The elderly are generally considered to be different from young people in terms of drug response and this applies particularly to quantitative differences. While altered drug handling is a major potential source of difference in responsiveness to drugs, the relative contribution of pharmacokinetics and pharmacodynamics to this difference is not clear. In the present review we have examined the available data on pharmacokinetics in the elderly.In the past, data pertaining to animal models have been extrapolated to man and in the absence of human experimentation these assumptions have tended to hold sway. This is best exemplified by studies on drug absorption. The absorption of actively transported substances may in fact be diminished in the elderly. However, most drugs are absorbed by passive diffusion and the recently available evidence in man indicates that there is no age-dependent change.While definitive data on the effect of old age on drug metabolising ability in animals is available, no direct assessments have been made in man. Many of the studies carried out using drug plasma half-life and clearance assessments are complicated by changes in distribution. This is best illustrated by a definitive study with diazepam, in which marked prolongation of plasma half-life was accompanied by an increase in apparent volume of distribution in the elderly. This latter change influences plasma drug clearance and, possibly, drug concentration at its site of action. Thus, the implications for drug effect of such changes in volume of distribution remain to be clarified.In theory, the rate of elimination of antipyrine should provide a good index of drug metabolising ability. Both plasma half-life and clearance values suggest a decrease in metabolism in the elderly. No other drug has been studied as intensively and the evidence for a diminished metabolism of other drugs in the elderly is less definite. Thus, while it is likely that the metabolism of some drugs is impaired in old age, it is not possible at this time to generalise with regard to the effect of age on drug metabolising ability in man.It is also difficult to generalise about age-related changes in plasma protein binding of drugs. With some drugs, binding to plasma protein does not appear to be altered and for two drugs - warfarin and phenytoin, the findings of different investigators conflict.Diminution of glomerular filtration rate, renal plasma flow and associated tubular function with age have been well documented. Drug clearance comparisons between old and young have been carried out for only three renally excreted drugs - digoxin, propicillin and sulphamethizole. With digoxin and sulphamethizole, the evidence is that renal excretion is diminished in the elderly. With propicillin, changes in volume of distribution predominate, resulting in higher plasma levels in the elderly but similar percent recovery in urine. In the remaining studies, drug plasma levels and/or plasma half-life values indicate that older patients are exposed to higher plasma concentrations of drugs and while the data is insufficient to explain the findings in kinetic terms, it is likely that diminished renal excretion is mainly responsible.In conclusion, with the exception of renally-excreted drugs, there is at present insufficient data on which to make recommendations with respect to doses of drugs in the elderly. The importance of changes in volume of distribution are not clear and data obtained from single dose studies are not necessarily pertinent to multiple dose continuous therapy. In addition, many elderly patients have multiple pathology and the findings from studies on healthy subjects may not be applicable. Future studies must combine pharmacokinetic and pharmacodynamic aspects in the relevant clinical setting so that the practical significance, if any, of altered kinetics emerges.