Hana Field: Iris, hey — I have to tell you, I had the strangest moment at the pharmacy yesterday, standing there holding a bottle of fish oil and genuinely not knowing if the dose on the label was, like, a suggestion or a ceiling.
Iris Holm: And you couldn't tell from the bottle.
Hana Field: Nothing. Just "take two softgels daily" and a picture of a salmon. And that's — okay, so that's actually exactly where we're going today, because there is a principle that should answer that question, and it's so old and so fundamental and somehow we still manage to ignore it completely.
Iris Holm: Dose makes the poison.
Hana Field: Paracelsus. 1493 to 1541. A Swiss physician who said, essentially, every substance will kill you — the only variable is how much. And we have spent five centuries building the math to prove him right, and people are still buying supplements that could genuinely harm their liver because the label says "natural."
Iris Holm: Right, but here's what makes it concrete — aspirin. Five hundred milligrams, you get pain relief. You push past roughly 150 milligrams per kilogram of body weight, you're in toxicity. Past 500 milligrams per kilogram, you're looking at potential lethality. Same molecule. Nothing changed but the number.
Hana Field: Same molecule. And the one that really stopped me — water. Hyponatremia. Endurance athletes have actually died from drinking too much water, documented cases, because the sodium in their blood diluted past the point of function. Water.
Iris Holm: Which means the question isn't "is this thing safe" — it's what dose is safe for which body. And that's the thing we don't have a clean answer to.
Hana Field: And that's the thing that I keep tripping over — like, we know it's not on/off, we know it scales, but I don't think most people actually have a picture of what that scaling looks like.
Iris Holm: Dimmer switch. That's the whole thing. Not a light switch — a dimmer. Below a certain point, the room stays dark. No effect at all. Then there's a range in the middle where small turns produce real, useful light. Crank it too high, the bulb burns out.
Hana Field: Oh, that's — yeah, that lands.
Iris Holm: The actual shape is called sigmoidal — S-shaped on a log-dose axis. Three phases. Flat at the bottom, that's sub-threshold, your body literally doesn't register the substance. Then a steep climb, the therapeutic zone, where small dose changes produce massive effect changes. Then it plateaus — and the plateau tips into harm.
Hana Field: So the middle section, the steep part — that's the therapeutic window. And I've heard that phrase, but what makes one window narrow and another wide? Like, what's actually different?
Iris Holm: That's the therapeutic index — it's a ratio. Toxic dose divided by therapeutic dose. And if that number is, say, two, you're in trouble. You double the effective dose and you're already in danger. A wide window means the ratio is large — you've got room. A narrow one means the gap is basically nothing.
Hana Field: And aspirin actually sits in that narrow zone for certain people.
Iris Holm: Right — which is exactly why the 20th century had to put numbers on this. EC50, the dose that gets you fifty percent of the maximum effect, that's what made Paracelsus's intuition into something you can actually compare across drugs. But the complication is that EC50 isn't a fixed constant. It shifts with age, liver function, body weight. The number that clears the steep part of the curve for one person might already be on the plateau for someone else.
Hana Field: So the curve is universal as a shape — but where any individual sits on it, that's a completely different question.
Iris Holm: And that's exactly where 'natural' breaks down — because the curve doesn't check the label. Turmeric, fish oil, vitamin A — all of them have toxic thresholds. The curve doesn't care what aisle you bought it in.
Hana Field: Which is — I mean, that's the thing that actually gets me. Paracelsus figured this out in the 1500s, and we have five centuries of math behind it now, and someone is still standing in the pharmacy reading 'natural' on a fish oil bottle and thinking, okay, I'm exempt. Like the word 'natural' is doing this enormous load-bearing work that physics simply doesn't allow it to do.
Iris Holm: It's emotional categorization. 'Natural' gets filed in a different mental folder than 'drug.' Same curve, different folder.
Hana Field: And the folder is wrong.
Iris Holm: Here's the part that sharpens it — potency versus efficacy. Two supplements, say turmeric and fish oil, both marketed for inflammation, both claiming roughly the same benefit. Same efficacy on the label, functionally. But their EC50 values could be completely different — meaning one might require a far higher dose to get you to that same effect. And the bottle doesn't tell you that.
Hana Field: Wait — so you could be taking more than you think just to hit the marketed effect, and that excess is exactly what pushes you up the curve toward harm.
Iris Holm: Toward the plateau, yes. And the plateau isn't neutral — it tips into toxicity. The FDA doesn't set reference thresholds for salicylate for fun. Blood concentrations above 300 micrograms per milliliter — that's their clearly toxic marker for aspirin. That number exists because the curve demanded a number.
Hana Field: And you know what none of that population data accounts for — and this is actually where it gets worse before it gets better — is where any one person actually sits on the curve. Because population ranges are approximations, and we haven't even touched what happens when individual variation shifts your threshold entirely.
Iris Holm: That's the next layer. And it complicates everything we just said.
Hana Field: And the shift can happen without anyone noticing — that's the part that actually keeps me up. I keep thinking about a nephrology clinic, a patient on standard aspirin for heart protection, and they finally check her creatinine after two years. The number comes back wrong. Wrong toward danger. And nobody had told her her curve had moved.
Iris Holm: Because the 150 milligrams per kilogram threshold — that's a population number. It doesn't know her kidney function declined.
Hana Field: She's taking the instructions she has.
Iris Holm: Which is why therapeutic drug monitoring exists — TDM. You actually measure blood concentrations. You don't estimate from population tables, you pull the number from this specific patient and you ask: are we inside the window, or have we drifted past it? The FDA's 300 micrograms per milliliter marker for salicylate toxicity — TDM is how you find out if she's already there.
Hana Field: So TDM is basically the admission that population ranges aren't enough.
Iris Holm: It is. And it's still not universally accessible — which, frankly, is a separate problem. But even TDM assumes the curve is a single sigmoid. And that assumption, actually, is where hormesis walks in and breaks things.
Hana Field: Wait — a biphasic curve? Like beneficial at very low doses, then harmful?
Iris Holm: For some substances, yes. A stimulatory effect at low doses before inhibition kicks in. That's not the simple threshold model — it's a different shape entirely. And there's actually an open-source tool called CurveCurator that applies a recalibrated F-statistic just to classify which type of curve a substance is following, because the computational problem of fitting these curves reliably is still unsolved. That's not historical. That's current research.
Hana Field: So we're handing patients population averages, and the underlying math for individual curves is still — I mean, that's the thing nobody in the pharmacy aisle knows.
Iris Holm: And nobody has mapped the collision. That's the part that actually breaks the Paracelsian promise — not because Paracelsus was wrong, but because he was describing one substance. The moment you're taking three things simultaneously, the graded response you'd predict from any single curve becomes something else. A population-level risk you didn't sign up for.
Hana Field: And that's — I mean, that's most people, right? Not one drug. Three things, a supplement, a prescription, maybe an over-the-counter. And the curve that was mapped for each one individually just doesn't account for what they do to each other. Five hundred years of 'the dose makes the poison,' and the clean single-substance rule quietly falls apart the minute someone's Tuesday morning looks like most people's Tuesday mornings.
Iris Holm: The honest version is this — the curve is universal. Your position on it is yours alone. And we've barely started building systems that treat it that way.
Hana Field: That's where I keep landing. Paracelsus gave us the question in the 1500s, the 20th century gave us EC50 and TDM, and most people are still navigating dose by reading a label written for a population they may not belong to. That's not a failure of science. It's a failure of translation.
Iris Holm: Good place to stop. Genuinely.
