Govind persad at the podium

When All You Have is a DALY, Everything Looks Like Disease

Recently the Petrie-Flom Center’s annual conference brought together medical experts, bioethics scholars, and disability advocates to rethink how medical systems and public health policies can engage with disability.

During the many fascinating panels at “Beyond Disadvantage: Disability, Law, and Bioethics” one idea that sparked some debate was whether we should use quality/disability adjusted life years (QALY/DALY) to set priorities for who gets access to healthcare resources. Prof. Govind Persad, a bioethicist at Johns Hopkins and blogger for Bill of Health, suggested in his presentation that we should dramatically reduce the role of such utilitarian calculations in priority setting, instead focusing on the genesis of the disadvantage.

However, Prof. Elizabeth Barnes of UVA (who also presented at the conference) responded that while this did answer some of the concerns surrounding the use of DALYs, many disabled individuals are fine with the idea of a utilitarian intervention. The issue, according to Barnes, is simply that DALYs do not accurately reflect quality of life.

Although it didn’t get much discussion at the time, due to the flow of the post-panel Q and A session where DALYs were briefly debated, this topic deserves more attention. In fact, the closer you look, the worse DALYs seem as a determination of an individual’s quality of life.

Consider two hypothetical blind people, one of whom was blind from birth. This person has developed ways to navigate the physical and social worlds without sight, and chose a profession that didn’t require it. The second person was born with sight and dedicated their life to becoming an athlete, only to lose their sight due to disease or accident.

In my opinion, the second individual likely suffers more due to blindness more than the first, because becoming blind interfered with their goals, while they also lacked the coping mechanisms the first person had developed over a lifetime.

Despite this huge variance in potential impact, DALY calculations simply give blindness a disability weight of 0.195. This seems odd to me, because with the exception of blindness, the first person’s life and my own are very similar. However, according to the DALY, a year of his life is worth 4/5s of mine.

Positive psychology suggests that there are five categories of experience that contribute to happiness: pleasure, engagement (with some sort of challenging task), relationships (which studies have shown to be a reliable indicator of happiness), meaning, and accomplishment.

It’s hard to see how blindness alone, with no other intervening factors, would necessarily cause a detriment to any of these categories.

So, if DALY is not actually a measure of happiness of life satisfaction, then what is it?

Prof. Barnes suggested that what the DALY actually measures attitude, or bias for typical function. There is even some direct evidence for this claim. DALYs are calculated by asking predominantly able-bodied individuals whether they would take a treatment that would cure a condition with a varying risk of death, or how much they would shorten their lifespan in order to avoid some disability or illness.

However, as Rebecca Dresser illustrated in her talk about dementia, able-bodied individuals cannot always accurately judge how much a given disability will affect their quality of life, either because of stereotypes, or because they are transformative experiences.

While it is controversial, there are potential economic reasons for how disabilities are weighted when calculating DALYs. It is true that none of the questions used to calculate DALY are directly related to cost or other economic factors, but it cannot be ignored that the disabilities with the largest detriment to DALY (schizophrenia, dementia, etc.), are also those that require the most care.

While this is not necessarily directly related to the economics of care, it is likely influenced by our cultural views about those who are unable to support themselves, how we discuss the prospect of future disability (language like “not wanting to be a burden”), and knowing that our families will have to bear the cost of care.

This makes sense in the context for which DALY was developed as a way to guide public health policy (although this is not to suggest that there are no problems with the way the DALY is calculated). In theory, highly individual factors, such as life goals, should even out when distributed over the entire population. Those people who experience disability as a positive, or who have a strong social network despite their disability, get evened out along with those who are particularly severely harmed.

Furthermore, the DALY is normally used to determine which diseases and disabilities the government should invest in treating or accommodating (as in, the more DALYs gained per dollar spent, the better the investment), not for highly individualized treatments.

However, this can still lead to unjust discrimination against the disabled.

For instance, we might consider that there are some diseases that are highly co-morbid with certain disabilities, like Parkinson’s with dementia. If the co-morbid disease wasn’t common among the general populace, using DALYs to determine how we should fund its treatment would systematically move resources away from disabled populations, thus reducing their access to care, even if they aren’t directly discriminated against.

DALYs are still a controversial measurement, and are open to dispute on the population level, but at least they do have a certain degree of utility. Analyzing the circumstances of each individual would be impossible. However, they are a wholly inappropriate tool to use for individual health care determinations. At best, DALYs are only modestly helpful. And at worst, they can be highly misleading and subject the patient to further discrimination.

 

John Hylton is a student research associate at the Petrie-Flom Center.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.