What is a disease? 13 Feb 200818 Sep 2017 Biology does normativity all the time. There are things that are the “normal” type of state of a species, an organism, an ecosystem, and so on, and things that are abnormal. But the puzzling thing is that all philosophers know, since David Hume, that normativity doesn’t develop out of facts. So no amount of factual statements about species, organisms and ecosystems will give a definition of what is normal. A suitably abstract introduction to the Seed Masters’ Imperative: “Tell us what a disease is”, right? Hey, I’m a philosopher. What did you expect? One of the normative words of medicine, as well as biology, is “disease”. It’s the opposite of “ease”, after all. It’s something that interferes with the ordinary flourishing of an organism. But that doesn’t help us much. We seem to be left with a matter of taste. And this may be what evolution implies too. For example, Type II or late onset diabetes is the outcome of a hunger for sugars that in our ancestry was in fact a good thing. But now that beets and canes are routinely harvested for sugars, we stress the pancreas beyond its ability to adapt, and it breaks down, causing insulin producing cells to die off. Hence, diabetes. At one point, our lust for sugars would drive us to find the odd bit of honey or sugar ant. Now it leaves us obese and suffering a “disease”. However, I think we can give a relatively straightforward definition of disease in evolutionary terms, which overcomes the normativity problem. But it has a wrinkle or two… A disease is any state or effect of organic processes that lowers the relative fitness of organisms. So, let’s unpack this a bit. A state includes the diabetes, which is not caused by the action of any other organism than the one doing the sugar feasting – us. An organic process includes the processes of other organisms, though, the ones we usually call pathogens or parasites. Vulnerability to infection or parasitism leaves the host organism in a less fit state. Nutrients are stolen, the host’s body becomes food for the infective agent, systems start to break down from the “normal” state. In the case of viral infection, the parasites aren’t organisms, but the effect is the same. Host resources and systems are “subverted” (another normative term – watch out for them, they are inevitable) by the virus. And this is a Bad Thing, right? Well, depends whose perspective you are going to take. From the host’s perspective, yes, it’s a Bad Thing. It lowers the host’s fitness relative to those of its population that are uninfected or are resistant to infection. From the species’ perspective, however, it may be a slight hiccup in the evolutionary trajectory. Sure, we got infected by various pox viruses and the plague. Now we have genes that confer resistance, such as CCR5 gene. AIDS is a disease of humans, but simians, felines and other organisms with related viruses just carry them without much impact. In a few hundred generations, we will too. But, and here’s the wrinkle, from the pathogen’s perspective, it’s of course a Good Thing. The fitness of the pathogen is the Good “sought” here, and the epidemiology of a “disease” is the population genetics of the pathogen. In fact they are mathematically intertranslateable. Pathogens, parasites and predators all share a feature: from the host’s or prey’s perspective they are fitness lowering, but from their perspective they are just transacting the ordinary business of evolution. So should we call predators disease causes? Well it would be the extreme, as predation generally kills its “hosts” outright, while “diseases” generally lower fitness but not eliminate it altogether. But predation doesn’t extinguish fitness along with the organism either – the organism may have had some progeny before it becomes lunch, and it may be protecting its close kin by dying. So there is no clear line between disease and predation. Once you conceive of infectious disease as a matter of fitness levels of interactions between distinct species, you start to see things differently, as Paul Ewald in a classic book, Evolution of Infectious Disease, pointed out. There used to be a story that virulence, which is the measure of how effectively the pathogen exploits the host’s resources, would slowly evolve to become relatively benign: the pathogen would become a commensual. This idea, which was developed by Sir Macfarlane Burnet and Sir Peter Medawar back in the 50s and 60s, was based on the understanding that evolution occurred, but they took an unfortunate one-sided view, that of the host. Not surprising, given that we are usually the hosts referred to. Ewald pointed out that this only works if the genetic interests of the disease coincide with the genetic interests of the host. If the rate of infection is roughly equivalent to the generational time of the host organism, then commensuality will evolve. But if it is a rapid infector, then selection will drive an increase in virulence, simply because the ones that exploit the host more efficiently will out reproduce the other forms. So the take home lesson is: inhibit infection to cause a decrease in the virulence of diseases. Wash your hands and the food you prepare. Use condoms. Vaccinate. Take any and all measures to prevent the spread of disease and the ones that persist will tend to become, rather rapidly, the ones we can live with. But back to the normativity question. What counts as a disease? When is it disease? Organisms that employ other organisms typically affect the phenotype of those organisms. The phenotype