Education And Debate Economics notes

Discounting

BMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7214.914 (Published 02 October 1999) Cite this as: BMJ 1999;319:914
  1. David J Torgerson, senior research fellowa,
  2. James Raftery, professorb
  1. a Department of Health Studies and Centre for Health Economics, University of York, York YO1 5DD
  2. b Health Economics Facility, University of Birmingham, Birmingham B15 2RT

    Until recently it has been common practice in economic evaluations to “discount” both future costs and benefits, but recently discounting benefits has become controversial. Discounting makes current costs and benefits worth more than those occurring in the future because there is an opportunity cost to spending money now and there is desire to enjoy benefits now rather than in the future. The reason why current spending incurs an opportunity cost relative to delayed spending is that a monetary investment yields a real rate of return and therefore there is a cost to spending money in the present.

    For example, if £100 were invested with a nominal return of 10%, in one year's time it would be worth £110; if inflation was 4% this would result in a real return of £6 on every £100 invested. If for some reason £100 of healthcare spending were delayed for one year then (assuming prudent investment) we could expect that in one year's time we would have £106 for healthcare investment.

    To take into account the opportunity cost of investing now rather than waiting one year we have to discount future costs. Therefore, if two healthcare interventions both released £100 in savings but for one we had to wait a year, then, all other things being equal, we would adopt the intervention that saved £100 now. This is because the £100 released now, if invested, would produce an extra £6 in a year's time (with a discount rate of 6%).

    Failure to discount the future costs in economic evaluations can give misleading results. For example, an evaluation of cystic fibrosis screening revealed a cost of £80 000 for detecting and terminating one affected pregnancy.1 This cost was compared with the future excess costs of treating an individual with cystic fibrosis, which was estimated to be £5000 a year over 25 years. As cystic fibrosis screening benefits (£125 000) outweighed the costs (£80 000) it was concluded that screening represented good value for money. However, if the averted costs had been discounted (at 6%) then these would have been only £63 917, which alters the study's results (though not if the discount rate were only 4%).

    Discounting future costs is uncontroversial and until recently so was the process of discounting health related benefits. The main argument against discounting health benefits is that health, unlike wealth, cannot be invested to produce future gains.2 The Department of Health has thus recommended that health related benefits should not be discounted,3 4 though more recent advice suggests future health benefits should be discounted but at a very low rate of 1.5%-2%.5

    An important reason for discounting future costs and benefits is “time preference,” which refers to the desire to enjoy benefits in the present while deferring any negative effects of doing so. Examples of human behaviour which implicitly discount future health effects abound. For instance smoking and drinking give current pleasure while incurring future (discounted) detrimental health effects. Indeed, research has indicated that smokers value future health benefits at a lower rate than non-smokers.6 This desire to enjoy pleasurable benefits in the present time is often reflected in differential pricing of goods and services. Consider the hire of a video for home viewing. Despite the increased cost of newly released videos, many people are willing to pay the extra cost rather than wait until the price falls.

    Effects of discounting on the cost effectiveness of hip fracture prevention

    View this table:

    Failure to discount future health related benefits will tend to show more favourable cost effectiveness ratios compared with discounting. For instance, an evaluation of two view mammography for breast screening showed an undiscounted marginal cost per life year of £1200.7 However, discounting the life years (at 6%) increased the marginal cost per life year by 74%, to £2092.

    If future health benefits are not discounted this implies that health gains achieved this year and those achieved in 20 or 30 years are of equal value. As an example, let us assume that about £70 000 is available for hip fracture prevention in 100 women, and there are two strategies under consideration: 10 years of hormone replacement therapy (given to 50 year old women), which prevents 50% of fractures in 30 years' time; or 10 years of calcium and vitamin D (given to 70 year old women), which prevents 30% of hip fractures in 10 years' time.

    In the table we show how discounting health benefits alters the relative cost effectiveness of the two interventions to prevent hip fracture Without discounting, hormone replacement therapy produces a lower cost per avoided hip fracture than vitamin D and calcium. On the other hand if the hip fracture benefits are discounted, then the reverse is true.

    In this example the decision on which preventive strategy to adopt is heavily influenced by discounting. Whether we discount health benefits, and at what rate, depends on how much value society places on current health benefits compared with future benefits. Intuitively it would seem best to be able to prevent hip fractures in the next 10 years rather than wait 30 years for this health benefit What little research there has been into society's preferences about current health benefits compared with future ones suggests that people value current health benefits more highly than future ones and that people actually discount future health gains more highly than future wealth gains.6 Many economists still hold the view that future health gains should be discounted.9 In most economic evaluations the choice of discount rate will not affect the relative ranking of the interventions under evaluation. However, it is good practice to establish whether the evaluation results are critically affected by the discount rate by a sensitivity analysis using different discount rates.

    Footnotes

    • These notes are edited by James Raftery (J.P.RAFTERY{at}bham.ac.uk)

      This is the eighth in this series of notes

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial

    Subscribe