Information regarding the pharmacokinetic (PK) and pharmacodynamic (PD) properties of a drug provides the basis for optimizing dosing. PK-PD information should be obtained from patients representative of the overall target population, but in many tropical hospitals or health care facilities it may be medically hazardous or logistically difficult for an ill patient or a young child to be sampled repeatedly. Traditional methods used to determine the pharmacokinetic properties of a drug require analysis of a large number of blood samples per subject. However, using modern statistical methods, sparse datasets (i.e. with assay results from only a few, or as little as one blood sample per subject) can now be analysed by a method termed 'the population approach'. Modern assay techniques can often be adapted to small blood volumes allowing finger prick blood samples to be taken. One of the major aims of the population approach is to distinguish and characterize patient and disease contributors to inter-individual variance in drug pharmacokinetics. The purpose of this paper is to explain the basis of the population approach, to highlight its advantages compared to traditional methods of analysis, and to review the application of the population approach to data from field studies of antimalarial drugs. The design of population pharmacokinetic studies is also discussed briefly. The principles discussed in the paper are also applicable to pharmacodynamic data.