AbstractQuality‐adjusted life‐years (QALYs) calculated from time tradeoff (TTO) based preferences have a time preference component. To impose a conventional discount rate on these implicitly discounted QALYs introduces some degree of double discounting. The purpose of this study was to determine the magnitude of the double discounting effect and the effectiveness of a suggested method for avoiding double discounting in a TTO‐based QALY model. Our analysis used holistic and composite preference scores obtained with the TTO technique in a prior study of four hypothetical treatment paths in type 2 diabetes. Discounted composite preference scores were significantly discrepant from holistic preference scores. Adjusting TTO‐based quality weights prior to external discounting reduced the discrepancy only slightly. Since time preference effects may vary with health state context, the double discounting effect needs further investigation. Copyright © 2002 John Wiley & Sons, Ltd.