is how the organism’s genes are expressed in a given situation or environment, including, of course, all the pathogens the organism is exposed to “ordinarily”. But phenotypes only have fitnesses in particular environments. To take a silly example, no human phenotypes are fit in any way once you put them in a vacuum. So the fitness value is relative to the environment, as the diabetes example shows. But it is also relative to the other phenotypes (and hence genes) in the population’s gene pool. The diabetes inducing genes may be the fittest there are in that population of hominids. What counts as a disease is highly local, evolutionarily speaking. In environments that include sources of folic acid in maternal diets, certain genes form quite “normal” vertebra, and hence spines. But in folate-poor environments, they cause spina bifida, in which the vertebral column incompletely encloses the spinal cord. Are these genes diseases? Well, not unless genes arise that form the spine “properly” in the absence of folates. Then we’d probably say that the “normal” genes of today would be diseases. And that is my point. Disease is a normative concept relative to some environment, gene pool, and the interests of the organisms (usually us). In the general dance of evolution, it’s just the dancers that miss a step and stumble… Ecology and Biodiversity Evolution General Science
Ecology and Biodiversity Measuring extinctions 24 Jan 2008 In 1972, David Raup published an influential paper on taxonomic diversity during the Phanerozoic. In that paper, he estimated extinction rates based on the number of fossil families and genera for the period and before and after. The idea was to estimate the “kill rate” of major disruptions in earth’s… Read More
Evolution Philosophy is to science, as ornithologists are to birds: 3. Science is a Dynamic Process 6 Jun 200724 Nov 2022 In this post, I want to propose my own view, or rather the views I have come to accept, about the nature of science. [Part 1; Part 2] Read More
Education British schoolkids’ teddies win space race 6 Dec 200818 Sep 2017 I just love this: Cambridge University’s Space Flight club got local school children to make space suits for these teddy bears, which were attached to a helium balloon that rose to 30km, enough to see the curvature of the earth. All teds were recovered safely. I expect Prof. Steve Steve… Read More
There are norms and there are norms. Some norms are normative or evaluative, in the sense that they prescribe or guide actions. But there are other kinds of norms that are not normative in this way. For example, there are statistical norms that are, presumably, just matters of fact. There are also theoretical norms, otherwise known as idealizations, that are useful when characterizing how systems behave in the absence of “disturbing influences” (think of null hypotheses “natural state” characterizations). It’s worth noting, too, that theoretical norms can be quite useful even if the conditions they describe are unrealizable. While neither statistical nor theoretical norms are inherently normative, they can, of course, be adopted as normative norms. I tend to the view that the norm of health is a theoretical norm which, for obvious reasons, is usually treated as a normative norm. And I view diseases (I prefer the terminology of ‘pathological conditions’) as deviations from that norm. This is the only approach I’ve found that fits well with the idea that medicine is a discipline with a distinctive goal, the promotion of health at the level of individual organisms (called patients), and a distinctive domain of problems consisting in all the ways organisms can deviate from the state of health. It also means that whether or not some condition consitutes a disease is a matter of fact — this is rather important if medicine is to be a scientific discipline. So, I’d say that perfect health is an idealized condition in which the natural functional capacities of an organism’s parts and processes are maximized. There’s unlikely to be a unique such state, so there are probably multiple ways in which health could be instantiated for a given organism. Also, since we’re talking about a theoretical idealization, it’s probably the case that no organism is ever perfectly healthy. With that much too sketchy account of what health is, diseases are deviations from maximal functional capacity. There’s a lot more to the philosophy of medicine than what passes for bioethics.
If fitness here means ability to reproduce then the proposition that disease is what makes an organism less fit is clearly false. In humans disease is age related and most diseases that kill us kill us after our reproductive years are over. Maybe you mean fitness in another sense, but what it is you haven’t made clear. Work on centenarians shows that females can live for many decades in a chronically ill, debilitated state. Males of the same age are healthier and more robust. But when men get sick, they die quickly, so women outnumber men in the “old old.” The women are more fit in the sense that they live longer but they live longer with lots more of what most everyone calls disease. So I don’t think you have solved the puzzle of what “disease” is. That there is a normative and cultural aspect is undeniable. We have things that were once considered normal variants that are now considered disease and vice versa (example of the former: eccentric behavior now considered part of the autism spectrum; examples of the latter; homosexuality). But that’s an observation, not an analysis. You haven’t convinced me.
