and are defined through the structural model and the covariate model.

Represents variability between individuals in a popPK model:

• Between-subject variability (BSV)

• Residual variability

• Inter-occasion variability (IOV)

• Interindividual variability (IIV)

Answer explanation

Interindividual variability (IIV) refers to differences in PK parameters between individuals in a population. It is also known as between-subject variability (BSV).

Which of the following statements best describes shrinkage in a popPK model?

• It reflects measurement error in concentration data

• It shows how much individual estimates are pulled toward the population average

• It measures how much drug is eliminated

• It indicates when covariates are statistically significant

Answer explanation

Shrinkage occurs when the data are not informative enough (due to sparse sampling or high variability) to estimate individual parameters accurately. In such cases, the model “shrinks” individual estimates back toward the population mean.

What is the purpose of the residual variability in a popPK model?

• To describe typical PK parameters

• To model structural model error

• To account for noise or unexplained variability in observed concentration

• To identify appropriate covariates

Answer explanation

Residual variability captures measurement error, biological fluctuations, data handling issues, and other unexplained variability.

Which of the following exponents is used for allometric scaling of CL?

• 1

• 0.75

• 0.5

• 0.25

Answer explanation

CL is scaled with a fixed exponent of 0.75 while Vd is scaled with an exponent of 1

In allometric scaling, we only include weight as a covariate if it is statistically significant:

• True

• False

Answer explanation

False, weight is typically included by default using allometric scaling, regardless of statistical significance.

What does the number 98 in the equation represents?

• The average drug concentration

• The typical CL in the population

• The median CrCL in the population

• The exponent describing the strength of the covariate effect

Answer explanation

The number 98 represents the median CrCL in the dataset. It’s used to normalise the covariate so that θ_CL reflects the typical CL for a patient with CrCL of 98.

What is the main purpose of covariates in a popPK model?

• To describe unexplained variability between individuals

• To account for measurement error in the observed data

• To define the structural behaviour of the drug in the body

• To explain predictable differences in PK parameters across individuals

Answer explanation

Covariates are used to explain predictable, often biologically plausible differences in PK parameters. Unlike random effects, which account for unexplained variability, covariates help reduce it by incorporating known factors.