'SH-SY5Y cells cultured on a 384 microwell plate', P. Denner 'SH-SY5Y cells cultured on a 384 microwell plate', P. Denner

What Would it Mean to End Disease?

A couple of months ago, Priscilla Chan and Mark Zuckerberg, via their philanthropic Chan Zuckerberg Initiative (CZI), pledged $3 billion over the next decade to help fund medical research with the ultimate goal of curing, preventing, or managing all diseases by the end of century. While researchers and administrators within the medical community have expressed practical concerns over the plausibility of meeting these goals, I will focus on some conceptual concerns.

Attempting to cure, prevent, or manage1 all diseases requires identification of what diseases are. Doing so in a satisfactory way tends to be a more difficult task than what might be expected. If diseases are conditions that CZI is motivated to cure, prevent, or manage, then plausibly, they are conditions that warrant medical attention. By medical attention, there are two main things that I have in mind: individual treatment and biomedical research, or the development of interventions of the sort that the CZI hopes to fund.

What we identify as conditions of disease, then, are conditions that we should feel justified in treating and researching. What we’re looking for is some set of rules whose observation can direct us to conditions that medicine should treat and research.

While he doesn’t offer such rules himself, Norman Daniels offers some guidance about what makes a disease a condition worth treating. In his discussion of health care, Daniels argues that medical institutions are unique in their ability to entirely equalise certain opportunities among individuals. This is because adequate health care can fully restore normal functioning in cases where such functioning is impaired. At this point, then, we can think of a disease as a condition that should be treated and researched because diseases are impairments in normal functioning that compromise people’s opportunity to participate in society.

The concept of normal functioning is, itself, a debated topic and it has been interpreted in a number of ways. Daniels endorses one account of disease offered by Christopher Boorse. In Boorse’s account, the ‘function’ of normal function can be attributed to a body part (e.g. the heart) or a system of body parts (e.g. the reproductive system) and is its contribution to a person’s survival or to reproduction. The ‘normal’ of normal function is statistical normality determined according to a person’s natural reference class (e.g. age or sex) whose members all share a functional design. Conversely, for Boorse, a disease is any reduction in function below normal efficiency.

There are a number of issues with using this account for our present purposes – again, the identification of diseases that medical research can and should cure, prevent, or manage.

First, doing so requires a precise understanding of reference class, which Boorse defines in terms of natural uniform functional design. As Elselijn Kingma observes, it’s not obvious what we are to take as natural uniform functional design. Indeed, it can be quite difficult to do so without making prior assumptions about what disease is – the very concept that Boorse’s account is proposed to explicate. As a result, the determination of the appropriate reference class against which to base statistical atypicality ends up being difficult to disambiguate.

However, even if we had clear and appropriate criteria for identifying the set of functional reference classes, it’s not clear how to determine statistical atypicality. As Boorse admits, the point where a given body’s functional capacities are reduced below normal function is determined arbitrarily. This issue might be less pressing when Boorse’s account is used simply to describe what diseases are. However, this issue severely compromises this account’s use for that actual identification of diseases.

Aside from the difficulties of using this account to identify diseases, a third concern arises – one that is particularly relevant to CZI’s stated goal: if we understand disease as a statistical entity, disease will necessarily persist so long as there is functional variation among people in the world. The group of conditions that we identify as diseases will be a continually moving target based on the distribution of functional capacities of different bodies in a given population.

'Typical Laws of Heredity, III', Sir Francis Galton
‘Typical Laws of Heredity, III’, Sir Francis Galton

Outside of articulating a conceptual definition of disease, we can identify all diseases by looking to whatever a relevant institution, for example, the World Health Organization (WHO) identifies as the cases of disease. Indeed, the WHO publishes a manual of this very sort: The International Classification of Diseases and Related Health Problems (ICD), which is currently in its 10th edition, and which was endorsed by the WHO’s World Health Assembly in 1990 (work on the 11th edition is well underway and its publication is expected in 2018)2.  

Each edition of the ICD has introduced changes to the collection of diseases included in its predecessor. There are at least two reasons for these changes: the discovery of health conditions that had been present in a given population and the appearance of new health conditions that had not existed previously in any population. Accordingly, curing, preventing, and managing all disease requires that we have discovered all health conditions in a given population as diseases and that no new health conditions will arise in the future. Accomplishing these tasks is difficult in light of Sheena Cruickshank’s observation that, as our immune systems change, so do the diseases with which we are saddled. So, again, we have the aforementioned moving target problem rendering the goal of ending all diseases very tricky to meet.

To be clear, I am not criticising CZI’s funding of medical research. Indeed, given the uncertainty regarding the future budgets of public funding bodies like the National Institutes of Health, the world’s largest funding body for biomedical research ($32.3 billion per year), private funding of the sort that CZI is contributing may become much more important moving forward.

I also want to clarify that by raising conceptual concerns about the cure, prevention, or management of all diseases, I am not at all expressing doubts about the cure, prevention, or management of particular diseases. Scepticism about the eradication of disease does not imply scepticism about the discovery of a cure for cancer or AIDS. Indeed, I think that CZI’s contribution may get the medical community closer to such cures, and hopefully others as well.

So then, what is the point, here? I think that by raising concerns about the very idea of ending disease, I’ve also raised concerns about corresponding ideas of medicine and what this institution is for. Rather than thinking about medicine as a project with a particular end goal (e.g. to end all disease), it might be more apt to think about medicine as a tool whose refinement through more and more research allows it to be more effective at helping a population cope with bodily distress. Relatedly, we can all agree that a toothbrush is a tool whose function is to clean teeth. No matter how effective we make a given toothbrush – we can arrange the bristles in a variety of strategic configurations, include an ergonomic handle, or even power it with a small motor – that toothbrush will not end dirty teeth. However, just as the persistence of dirty teeth in the world ought not be thought of as a failure of the toothbrush, the persistence of disease in the world ought not be thought of as a failure of medicine.


Nick Ichien studied Psychology at NYU and Philosophy of Social Science at LSE. He wrote this article and is concerned this biography is too sparse.

Recommended readings:

SEP Entry on ‘Concepts of Disease and Health

Chan-Zuckerberg Initiative website

Christopher Boorse’s Biostatistical Theory of Health

FacebookTwitterReddit

Leave a Reply