Proportional Heuristics in Time Tradeoff and Conjoint Measurement

Abstract

The time-tradeoff (TTO) test is widely used to measure quality of life for different health states. Subjects are asked to equate the value of living a given period in an inferior health state to the value of living a shorter period in good health. Applications of TTOs have been criticized based on the fact that the value of future life duration is taken as the future life duration itself. The authors show that for a health state in which a subject does not want to live longer than a specified amount of time, subjects' responses do not comply with the assumption that the value of the period in inferior health is equated to the value of the shorter period in good health. Actually, preference reversals with respect to such a health state point to the use of a proportional heuristic in the TTO test. Comparisons of the TTO test in these subjects with category scaling and difference measurements also favor a proportional inter pretation of the TTO test. In tests based on conjoint measurement, these subjects also appear to use a proportional heuristic. Consequences of the use of the TTO test and conjoint measurement in quality-of-life models are discussed.

The gap effect: discontinuities of preferences around dead

“…Such a preference is called a maximal endurable time (MET) preference [15]. This phenomenon has been observed in a wide variety of populations including students [16,17], patients [14], and physicians [15]. When MET preferences apply, independence between duration and quality (as implied by the QALY model V ¼ L Â Q) does not hold because the slope or quality weight for state B i decreases from positive to negative with increasing duration.…”

A consistency test of the time trade-off

“…Stalmeier, Bezembinder, and Unic (1996), Stalmeier, Wakker, and elicitation task vary. Time trade-off measurements that use health states clearly preferred to death and that do not vary the endpoints, which is the common case in cost-utility analysis, appear to be on much firmer ground.…”

A Direct Method for Measuring Discounting and QALYs More Easily and Reliably

“…[2][3][4][5] Only few attempts have been made to avoid the interaction between discounting and quality of life, and these invoke extraneous factors such as risky or interpersonal (utilitarian) aggregations. 1,[6][7][8][9][10] Then attitudes toward risk and welfare intervene and generate additional biases. [11][12][13][14][15][16] This article presents a method to measure discounting within the QALY model that avoids the aforementioned problems: Our method (a) needs no extraneous factors, (b) is not affected by the interactions between discounting and quality of life, and (c) uses stimuli that can be simpler and more realistic than those for existing methods.…”