ImpactMatters has been acquired by Charity Navigator.

Learn more on charitynavigator.org

ImpactMatters has been acquired by Charity Navigator.

Learn more on charitynavigator.org See an example of the new impact rating

Blog

Tips and information on effective giving

What are Disability-Adjusted Life Years?

Comparing disparate health interventions

How would you compare an intervention that prevents one person from going blind to another that saves the life of a child? Saving a life seems like the clear choice, but does that change once you learn that it costs $50 to prevent blindness and $2,000 to save a child’s life? Does it change if saving the life costs $50,000? If preventing blindness costs $2? Where is the line, and how can you come to know it?

We faced these questions at ImpactMatters. Take our list of Top Health Nonprofits. In ranking the impact of nonprofits within the health field, we needed a metric we could use to compare disparate interventions — one that could tell us where to draw the line between saving a life and preventing “x” cases of blindness. We follow the precedent of the broader health community: To standardize our intervention-specific estimates, we use disability-adjusted life years.1

Disability-adjusted life years (DALYs) measure the years of healthy life lost to a disability or death. “Disability” is a general term here. It describes any condition that impairs the physical or mental health of an individual, ranging in severity from near-sightedness to a parasitic disease. Health interventions aim to avert DALYs, since averting one DALY means giving back a year of healthy life to someone. 

The power of the DALY lies in its ability to capture both ways in which a disability affects a person’s health: length of life and quality of life. Before the DALY, health professionals primarily used mortality to assess the value of an intervention.But mortality is too blunt a tool. Interventions that focus on improving the quality of life — such as interventions that combat blindness — have merit even if they save few lives. The DALY is cleverly designed to capture both dimensions. To calculate DALYs, one takes the sum of the years of life lost to death and the years of life lost to disability.3

Calculating years lost to death is straightforward: If a given condition causes someone to die prematurely, calculate how many extra years of life that person would have lived if not for that condition. 

The nuance of the DALY calculation lies in its second component: years of life lost to disability. Years of life lost to disability takes the years that someone lives with a disability and uses a disability weight to convert it to the equivalent years of death.4 Disability weights answer the question: how many years of living with a disability would it take for someone to prefer dying a year early? If a disability is relatively benign, like mild acid reflux, that conversion rate is low. Indeed, mild acid reflux symptoms have a disability weight of 0.0115; many people would prefer to live with mild acid reflux for 90 years than die a year early. In contrast, disability weights are high for intense disabilities like severe opioid dependence. A disability weight of 0.697 demonstrates that many would prefer dying a year early to experiencing that level of opioid dependence for two years.

Let’s return to the Top Health Nonprofits. We wanted to identify the nonprofits that did the most good with the fewest dollars. To compare apples to apples, we converted dissimilar outcomes like blindness, malnutrition and brain damage into the same metric, DALYs. Sightsavers took the top spot, spending just $50 per case of blindness prevented in India, which roughly translates to $8 per averted DALY. Sightsavers ranks closely to Integrate Health, a community health nonprofit focusing on women and children in Togo. Integrate Health saves a child’s life for $2,000, which roughly translates to $30 per DALY averted. Our use of DALYs enabled us to see that vision care and community health interventions can both be highly cost-effective ways of improving health.

Standardizing disparate outcomes raises weighty, provocative questions. Does the death of an adult feel worse than the death of an infant for you? Do you disagree with the standard rates used to convert years spent with a certain disability to equivalent years of life lost? Should those rates of conversion be universal or are there important differences in the way people experience health conditions across, say, countries, gender and income level? We grapple with these questions too. But while tweaks like age-weighting might, some argue, improve the DALY, it remains the best metric developed to date to quantify and compare health effects, at least within an order of magnitude. Without a standardized metric like it, comparing health interventions would be guesswork: bias-driven and error-prone. The DALY enables us to check our gut intuition. It is a powerful tool, created by thoughtful people to solve a serious problem.

1 While the quality-adjusted life year (QALY) is the most common integrated metric used in assessing the cost-effectiveness of health interventions in high-income countries, most of the health interventions we analyze take place in low-income countries. For health interventions in low-income countries, DALYs are standard. See Chapter 3 of the Institute of Medicine’s 2006 Valuing Health for Regulatory Cost-Effectiveness Analysis for more information on alternative assessment metrics.

2 For a discussion on the history of the DALY, see Chen et al. (2015).

3 For more detail on the DALY calculation and its variations, see the World Health Organization's (W.H.O.) summary.

4 See the W.H.O.’s summary for more information on different methods of using disability weights to calculate years of life lost to disability.

5 The disability weights used here are drawn from the Global Burden of Disease Study 2017 Disability Weights. Weights were determined using a combination of panel-based and case-outcomes based methods. Panel-based methods present to a panel of health professionals or representatives of the general population a vignette describing the effect of a given health condition. Members of the panel then score the health status of interest between zero and one. Case-outcomes methods, on the other hand, survey individuals actually affected by the health condition in question on their generic health status in order to determine weights.