Much of the literature on the value of life is based on the valuation of small reductions in mortality risk with many remaining life years if the fatal outcome is avoided. In contrast, this paper explores valuations of interventions which with varying probability levels offer smaller or moderate life year health gains. We interviewed 2900 respondents about hypothetical therapies that involved life year gains in the range of 1-180 months with a probability of 1 to 1/180, presented both in individual and societal perspectives. The results of the study indicate that the value of the hypothetical treatments is not a simple function of probability and gain. Rather, respondents seem to adopt thresholds when they value treatment offers. This results in kinked utility functions where the expected individual utility is significantly decreased when the gain in life expectancy is 6 months or less, and markedly increased if the probability of gains exceeds 1/12. There were only small differences in valuations across the individual and societal perspectives, suggesting that preferences for dispersion of health gains are not only a reflection of equity considerations. If the results of this study reflect widespread preferences, the standard methods in cost-effectiveness/cost-utility analysis may misinform decision makers.