Disability Weights Elicitation
Also known as: Health-State Valuation, Disability Weight Estimation, Paired-Comparison Disability Weighting, GBD Disability Weights Method
Disability weights elicitation is the methodology for assigning each health state a number between zero and one that represents the level of health loss it entails, where zero is full health and one is a state equivalent to death. These weights are the ingredient that converts time lived in less-than-full health into the years-lived-with-disability component of summary health metrics, but the technique here is the valuation itself rather than the downstream metric. The dominant modern approach, developed for the Global Burden of Disease 2010 study by Salomon and colleagues, abandoned older clinician-based valuations in favor of large population surveys that ask ordinary people to make simple paired comparisons between described health states. Because such comparisons yield only orderings, the method uses a probit regression to recover a latent severity scale and then anchors that scale to the zero-to-one disability-weight interval. The result is a set of weights grounded in common values held across diverse populations, describing functioning loss in terms compatible with the biopsychosocial view of health embodied in the ICF.
Key highlights
- Replaces fragile direct cardinal judgments with easy, reliable paired comparisons, improving respondent consistency and data quality.
- Draws on large, diverse, multi-country population samples so the weights reflect broadly shared values rather than a single expert group.
- Uses a transparent probit model that recovers a full latent severity scale from purely ordinal choice data.
- Frames health states by their functioning consequences, aligning valuation with the ICF's lived-experience view of disability.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use disability weights elicitation when you need defensible numerical weights for health states to feed summary measures of population health, to compare the non-fatal burden of different conditions, or to incorporate functioning loss into cost-effectiveness and priority-setting analyses. The paired-comparison, population-based approach is appropriate when you can field surveys to reasonably large and ideally diverse samples and when you want weights that reflect shared social values rather than expert opinion. It is less suitable when only a handful of respondents are available, when you require a person's own valuation of their individual health for clinical decision making (where preference-based utility instruments fit better), or when the goal is to describe rather than to value functioning — for description the ICF and its core sets are the right tools. The technique values states; it does not by itself produce the burden metric, and it should not be confused with the disability-adjusted life year it helps compute.
Strengths & limitations
- Replaces fragile direct cardinal judgments with easy, reliable paired comparisons, improving respondent consistency and data quality.
- Draws on large, diverse, multi-country population samples so the weights reflect broadly shared values rather than a single expert group.
- Uses a transparent probit model that recovers a full latent severity scale from purely ordinal choice data.
- Frames health states by their functioning consequences, aligning valuation with the ICF's lived-experience view of disability.
- Weights are sensitive to how lay descriptions are written, so framing choices can shift the resulting numbers.
- Anchoring ordinal comparisons to an absolute zero-to-one scale requires additional assumptions and items that are themselves debatable.
- Population valuations may diverge from the lived experience of people who actually have the condition, raising fairness concerns.
- Comparisons describe states in isolation, abstracting away comorbidity, adaptation over time, and context that affect real health loss.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
Why use paired comparisons instead of asking directly for a number?
Because people are poor at producing reliable cardinal numbers for abstract health loss but good at simple comparative judgments. Asking 'which of these two people is healthier?' is an easy, stable task, and answering many such pairs yields far more consistent data than asking someone to state, for example, that a condition equals 0.3 of the loss of death. A probit model then reconstructs the underlying severity scale from the pattern of choices. The cardinal weight is recovered statistically rather than asked for directly, which is more robust.
What do the numbers zero and one mean on a disability weight?
By convention, a disability weight of zero corresponds to full health and a weight of one corresponds to a health state judged equivalent to being dead; intermediate values represent proportional health loss. A weight is not a judgment of a person's worth or of their quality of life overall — it is a population-derived estimate of how much health is lost in a described state. Anchoring the latent scale recovered from paired comparisons to this zero-to-one interval is a distinct estimation step that requires its own assumptions.
How is this different from a disability-adjusted life year?
Disability weights elicitation produces the weights; the disability-adjusted life year is a separate metric that combines those weights with time. Specifically, a weight multiplied by time lived in a state gives years lived with disability, which is added to years of life lost to produce the burden metric. This method stops at valuing the states. Treating the elicited weight as if it were the burden measure conflates an input with the composite indicator built from it.
Sources
- 1.Salomon, J. A., Vos, T., Hogan, D. R., et al. (2012). Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010. The Lancet, 380(9859), 2129-2143.
- 2.World Health Organization. (2001). International Classification of Functioning, Disability and Health: ICF. Geneva: WHO.ISBN 9789241545426
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Disability Weights Elicitation. ScholarGate. https://scholargate.app/disability-studies/disability-weights-elicitation