"It worked for my aunt": why stories aren't evidence
Your aunt's experience can be real. It still cannot carry a rating.
Checked against the sources ยท Updated 17 Nov 2026
In short
- A family story is enough evidence to accept a health claim
- The Cochrane Handbook says very low certainty is the right level for unsystematic clinical observations, such as case reports. One person's story is that kind of evidence.
"It worked for my aunt": why stories aren't evidence
Picture Sunday lunch. The family is a composite, not a real one, and you almost certainly know one like it. You are at your Tita's house, and someone is passing a phone around the table with a shopping link already open. A cousin's friend tried a powder for a month, and her knees feel better. There are photos. There is a discount code. Because you are the one who reads, the eyes drift toward you.
So here is the honest answer, the one you can say out loud without hurting anyone. Her experience can be completely real, and it still cannot carry a health claim. One person's result sits at the bottom of the evidence researchers use. That is not an insult. It is how evidence gets weighed. Once you can see why the weighing works this way, the table is easier to answer.
What one month actually contained
Her knees felt better. Pain also changes on its own. Sleep changed, or the weather did, or one kinder week changed how her whole body felt. She may have started other things, or stopped them, and nobody wrote it down. The month was a whole life, and the powder was one new thing inside it. Any of those changes could sit next to the easier knees, and the chat has no way to sort them.
There was no second person living the same month with everything held the same except the powder. And the neighbors who tried the same jar, felt nothing, and stayed quiet never made it into the chat. Quiet disappointments rarely get a turn with the phone. The table hears the month that went well, because that is the month someone wanted to share.
One person, one outcome, no comparison: researchers call that a case report. Doctors still write them, because a single odd result can be the first hint of something worth studying. A hint can start the question. It cannot be the answer. You can be glad her knees eased and still not know what eased them.
How reviewers weigh it
Cochrane, the network that reviews health evidence, puts that kind of report, and a handful of them together, at the bottom. They are what you get when someone notices what happened and does not set up a fair comparison. Reviewers rate the whole set of evidence for a question, not one person as a pass or fail. They are weighing the claim, not the person who brought it to lunch.
Careful studies that watch groups of people over time already sit higher than a single story, and they are still not the strongest evidence. A story about one person begins below those. From her month alone, a reviewer cannot say what the powder would do in someone else's month.
Sitting at the bottom does not mean she is lying. It means that from her story alone, nobody can know what the powder would do for her cousin, or for you. The rule is about the claim traveling from one person's month to everyone else at the table.
Why the story feels heavier
You know the face. She is standing right there, moving better than she was, and no table of numbers has ever brought you soup when you were sick. That pull is real, and it is social. A familiar person will always feel more solid than a paper she has not met.
A familiar face does not add people to the evidence. A hundred stories retold in the chat are still a hundred single stories. Retelling does not turn them into a comparison, and it does not create the quiet neighbors who never posted. Stories can tell researchers where to look next. They cannot tell the table what worked, or what will work for the next person who orders the same jar.
Honest limits
She may not have imagined the relief. Bodies do change. Plenty of worthwhile research began as one patient's strange result that a curious doctor refused to ignore. Keeping the door open for that kind of hint is not the same as treating the hint as proof. The claim being checked is the shortcut, not the storyteller: that a family story is enough evidence to accept a health claim. That shortcut is what gets a red flag.
Any rating names the claim, the studies, and the date. It does not name your aunt.
What to say at the table
Separate the person from the claim, out loud. You can hold the line without making anyone feel small.
- Thank her, then ask. You are glad the knees eased. Then: what else changed that month, and has anyone tested this in other people with knee pain? The first sentence is about her. The second is about the jar.
- Offer to look together. One person cannot tell you it was the powder. If there is a study in people, read it with her. That turns you from the family skeptic into the person who will sit with the paper.
When a story needs a doctor
A story stops being harmless when it is used to stop a prescribed medicine, or to treat chest pain, fainting, or a sudden new symptom. Those are not questions for the table. They need a doctor who can examine the person. No story, however sincere, can do that exam.
---
Sources
- Cochrane Handbook, chapter 14. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-14
- Balshem H, et al. GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology. 2011. https://doi.org/10.1016/j.jclinepi.2010.07.015
General guidance, not medical advice or a diagnosis. Talk with your doctor about your own health.
How strong is the evidence?
What the research says: The Cochrane handbook says very low certainty fits unsystematic clinical observations such as case series and case reports. A GRADE paper sets four levels of quality, high, moderate, low, and very low, and starts even observational studies at low.
What it doesn't say: This does not mean your aunt imagined it, or that a story never points to something worth studying. It means her result is not a reason for the whole chat to copy the bottle, the fast, or the gadget.
Cochrane Handbook, chapter 14
How we rate: the Flamel MethodTry this
When a relative shares what helped them, thank them, then ask what else changed that month and whether any study in other people found the same thing.
That counts.
See a doctor if The story is being used to stop a prescribed medicine, or to treat chest pain, fainting, or a sudden new symptom.
Get delightful living tips directly in your email
Join freeSources
- Cochrane Handbook, chapter 14
- Balshem H, et al. GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology. 2011.
Last reviewed by Flamel Staff on 17 Nov 2026
Flamel is not medical advice and does not diagnose. If you have symptoms or readings in the red zone, see a doctor.