OBJECTIVES: Several large scale surveys conducted to measure time-preferences of people in various fields, other than those in the health outcomes, have shown that time-preferences follow non-constant exponential discounting. In pharmacoeconomic studies, however, all outcomes measures including quality-adjusted life-years (QALYs) use discrete discounting. Disability-adjusted life-years (DALYs) is an exception that is discounted using the non-constant exponential discounting approach. The objective of this study is to review the current literature on time-preferences specific to health outcomes and compare the differences between QALYs obtained through discrete discounting and non-constant exponential discounting approaches.  METHODS: We searched PubMed and EconLit for methodological studies examining time-preferences specific to health outcomes. We projected gains of 0.1 QALY/person/year over 1 to 75 years in a hypothetical dataset of 1000 persons. We calculated differences in present values of QALYs obtained through discrete discounting and non-constant exponential discounting approaches, i.e. QALYs from discrete discounting subtracted by QALYs from non-constant exponential discounting, at discount rates of 1.5%, 3%, 5%, and 7%, from 1 to 75 years.  RESULTS: We found no studies that examined discounting approaches specific to health outcomes. The differences in present values of QALYs, at 25, 50, and 75 years, respectively, were: 1) 0.28%, 0.55%, and 0.83%, using a 1.5% discount rate; 2) 1.1%, 2.2%, and 3.3%, using a 3% discount rate; 3) 3%, 5.9%, and 8.7%, using a 5% discount rate;  and 4) 5.7%, 11.1%, and 16.1%, using a 7% discount rate.  CONCLUSIONS: We found no published research comparing discrete discounting to non-constant exponential discounting approaches for QALYs. Over long time horizons, we found small but conceptually important differences between QALYs estimated by these approaches. Therefore, we recommend future studies to address time-preferences specific to determine if non-constant exponential discounting is relevant to health outcomes such as QALYs.