Feb. 15th, 2009

haggholm: (Default)

For those keeping track, I have two lists of allergies—the first is my list of clinically confirmed allergies; the second is my list of foods that I will not eat because I am clinically confirmed to be allergic to them or because they are related to such foods and I wish to avoid cross-reactivity or because personal experience leads me to strongly suspect that I’m allergic to them. The latter I wish to have clinically tested. I get the impression that simple scratch tests don’t have impressive sensitivity (they are prone to false negatives: The scratch test fails to reveal an allergy that is real) but very high specificity (they are not prone to false positives: If the scratch tests reveals an allergy, it’s very unlikely to be spurious).

  • Some legumes: Peanuts, soya beans, peas, chick peas, lupini beans, fava beans, kidney beans.
  • Some tree nuts: Almonds, hazelnuts.
  • Chicken: This test was more suggestive than conclusive, but it has been tested and it definitely wasn’t a clear negative.
  • Legumes: Because I react to most legumes, I avoid the family as a whole.
  • Tree nuts: I actually tested negative on reactions to brazil, cashew, and pecan nuts, but the risk of cross-reactivity is high and the allergy can be fairly severe; if I ate nuts, I’d risk a high probability of strong reaction in the future.
  • Chicken and turkey: Both seem anecdotally to cause reactions, and the chicken test makes me extra suspicious. I don’t know about other poultry, like ducks or geese.
  • Some fresh fruits: Apples, plums, papaya, and strawberries at the moment; this list would likely grow quickly if I ate more fruit.
  • Button, criminy, and portobello mushrooms.

If you’ve really been keeping track, you’ll note that mushrooms is a new entry. It hasn’t been clinically tested at all, because a couple of years ago, when I had my tests, it wasn’t on the board. In fact, I’m considering today’s reaction the clincher. To illustrate the difficulty of figuring this stuff out, and hopefully convincing you that I am applying critical thinking to my own health—also, because I feel like writing about it—I’ll go over it, though very briefly.

The problem with food allergies is, of course, that short of testing everything you might eat, you can’t know when they’ll appear; and even that doesn’t account for the all-too-common development of new allergies (when I was a child, my list consisted simply of nuts, and peanuts). I have no idea about any new allergy until I eat something that contains the allergen and I feel ill. Today was a particularly annoying case of it—I had run out of bread at home and so gone out for breakfast; my breakfast consisted of a roasted vegetable sandwich with portobello mushroom. I subsequently went to the gym, did a bit of rolling (very badly, but I don’t blame allergies for that), then started feeling very ill and dizzy. After I sat down and my head cleared a bit, the nausea settled into a form that I have learned to associate with mild allergic reactions. But to what?

The problem with figuring this stuff out is that I rarely come across my allergens in isolation. The sandwich didn’t just contain mushrooms, after all; it also had onion, bell pepper, artichoke, and various other things (I don’t even recall all of them—I just noted that none of them were known allergens to me). This is obviously a perfect setup for conflating correlation and causation—I ate the sandwich; the sandwich had mushrooms in it; I got sick—ergo, the mushrooms made me sick. I know of no better way to identify allergens (save allergy tests requiring a referral and a long wait) than to simply note the occasions when I feel thus ill and try to identify common threads.

It gets extra tricky precisely because I have so many allergies. Did I feel sick today because of the mushroom, or something else in the sandwich? Did it contain some sort of legume, or perhaps hummus, that wasn’t mentioned? Was there some contaminant? Or, for that matter, did I feel ill for some entirely separate reason? This clouds the picture at every occasion of a presumed reaction, making it pretty hard to be sure. It doesn’t help to be aware that humans are notoriously prone to a false positive bias (presumably because, in an evolutionary context, seeing imagined patterns is only a little bit costly in terms of energy, whereas failing to see real patterns is often lethal: A limited heuristic budget predisposes us to finding patterns whether they exist or not).

The strongest datum supporting my conclusion occurred the other week, when I made a warm pasta salad I’ve taken to very frequently making for work lunches. I usually make it with whole wheat penne, sautéed red onion, steamed kale (and sometimes spinach), pieces of sundried tomato, and pieces of capocollo ham. (It’s quite tasty.) The other week, however, I was out of capocollo but had some mushrooms—so I sautéed them as a substitute. After eating that, I felt mildly ill, even though I’ve always felt fine when eating the same dish made with capocollo and no mushrooms. As far as it’s possible to isolate variables, I’d say that did it. But even so, that alone was hardly ironclad—all the confounding factors still applied; and at the time, how was I to hold mushrooms as a culprit over, say, kale? I’d never reacted poorly to kale, of course, but I didn’t think I was allergic to mushrooms either! —Well, a number of events like today’s, coupled with that much stronger datum, has lead me to this reluctant conclusion: I’m fairly certain (say, 85%­–90%) that I am, in fact, allergic to these mushrooms; and I shall no longer eat them, if I can avoid it.

