How we write
Four rules, a grading scale, and a list of things we will not print. It is here so you can disagree with it — a standard that is not published cannot be held to.
1. The answer goes first
Every guide opens with the direct answer in one block, before any context. If a reader leaves after that block they should have got what they came for. Everything after it is for the reader who wants to know why, and for the reader whose situation is not the common one.
2. The evidence gets graded
Four bands, applied to the guide's central claim:
| Grade | Means | How to read it |
|---|---|---|
| Strong | Multiple consistent human trials, or a major guideline built on them | Act on it |
| Moderate | Direction well supported, size of the effect uncertain | Reasonable to act on |
| Limited | Plausible mechanism, thin or mixed human evidence | Do not lean on it hard |
| Contested | Credible researchers currently disagree | Know that the debate exists |
A limited grade is not a warning that the guide is bad. It is us telling you how much weight the claim can carry, which is information most health writing withholds.
3. Every figure names its source
Sources sit at the bottom of each guide paired with the specific claim they support, not as a decorative link list. We prefer published clinical guidelines, USDA FoodData Central, NIH fact sheets and national health services over other health content sites. Where a reference range depends on the assay or the population it came from, the range table says so — a number printed without that context is the single biggest reason people misread their own results.
4. Every guide marks where it stops
Every guide carries a when to stop reading and see someone block, and it is a required field in our content system rather than a legal footer. A guide that cannot fill it in honestly does not get published. That is the whole reason this desk can write about health without pretending to practise it.
What we will not publish
- Statistics we cannot attribute. No invented sample sizes, cohorts, odds ratios or survey results. If we cannot name where a number came from, the number does not appear.
- A negative claim from silence. If a source does not address something, we say nothing rather than inferring that it does not exist.
- Dosing or treatment instructions. That is the boundary this desk marks rather than crosses.
- A guide without an evidence grade and a boundary. Which is why some obvious topics are missing from this site.
Money
No affiliate links, no paid placements. Exactly one guide across the site reviews commercial products; it is in tools and it discloses on the page which figures came from the companies themselves rather than from an independent party.
Corrections
We correct on the page and state what changed. We do not edit quietly. If a source here does not say what we attributed to it, that is the most useful thing you can tell us.