Well revere, I understand your point, but I think perhaps I haven’t made mine clear. First of all, I am using the term “fitness” here in its evolutionary sense. A woman may be fitter than another who is active and be bedridden, so long as she has more viable progeny after some number of generations than the other. Moreover, I am not talking about the fitness of a singular organism. I am talking about the lowering of fitness of organisms plural. By definition, that implies, in fact, of the alleles of organisms, or the types of organisms. That women outlive men is, in some sense, either a demonstration of my point or irrelevant to it, depending on which escape hatch I want. If women outlive men because doing so contributes to their Hamiltonian inclusive fitness (which here means their close relatives are fitter because they are alive) then disease that lowers that affects fitness even if they are well past reproductive age. If they outlive men because of some fitness unrelated factor (say, higher testosterone levels cause more mitochondrial damage, to invent an example) then it is not disease that kills men earlier, but merely age. You, as a medical person, think of disease in terms of the effects on individuals. I am suggesting this is a bad way to conceive of disease (but not of medical conditions, which is another thing), because it gathers together many disparate facts and processes. If there’s one thing (or even just one thing) that philosophy can offer to the sciences, it’s attempts to make language unambiguous. Here I am suggesting there is something that tracks natural facts, based on evolution. That we socially include other matters, like individual comfort, doesn’t make the category a natural one.
Hmm. Still not convinced by the definition that lowering fitness = disease. For example, vasectomy fairly clearly lowers a man’s future reproductive fitness! It would seem odd at the least to classify it as a disease process.
That doesn’t follow. If the resources one can save by not fathering further children can be devoted to other children (either your own or those of near kin) then a vasectomy can raise fitness. Also, cultural fitness and biological fitness are not the same thing. People can do things that raise their cultural fitness but lower their biological fitness. In that case I would say that was a disease – like self-harming behaviour.
John – You’re not going to persuade physicians that disease is a population level phenomenon. Their experience every day speaks to the contrary. Even using population measures to define the “normal range” is fraught with problems, since it doesn’t allow for stable attribution of disease at the individual level; i.e., a patient whose condition is unchanged might be getting “better” or “worse” relative to a shifting population norm. To a clinician, saying that person’s health status has changed in the absence of any “internal” change is just bonkers.
“He was constantly reminded of how startlingly different a place the world was when viewed from a point only three feet to the left.” (the late Douglas Adams) Most models are wrong, but some are useful. (paraphrase of George Box) Thanks, John, for describing your perspective of this phenomenon.
“He was constantly reminded of how startlingly different a place the world was when viewed from a point only three feet to the left.” (the late Douglas Adams) Most models are wrong, but some are useful. (paraphrase of George Box) Thanks, John, for describing your perspective of this phenomenon.
What *is* the difference between a disease and a medical condition? Is it the length of time one suffers from the ailment? Does it matter that the problem is caused by a pathogen? Can something be a disease and a condition?
What *is* the difference between a disease and a medical condition? Is it the length of time one suffers from the ailment? Does it matter that the problem is caused by a pathogen? Can something be a disease and a condition?
In my opinion, medical conditions are defined by their relationship to medicine, which is at least partially a construction of the social attitudes and technologies of the day. Most of what I call medical conditions are what I call diseases, but not all of them. A great many MCs are defined by their being normatively deprecated or disliked, such as the effects of old age, but that doesn’t make them diseases. Pathogens and other fitness lowering conditions are medical conditions if they are recognised by the health standards of the society as medical conditions. We would hope that most diseases would be so recognised, but at least a few are not diseases but disapprovals.
John – it’s a thought experiment, so I can explicitly say that the hypothetical vasectomy does indeed reduce biological fitness. I still don’t think it’s a disease process. Or how about the simple choice of deciding to remain childless? For information, the hypothetical person making this choice has no brothers or sisters, and even if he did, he wouldn’t want to help raise them. Is the act of making that choice a disease? Or how about someone who marries a divorcée and acquires a step-family? They are now putting their time and effort into raising someone else’s children. Was their wedding a symptom of a disease?
