Discounting Health

The application of a discount rate to future health outcomes within an economic evaluation, converting them into a common present-day value.

Source: Parsonage & Neuburger 1992

Because the justification that applies to money does not transfer straightforwardly. Money can be invested and grow, so deferring a cost genuinely reduces the sacrifice. Health cannot be banked in the same way, and a year of healthy life in twenty years is not obviously worth less than a year now to the person living it. Discounting health therefore requires a different argument, usually that society weighs health accruing later less heavily, or that health will be cheaper to produce in future because technology improves.

Source: Parsonage & Neuburger 1992

Interventions producing benefits far into the future gain enormously in apparent value, since a life saved in fifty years counts the same as one saved today. Taken to its conclusion, any programme with perpetual benefits would dominate every alternative, which most people regard as a reductio. Not discounting health while discounting costs also creates a paradox in which delaying a health-producing programme indefinitely appears increasingly attractive, because its costs shrink in present value while its benefits do not.

Source: Drummond et al. 2015

Most frameworks now specify equal rates, principally on consistency grounds, since differing rates generate the delay paradox and imply that the value of health relative to money changes over time. The argument for a lower rate on health rests on the observation that the two are different kinds of thing valued for different reasons. Practice has moved towards equal rates as the reference case, with differential rates retained in sensitivity analysis where the horizon is long enough for the choice to matter.

Source: Drummond et al. 2015

Those whose benefits accrue over decades: childhood vaccination, screening programmes, smoking cessation, and any preventive intervention whose costs fall now. The effect can be large enough to move such programmes from clearly cost-effective to clearly not, depending on the rate applied, which means the rate frequently determines the conclusion rather than the evidence. Interventions producing immediate benefit are barely affected, so the choice of rate systematically alters the relative standing of prevention against treatment.

Source: healtheconomics.wiki

The rate applied to health should be stated separately from the rate applied to costs, even where they are identical. Results should be presented at the framework rate, at zero, and at any alternative the analysis regards as defensible, so a reader can see how much the conclusion depends on the choice. Where the recommendation changes across that range, it should be reported as a finding rather than buried in a sensitivity table, since the decision then rests on a methodological judgement rather than on the clinical evidence.

Source: Drummond et al. 2015