Age Weighting

An early methodological adjustment to the DALY calculation valuing healthy years lived in young to middle adulthood more highly than years in childhood or old age.

Source: Murray 1994

The argument advanced for it was that a year of healthy life at different ages carries different social value because of the roles people occupy, with young and middle-aged adults supporting dependants both younger and older than themselves, so that ill health at those ages carries consequences beyond the individual. Supporting evidence was drawn from studies reporting that people asked to choose between saving lives at different ages tended not to treat all ages equally. The adjustment was presented as reflecting an observed social preference rather than as a claim about the intrinsic worth of individuals.

Source: Murray 1994

A continuous weighting function was applied to each year of life, rising from a low value in infancy to a peak in early adulthood and declining gradually through later life, so that years lost or lived with disability were counted at more or less than one according to the age at which they fell. Applied to a burden estimate, it raised the recorded burden of conditions affecting working-age adults relative to those concentrated in early childhood and old age. It was applied alongside a separate adjustment discounting years occurring further in the future.

Source: Murray 1994

The revision of the global burden estimates published in 2010 removed both age weighting and discounting, on the grounds that they introduced value judgements into a measure intended to describe the burden of disease rather than to prescribe priorities. The empirical basis had also been questioned, since the preference studies underlying it were few, drawn from limited populations and sensitive to how the choices were framed. The ethical objection carried substantial weight: a measure counting a year of healthy life differently according to the age of the person living it treats people unequally on a characteristic they do not choose.

Source: Salomon et al. 2012

Both were adjustments to the value attached to a year of healthy life, one according to when in a lifetime it occurred and the other according to how far in the future it fell, and both were applied as standard in the earlier estimates. They interact, since discounting reduces the weight of years lost far ahead and age weighting reduces the weight of years lived in old age, and together they substantially lowered the recorded burden of conditions causing death in childhood. Removing both raised the relative burden of those conditions.

Source: Anand & Hanson 1997

Burden estimates published before the revision are not comparable with those published after it, so a change in a country's recorded burden across that boundary may reflect the method rather than any change in health. Analyses and priority-setting exercises built on the earlier figures carry the embedded weighting whether or not their authors intended it. The episode also illustrates a general point about composite measures, that adjustments introduced for defensible reasons become invisible once the measure is in routine use, and are then difficult to identify in the conclusions drawn from it.

Source: healtheconomics.wiki