Population pharmacokinetics assumptions

M. O. Karlsson, E. N. Jonsson, C. G. WUtse, and J. R. Wade, Assumption testing in population pharmacokinetic models illustrated with an analysis of moxonidine data from congestive heart failure patients. J Pharmacokinet Biopharm 26 207-246 (1998). 

Classical Nonlinear Regression Assumptions with Application in a Population Pharmacokinetic Setting... 

Karlsson and Sheiner have provided excellent examples of the varied assumptions that come into play during population pharmacokinetic analysis. The decomposition of the assumptions into common categories and application of assumption testing to actual data sets are important concepts to grasp, and the manuscript itself is an excellent reference. Table 15.4 summarizes the assumptions discussed in Karlsson et al. and lists testing procedures to examine the validity of these assumptions. 

Assumptions and Assumption Testing Options for Population Pharmacokinetic Analyses... 

FIGURE B-2 Time course for VOC concentrations in blood after a bolus dose. Arrows indicate two of the many time points when a biomonitoring sample may be taken. Without prior knowledge, a useful screening approach is to assume that the sample was taken at the 5-hour arrow, well after the peak blood concentration. If that assumption is used across the population, a pharmacokinetic model can yield reasonably conservative bounding estimates of exposure dose. (Many exposure events might occur in a single day that could affect the concentration-time course of the VOCs in blood.)... 

The use of mixture models is not limited to identification of important subpopulations. A common assumption in modeling pharmacokinetic parameters is that the distribution of a random effect is log-normal, or approximately normal on a log-scale. Sometimes, the distribution of a random effect is heavy tailed and when examined on a log-scale, is skewed and not exactly normal. A mixture distribution can be used to account for the large skewness in the distribution. However, the mixture used in this way does not in any way imply the distribution consists of two populations, but acts solely to account for heavy tails in the distribution of the parameter. 

Box (1976), in one of the most famous quotes reported in the pharmacokinetic literature, stated all models are wrong, some are useful. This adage is well accepted. The question then becomes how precise or of what value are the parameter estimates if the model or the model assumptions are wrong. A variety of simulation studies have indicated that population parameter estimates are surprisingly robust to all kinds of misspe-cifications, but that the variance components are far more sensitive and are often very biased when misspeci-fication of the model or when violations of the model assumptions occur. Some of the more conclusive studies examining the effect of model misspecification or model assumption violations on parameter estimation will now be discussed. 

---