A recent extension of clinical evaluation is "economic evaluation", which seeks to characterise each relevant alternative health care strategy in terms of a summary measure incorporating the costs and benefits of such strategies. In an economic evaluation, separate measurements of resource volumina and resource prices on the cost side, and separate measurements of survival and quality of life effects and valuation of these outcome effects on the benefit side are required. From these effect parameters, which should be calculated for all competing strategies considered in the analysis, the relative cost-effectiveness of one strategy as against the other can be derived. The degree of generalisability of the study results determines the validity of economic evaluation in decision-making. This depends on the generalisability of the clinical findings, and in this respect the so-called "piggyback" economic evaluation, which is added to a clinical trial, has its limitations. In the field of cancer, specific attention should be given to costs and effects occurring after non-mortality endpoints, to patient and family costs and to variations in treatments between settings of care. It is argued that conventional clinical trials and economic evaluations will integrate further in the future.