The dish-to-person gap

Almost every claim about a compound starts in a dish. Cells in a dish are given a concentration of something and something happens to them, and that is a real result. The question that decides whether it means anything for a person is whether a person's blood ever reaches that concentration.

Often nobody has looked. Sometimes somebody has, and the answer is that it does not come close. This page holds both, for every compound in the register.

There is nothing to fill in here and nothing to choose. This page cannot see you, and the two figures in each row are measurements somebody published, not calculations about anybody.

Why the distance matters

A dish can be given any concentration at all. A body cannot. Three things stand between the two, and none of them is a matter of trying harder:

What this page will not tell you, and why that is not coyness

The obvious next question is how much of something a person would have to consume to reach the laboratory concentration. This page does not answer it, and the reason is not that the answer is being withheld. It is that there is no answer to withhold.

Nobody has measured a way of reaching those concentrations. In the one entry here where both numbers exist, the two volunteers who had any detectable trace at all were already at the highest levels the study tested, and the study had been designed to keep escalating until something appeared. The arithmetic that looks as though it would answer the question produces a number with nothing behind it: not a measurement, not a finding, and not something anybody has shown a body can do. A number like that is more dangerous than the absence it fills, because it looks like knowledge.

So the register holds no amount of any substance, anywhere, for any compound at any grade. There is no field for one and nothing to compute one from. That is a property of the data rather than a promise in a sentence.

Every compound in the register

One of the sixty-nine has both numbers. That proportion is the finding, not a gap in this page: measuring what a person reaches is difficult, unglamorous and rarely funded, and the claims that circulate do not wait for it.

Both numbers exist 1

Both numbers exist, and the gap between them can be stated.

CompoundLowest used in the laboratoryHighest measured in a personGap
Turmeric, curcumin 11.32 micromolar 0.137 micromolar about 80-fold

What a person reaches is known, and what the laboratory used is not recorded here 2

Somebody has measured what a person reaches. Nothing on this page records what the laboratory used.

CompoundLowest used in the laboratoryHighest measured in a personGap
Vitamin C, intravenous not recorded 15000 micromolar no gap can be stated
Vitamin C, oral not recorded 100 micromolar no gap can be stated

The laboratory concentrations are recorded and nothing measures what reaches a person 1

The laboratory concentrations are recorded and nobody has measured what a person reaches.

CompoundLowest used in the laboratoryHighest measured in a personGap
Black salve 1.2 micromolar not measured no gap can be stated

Neither number has been measured 27

Neither number has been measured, so there is no gap to state and no basis for saying the compound reaches its target.

Not a comparison that applies 38

This is not a compound that reaches a concentration in the blood, so the comparison does not apply. Exercise, a way of eating, a conversation with a psychologist: these are not substances with a blood level, and asking what concentration they reach is not a hard question but a meaningless one.

What to do with this

If you have read that a compound does something to cancer cells, the question worth carrying to your team is not whether the laboratory result is real. It probably is. It is whether anyone has shown the compound reaches that concentration in a person — and on most of this page, the answer is that nobody has looked.

Each entry above links to its own page, where that question is written out with the sources behind it and a button that puts it on your question list.

The related tools are the evidence ceiling, for how far a study design can reach, and the claim grader, for working a claim the book never graded through the same procedure the book used.