Peter, you are still thinking from the perspective of the individual. Fitness is something that applies solely to populations, and in particular the relative success of heritable types in that population. That individual may not increase his fitness thereby, but if that behaviour is heritable, and it persists in the environment, it is not detrimental, and hence by my metric not a disease. Moreover, we allow that some behaviours might not be biologically determined. Having a vasectomy is likely at the end of mostly socially determined (in this case economically) behavior. It’s still a medical matter, but not a disease. On the other hand, in certain societies, it is regarded as a disease…
It seems to me that we need to distinguish between diseases and what are _regarded_ as diseases. And, if we think medical conditions are defined by what the medical community chooses or decides to include in their nosologies and diagnostic categories, then we need to ask for the basis of those choices/decisions, and ask whether the medical community has chosen or decided rightly. Are there non-diseases (non-pathological conditions) that are rightly deemed medical problems? If so, on what basis? And is that basis compatible with judgments about medical indications and medical necessity?
“A disease is any state or effect of organic processes that lowers the relative fitness of organisms.” Just to press the point that John’s populational account of disease is problematic, consider so-called “universal diseases,” like dental caries. If all the members of a population have a condition like this, it won’t lowever the _relative_ fitness of organisms. So on John’s account, it couldn’t be a disease. Something tells me that we’ve run off the rails.
“A disease is any state or effect of organic processes that lowers the relative fitness of organisms.” Just to press the point that John’s populational account of disease is problematic, consider so-called “universal diseases,” like dental caries. If all the members of a population have a condition like this, it won’t lowever the _relative_ fitness of organisms. So on John’s account, it couldn’t be a disease. Something tells me that we’ve run off the rails.
Just after speed-reading your pretty interesting entry, I want to tell you about a certain mistake you make in your argumentation: >>But if it is a rapid infector, then selection will drive an increase in virulence, simply because the ones that exploit the host more efficiently will out reproduce the other forms.
Bob k: Dental caries are not universal in populations, the severe forms are caused by bacteria that some people just never catch. That tendency seems to be at least partly genetic, or at least somewhat heritable in the social context (you get the bacteria from your primary caregiver, if you’re going to get them at all), so it still fits John’s definition. If you want to argue that a whole poplution might still have them, you’re into population-level selection, which John probably knows more about than both of us. It seems to me that a lot of the “medical” people commenting on this thread have at best an undergrad-level understanding of evolution… I hope that doesn’t sound rude.
Bob k: Dental caries are not universal in populations, the severe forms are caused by bacteria that some people just never catch. That tendency seems to be at least partly genetic, or at least somewhat heritable in the social context (you get the bacteria from your primary caregiver, if you’re going to get them at all), so it still fits John’s definition. If you want to argue that a whole poplution might still have them, you’re into population-level selection, which John probably knows more about than both of us. It seems to me that a lot of the “medical” people commenting on this thread have at best an undergrad-level understanding of evolution… I hope that doesn’t sound rude.
Chris – I’m aware that caries is not actually universal, since I’ve got a brother who’s gone 60+ years without. But the point is conceptual, which is perfectly legitimate when people are proposing definitions. So, I’ll stand by my point that a universal disease is a conceptual impossibility if disease implies reduced relative fitness. I’m a bit beyond the undergrad level, though I imagine there are still things I can learn about evolution. I’ve even studied Ewald’s work, and appreciate its significance for understanding some features of some diseases. But I don’t think, as apparently John does, that it provides a different way to understand what a disease is. I’d encourage anybody who’s interested in concepts of health and disease to at least familiarize themselves with the extensive literature on this topic.
Bob: fair enough, but even as a thought experiment I don’t like it. An evolutionary biologist would be very surprised to find every member of a large population being equally affected by the same disease. Wherever you have universal diseases, you have resistance. Malaria and sickle-cell anaemia are the famous, contradictory example of course. But there’s also genetic variation in HIV resistance, courtesy of ancient retroviral incursions in some populations, and even genetic variation in our rates of aging (what a friend of mine calls the pixie gene). I think John’s definition is, at the very least, defensible. Large-scale evolutionary explanations tend to be a bit confronting when you’re not used to them. It’s the cavalier way they treat individuals that get most people defensive.