Absolute Shortfall

Absolute Shortfall and severity measured in QALYs lost

Absolute shortfall is one of two measures NICE uses to decide whether a condition is severe enough for the health gains of a new treatment to count for more. The NICE manual (PMG36) sets cut-offs of 12 and 18 discounted QALYs for it, alongside cut-offs for proportional shortfall. The answer can raise the weight given to a technology's QALY gains by 20 or 70 per cent, which can change a funding decision.

NICE combines absolute shortfall with proportional shortfall to set a severity weight, the more severe of the two measures determines that weight, and the weight then enters the cost-effectiveness decision. The approach rests on value judgements and has limitations of its own.

Why severity is measured

In its reference case, NICE treats all QALYs as being of equal weight. Its methods manual (PMG36) allows decision modifiers to be taken into account, either qualitatively through committee discussion or quantitatively through QALY weighting, and in technology appraisals severity is the modifier applied as an explicit QALY weight. NICE defines severity as the future health lost by people living with the condition with standard care in the NHS.

The severity modifier was introduced in 2022 and replaced the end-of-life criteria that, since 2009, had allowed extra weight for treatments extending life in patients with very short life expectancy. NICE's methods review describes the new modifier as designed to be of similar overall size to the end-of-life modifier, so that it would be broadly neutral in opportunity cost.

How absolute shortfall is calculated

PMG36 defines absolute QALY shortfall as the total QALYs that people with the condition would be expected to have with current treatment over their remaining lifetime, subtracted from the total QALYs that a general population with the same age and sex distribution would be expected to have. The NICE Decision Support Unit guide to calculating severity shortfall writes this as a formula.

$$\text{AS} = Q_h - Q_c, \qquad \text{PS} = \frac{Q_h - Q_c}{Q_h}$$

where AS is the absolute shortfall, PS is the proportional shortfall, $Q_h$ is the expected remaining QALYs of the age- and sex-matched general population and $Q_c$ is the expected remaining QALYs of the people with the condition under current care. Both measures are forward-looking. The guide notes that they consider health from the point at which the technology is being considered, not past health.

The general-population QALYs come from life expectancy in national life tables combined with population EQ-5D values; the Decision Support Unit guide recommends period life tables for England for 2017 to 2019 from the Office for National Statistics. PMG36 requires both shortfall calculations to be discounted at the reference-case discount rate, and the guide states that carer impacts should be excluded. McNamara and colleagues publish quality-adjusted life expectancy norms for the English population with a companion calculator.

The condition QALYs are normally taken from the comparator arm of the committee's preferred cost-effectiveness model, so its survival and utility assumptions drive the shortfall. Where the model does not use a lifetime horizon, the comparator arm needs further extrapolation, and where several comparators or subgroups are relevant, shortfall is estimated for each.

From shortfall to a QALY weight

NICE considers absolute and proportional shortfall together and applies the weight implied by whichever shows the greater severity. If either measure falls exactly on a cut-off, the higher severity level applies. The weights are set out in Table 6.1 of PMG36.

The Decision Support Unit guide states that the weights are applied to the incremental QALYs of the new technology. PMG36 states that, when decision modifiers are applied, net health benefit should be presented at values of £25,000 and £35,000 per QALY, both with and without the QALY weighting, to show the effect on opportunity cost.

Worked examples from a DSU case and a young population

The first example is the hypothetical case in the Decision Support Unit guide, using EQ-5D-3L population values. A population with an average age of 62, of whom 33 per cent are female, would be expected to have 11.96 discounted QALYs in the general population, and has 6.031 discounted QALYs under current care.

$$\text{AS} = 11.96 - 6.031 = 5.93, \qquad \text{PS} = \frac{5.93}{11.96} = 0.50$$

where both results fall below the lowest cut-offs, so the QALY weight is 1. Because the matched population has only 11.96 discounted QALYs at age 62, absolute shortfall could not reach the 12-QALY cut-off even if the condition caused immediate death, so for older populations only proportional shortfall can deliver a weight above 1. With the EQ-5D-5L value set that NICE adopted in August 2026 for new topics (topics already under way continue with 3L values, under the interim methods statement PMG51), the guide reports 12.69 general-population QALYs, an absolute shortfall of 6.66 and a proportional shortfall of 0.52, which leaves the weight at 1. The second example is illustrative. A younger population has 20 remaining discounted QALYs in the general population but only 1.5 with the condition.

$$\text{AS} = 20 - 1.5 = 18.5, \qquad \text{PS} = \frac{18.5}{20} = 0.925$$

where the absolute shortfall of 18.5 QALYs meets the threshold for a weight of 1.7, while the proportional shortfall of 0.925 would qualify only for 1.2. Because NICE uses whichever measure implies greater severity, the weight is 1.7.