I will also try to get a referral to an allergy specialist again. I want this stuff confirmed.

On a side note to this, I looked up the mushroom in question and found that button, criminy, and portobello mushrooms are in fact the same species (Agaricus bisporus, or the table mushroom), harvested at different stages of maturity. This gives me some hope that even assuming that I am right and I am allergic to them (and my evidence is gathered from all three types, so their identity strengthens my conclusion) I may still be able to eat other mushrooms; especially as the fungi are a very, very diverse groups. (Many other edible mushrooms are not in the same genus or family; oyster mushrooms belong to the same order, Agaricales, whereas chanterelles are only related to the above at the level of the class Agaricomycetes.)

haggholm: (Default)

Reposted from a forum.

The Malthusian argument:

Unchecked population growth is geometric in nature (2x). Because resources are always finite, no population can grow unchecked for long; therefore, most individual organisms die.

Natural selection:

Deaths are either "random" or statistically controllable by their victims. An example of a "random" death is death by direct meteor impact. An example of a "controllable" death is predation—a fast rabbit may get killed by a fox, but out of a normal distribution of rabbits according to running speed, the lower end will be disproportionately prone to predation. As Louis pointed out, this leads to the mean running speed shifting up—evolution due to natural selection.

(I leave the logical third alternative out of my discussion: Death is either random with respect to fitness; or it favours the fit…but logically, we might have a situation of death specifically favouring the status quo by preferentially killing off those who would otherwise prosper due to superior fitness. This is a silly idea, however, nonsensical and unsubstantiated—I mention it only because it's logically a missing piece from the spectrum of possible explanations.)

Of course, this only applies under selection pressure—but with geometric population increase in the face of finite resources, almost every individual organism on Earth is under selection pressure (the exception being a newly introduced species that does grow geometrically in a new environment). The question is not whether there is selection pressure, but of what sort, and how strong it is.

If all deaths are random, then there's no reason to believe that any trait will change consistently—the fit are as likely to die as the unfit. If all deaths are probabilistically subject to phenotypic traits like running speed, photosensitivity, digestive ability due to gastric acidity, skin toughness, etc., then there will be a very strong selective pressure in favour of any favourable traits.

The truth is, of course, somewhere in between, but that doesn't mean that traits will not preferentially evolve—it just means that it's slower. If only half of all deaths are "controllable", then the pressure will be half as strong; if only a quarter, then a quarter… But since starvation, predation, and mate competition are in fact very common causes of failure to leave numerous offspring in nature, we may reliably infer that change does happen.

Universal Darwinism:

This is based on what Dawkins calls "universal Darwinism"—he postulates that if, anywhere in the universe, there exist a population of entities that copy themselves with high but less-than-perfect fidelity, and if they are competing for finite resources, then evolution by natural selection will occur. I find the argument very compelling. (I may not have done it justice in my very brief summary. Here is a longer account.)

Note that this argument does not require empirical observation—it's simple logic. Of course it needs to be supported by empirical observation: It's based on our understanding of genetics. Of course it only gives a very, very crude outline of the process: Evolutionary biology is a field with massive amounts of research and evidence and minor theories failed or successful; and the process is vastly complicated by genetic linkage, epistasis, and especially pleiotropy, which can make determining the fitness of an allele difficult. The basic idea, however, requires nothing beyond the Malthusian argument, Mendelian genetics, awareness of mutation, and a little bit of logic.

It is also poignant that out of all the theories ever proposed for the origin of life (let alone the genetic evidence of common descent!), Darwin's theory of evolution by natural selection, and its descendant in the form of modern evolutionary biology, is the only theory that has ever been at all successful—and that, though it may be the most controversial scientific theory since heliocentricism, it has survived intense criticism for a century and a half.

What alternative mechanism do you propose?


After I posted this, I received an positive private message with an idiom that, for some reason, amuses me to no end:

That was an absolutely cracking post Petter!

Me, I'm more of a nuts and bolts guy, but that was a compelling, and clear, big-picture view.

Louis

I made a cracking post. That is strangely satisfying.

Profile

haggholm: (Default)
Petter Häggholm

July 2025

S M T W T F S
  12 345
6789101112
13141516171819
20212223242526
2728293031  

Most Popular Tags