I was standing in a chemist on a Saturday morning, holding a bottle somebody had recommended to me, reading the back panel. I could feel an old reflex firing before I had consciously decided anything.
The front of the bottle said one thing in large, friendly type. The back said something much more careful in six-point grey. Between those two panels sat a number, and that number was doing an enormous amount of work.
Thirty years of reading vendor proposals will do that to you. So will a physics education that marked me down, over and over, for writing an answer without its units. Neither of those qualifies me to tell you what to put in your body. Both of them make me very difficult to sell to.
The reflex is not scepticism. People assume it is, and scepticism is the cheap version. It lets you dismiss everything without doing any work, and it feels like rigour while costing you nothing.
What I actually have is three questions, asked in the same order every time. The first is: what is this number measuring?
"Three hundred per cent more absorbable" is not a fact. It is a fact-shaped object.
Absorbable measured how? Into the blood, or into the tissue where the thing is supposed to act? Over four hours, or over a day? In how many people, in what condition, doing what else at the time?
I did not learn to ask that in a marketing course. I learned it from a physics education whose actual content I have almost entirely forgotten. A number without a unit and a stated measurement procedure is decoration. It has the shape of evidence without the substance, and the human brain accepts the shape very happily, because checking the substance is tiring and the queue behind you is getting longer.
Blood level is not tissue level. Tissue level is not effect. Effect in a dish is not effect in a man in his fifties who has been sleeping badly for a decade. Each of those is a leap, and a label makes all of them silently, in the white space between the number and the promise.
The second question is: compared to what?
This is the one almost everybody skips, and it is where the entire claim actually lives. "Twice as effective" is meaningless until you know the other half of the sentence.
Twice as effective as nothing? As a smaller dose of itself? As a cheaper form of the same ingredient, chosen precisely because it performs badly?
I watched vendors do this with network hardware for three decades, and it is the same move every time. You do not lie about your product. You choose the comparison. The number is honest, the measurement was real, and the reader still walks away with a false belief, because the reader assumed the comparison was a sensible one and nobody ever said it was.
A bottle has almost no room for that information. Even where it has room, it has no obligation to use it. The comparison group is the most important fact about any claim, and it is the fact most reliably missing from the packaging.
The third question is the uncomfortable one. What has been left out?
Who paid is a reasonable place to start. A Cochrane review by Lundh and colleagues, updated in 2017, pooled the studies it could and found that industry-sponsored drug and device research was about twenty-seven per cent more likely to report favourable efficacy results than research funded from other sources, and about thirty-four per cent more likely to arrive at favourable conclusions. The authors described it as an industry bias that the standard checklist reviewers use to grade a study's reliability does not explain.
That finding needs a caveat, and it is a large one. It is about drugs and medical devices, not about the supplement aisle. It does not transfer directly and I am not going to pretend it does.
What it does tell you is a smaller point, and a worse one. That effect turns up in the most heavily scrutinised, most expensively conducted, most tightly regulated research human beings do. The shelf you are standing in front of is not held to a higher standard than that one.
Funding is only the visible part, anyway. How many people were in it. How long it ran. What else changed at the same time.
Whether the version of the study that did not work was ever written up, or quietly stayed in a drawer. None of that fits on a bottle, and none of it is required to.
Which brings me to the thing I did not know until I went and looked, and which has been printed in plain sight the whole time.
Look at the small print on most of what an Australian chemist sells as a supplement and you will find an AUST L number. The Therapeutic Goods Administration is entirely open about what that means: listed medicines are not individually evaluated by the TGA for quality, safety and efficacy before they are supplied.
The ingredients have been pre-assessed as low risk and must come from an approved list. The individual product is not assessed. The sponsor, meaning the company that owns the product, certifies that it works as claimed, holds the supporting evidence itself, and the regulator can ask to see it afterwards.
I am not raising that as a scandal. It is a deliberate decision about where to spend limited regulatory attention, and it is defensible.
But notice what it does. It moves the job of evaluating the evidence away from an institution with pharmacologists on staff, and onto me, in an aisle, on a Saturday, holding a bottle. The number on the box is a registration, not a verdict.
For years I had quietly assumed it was closer to a verdict. Nobody misled me about that. I just never read it, in exactly the way I spent a decade never reading my own tiredness.
Here is where I am supposed to tell you what I concluded, and I am not going to, because the three questions do not do what people want them to do.
They will not tell you what to take. They will not save you money, and they will not settle an argument with a mate who swears the stuff changed his life. Asked honestly, they usually return the same answer, which is that you do not know, and the label was never built to tell you.
What they give you is smaller than certainty and more useful. They separate what you have been shown from what you have been told, and that gap is almost always wider than the front of the bottle suggests.
I still buy things. I am not standing above any of this. The man who reads vendor proposals for a living has absolutely been talked into a supplement on the strength of a number he never checked, and then felt slightly better for a fortnight, which proves nothing at all.
Go and open the cupboard. There will be something in there you bought for your health and are about halfway through, and it will have a number on the back of it: what was that number actually measuring, and compared to what?
This content is for educational and informational purposes only. It does not constitute medical advice. Always consult a qualified healthcare professional before commencing any health protocol or making changes to your treatment.