Erik Nord

Feb 9 2001


In cost-utility analysis of health interventions, and in burden of disease measurement, health problems of different degrees of severity are assigned numerical scores on a scale from zero to one. The scores are used to weight life years in calculations of Quality Adjusted Life Years (QALYs) and Disability Adjusted Life Years (DALYs). The scores are controversial for a number of reasons. One is that they may be seen as a devaluation of the lives of people with chronic illness or disabilities. While this is not necessarily the intention of the constructors and advocates of QALYs and DALYs, it is certainly the way in which people with chronic illness or disability tend to perceive these approaches.

I believe some of the controversy over health state scores could be avoided if those who develop and use them could be both more clear and more restrictive about what the scores are supposed to mean. More specifically, I think there is a need to distinguish between the value of a life with a given problem, the quality of life experienced with that problem, and the desirability of getting rid of the problem. These are three separate issues. It is a serious mistake to think that they can be measured in the same way or expressed in one single index.

I discuss each of these concepts in the following. But first I explain a little more about QALYs and DALYs.



The use of health state scores in economic evaluation and decision analysis goes back more than thirty years (Chiang, 1965; Fanshell et al, 1970). A scale running from zero (dead or as bad as dead) to unity (full health) was adopted. The more burdensome a condition is to the individual concerned, the lower is its score on the scale. Health state scores are used as weights for life years, such that for instance a year in a state with score 0.8 counts as 0.8 life years, and a given scenario of life years with chronic illness or disability comes out as equivalent to some lower number of life years in full health (Culyer et al, 1971). The down-weighting of years with health problems is called ’quality adjustment’, and life years thus downweighted are called Quality Adjusted Life Years (QALYs) (Patrick et al, 1973). In cost-utility analysis the benefit of a health intervention is expressed as the difference between the number of quality adjusted life years the individual faces with and without intervention. So if for instance an intervention takes a person from 0.6 to 0.8, and he/she get to live 10 years after that, 10x0.2 = 2 QALYs are gained.

The idea of scoring health states and life years on a 0-1 scale was later adopted also in Burden of Disease measurement (World Bank, 1993). Here the scale is turned around to become a severity scale, so that zero represents ’no health problem’ while unity represents being dead or problems equivalent to that. Life years adjusted for health problems are called Disability Adjusted Life Years (DALYs). The DALY is used as a unit of measurement, meaning the value of a healthy life year or the equivalent to that. A life year in a state with a severity score of 0.2 counts as 0.8 healthy years. A gain of such a year thus yields 0.8 DALYs. The burden of disease associated with that year is 0.2 DALYs.

The potential value of QALYs and DALYs lies in helping decision makers to compare the severity of different health problems, and thereby to set priorities in prevention and treatment. The problem with QALYs and DALYs is that if a health problem A is assigned a value of 0.8 (a severity weight of 0.2) in terms of treatment desirability, the following two interpretations are also possible:

    1. Life extending programs for people with problem A are deemed less valuable than similar programs for people in full health, all else equal.
    2. The quality of life of people with problem A is judged to be lower than than of people in full health.

The former proposition violates many peoples ethical views. The latter proposition is inconsistent with the way many people with chronic conditions or disabilities actually feel about the quality of their own life, and therefore insulting.