Stuart L. Beal and Lewis B. Sheiner at UCSF (late 1970s); expanded by Robert J. Bauer at ICON plc
Core Principles & Analysis
Developed in the late 1970s at the University of California, San Francisco by Stuart Beal and Lewis Sheiner, NONMEM introduced nonlinear mixed-effects modeling to estimate population pharmacokinetic parameters from sparse clinical blood sampling.
Models written in the NM-TRAN control language separate fixed effects (population mean clearance and volume of distribution, plus covariates such as body weight, age, and renal function) from random inter-individual (η) and residual (ε) variability.
Regulatory submissions to the FDA and EMA routinely rely on NONMEM estimation methods (FOCE-I, SAEM, and Bayesian MCMC) paired with companion tools such as Perl-speaks-NONMEM (PsN) and Xpose to justify dosing regimens in special populations.