Absolute and proportional shortfall compared

The two measures treat age differently. General-population QALYs are higher at younger ages, so a condition that removes the same share of remaining health produces a larger absolute shortfall in a younger population, and the largest possible absolute shortfall, on immediate death, equals the general-population QALYs, which fall with age. Proportional shortfall scales the loss by the remaining QALYs, so it gives more similar values across ages; in the extreme case of immediate death, a 10-year-old and an 80-year-old both have a proportional shortfall of 1.

NICE's methods review notes that both measures are used by other HTA bodies and that using them together mitigates the limitations of each. An analysis of NICE decisions by Wailoo and colleagues found that all decisions in the highest absolute shortfall category involved populations with a starting age below 20, and concluded that using both measures makes technologies for children and young adults more likely to receive extra weight. Norway formalises severity as absolute shortfall, while proportional shortfall has a longer history in the Netherlands, where Stolk and colleagues proposed it as combining the fair innings and severity-of-illness views of equity.

What the weight does to a decision

A severity weight effectively raises the cost-effectiveness threshold for the technology concerned, because each QALY it gains counts for more. A weight of 1.7 is equivalent to judging the unweighted cost per QALY against £42,500 to £59,500 rather than £25,000 to £35,000. The weight therefore matters for technologies whose unweighted cost per QALY lies above the normal range but within that range multiplied by the weight. The Decision Support Unit guide notes that when shortfall estimates are close to one of the cut-offs, judgement will be required, and NICE's methods review chose a two-step modifier rather than a sliding scale partly because shortfall estimates are inevitably uncertain.

The modifier does not apply everywhere. PMG36 states that it is not applied to highly specialised technologies, is not initially applied to technology appraisals of HealthTech, and that a severity weight is unlikely to be relevant in diagnostics evaluations. Technologies recommended after applying the modifier are considered relevant comparators for future evaluations in the same condition, provided they are recommended for routine use and represent established NHS practice, so they form part of the current care against which later shortfalls are measured.

Value judgements and criticisms

Weighting QALYs by severity is a value judgement. It gives priority to people with more future health to lose, at the opportunity cost of health for other patients whose care is displaced. Absolute shortfall looks only forward and so does not count health already lost. Hausman argues that NICE's severity weights are not well supported and have troubling implications, and that they find little support in egalitarian or prioritarian philosophy because they look only at future health rather than well-being in general. General-population life tables reflect only age and sex, so where a condition mainly affects people in more deprived groups, the Decision Support Unit guide notes that shortfall will overstate severity.

NICE's Board reviewed the modifier in 2024 and concluded that it was working as intended. Wailoo and colleagues found no statistically significant difference in the mean severity weight between appraisal decisions before and after 2022 (1.116 and 1.125), consistent with the aim that the new modifier should be similar in size to the end-of-life criteria it replaced.

Common errors in shortfall calculations

A common error is to calculate shortfall without discounting, or with a non-reference-case discount rate, when PMG36 requires the reference-case rate of 3.5 per cent, and the Decision Support Unit guide states that this applies even where the committee accepts a non-reference-case rate for the cost-effectiveness analysis. Another is to compare the condition with a general population that does not match its age and sex mix, which can move the estimate across a cut-off.

A third is to apply the weight to total rather than incremental QALYs. The weight multiplies the QALYs the new technology gains over current care, not the QALYs patients would have anyway.

Sources

• National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). Published 2022; updated 2026. Sections 6.2.10 to 6.2.21 and Table 6.1.
• Wailoo A. NICE DSU Technical Support Document 23: A guide to calculating severity shortfall for NICE evaluations. NICE Decision Support Unit, University of Sheffield. January 2024, updated September 2026.
• National Institute for Health and Care Excellence. Review of methods, processes and topic selection for health technology evaluation programmes: conclusions and final update. Appendix: further discussion and rationale for conclusions, methods. 2022.
• McNamara S, Schneider PP, Love-Koh J, Doran T, Gutacker N. Quality-adjusted life expectancy norms for the English population. Value in Health. 2023;26(2):163-169.
• Stolk EA, van Donselaar G, Brouwer WB, Busschbach JJ. Reconciliation of economic concerns and health policy: illustration of an equity adjustment procedure using proportional shortfall. PharmacoEconomics. 2004;22(17):1097-1107.
• Reckers-Droog VT, van Exel NJA, Brouwer WBF. Looking back and moving forward: on the application of proportional shortfall in healthcare priority setting in the Netherlands. Health Policy. 2018;122(6):621-629.
• Wailoo A, Bell Gorrod H, Dunning L, Kenny J, Leckenby E, Shah K. The NICE experience of designing and utilising severity weights. Health Policy OPEN. 2026;10:100156.
• National Institute for Health and Care Excellence. Interim methods statement: implementing the EQ-5D-5L value set (PMG51). Published 27 August 2026. Sections 1.10, 1.17 and 1.18.
• Hausman DM. Problems with NICE's severity weights. Social Science and Medicine. 2024;348:116833.
• National Institute for Health and Care Excellence. NICE Board says new method allowing greater weight to be given to severe diseases is working. News release. 25 September 2024.