Hey friend,
Last week a family member almost bought a supplement because of a headline.
You know the ones. "This ONE fruit reverses diabetes." He saw it, felt that little jolt of "wait, really?", and had his card out before he stopped himself and thought; hold on, let me actually check this.
Turns out the "study" was 14 people, over 10 days, and it wasn't even testing what the headline said it was testing.
He called me immediately and started asking lots of questions.
So if it can get to someone close to me, it can get anyone. Which is exactly why I want to hand you the same checklist I gave to him. Five questions you can run through your head before you believe (or share, or buy into) any health headline.
Bookmark this one. You'll use it a lot.
1. Is this "linked to" or does it actually "cause" it?
These two get swapped constantly, and it's the oldest trick in the book.
Say a study finds people who eat more organic food have less heart disease. Sounds like organic food is the hero, right? Except people who buy organic often also earn more, exercise more, and see a doctor regularly. Any one of those could be doing the real work.
For years, doctors believed hormone therapy protected women's hearts, based on studies that just tracked what women were already doing. Then a proper randomized trial ran the numbers and found the opposite. The earlier studies were mostly picking up on the fact that women who chose hormone therapy tended to be healthier to begin with not that the therapy itself was protecting them.
Ask yourself: did the study actually test this, or did it just notice two things happening together?
2. How many people were actually in the study?
Before you believe "New Extract Boosts Metabolism by 30%," scroll down and look for the sample size, usually written as N =.
If it's 12 people over one weekend, that's not a finding, that's a coincidence waiting to be reported. Bodies are wildly different from person to person. You need hundreds, ideally thousands, of people before a pattern means anything for the rest of us.
3. Was this tested in humans, or in mice and lab dishes?
Mouse studies and cell studies matter, they're a real first step. But you are not a 20-gram mouse, and you definitely don't live your whole life in a petri dish.
Here's the number that stuck with me: researchers estimate that around 90% of treatments that look promising in animal studies fail once they're tested in actual humans. Ninety percent. If the article doesn't clearly say "in humans," file it under "interesting, not actionable yet."
4. Did they test one clear idea, or twenty random ones until something stuck?
Imagine a researcher testing whether peanuts cause acne. Instead of one clean question, they test 20 different peanuts against 1,000 teenagers. By pure chance, the groundnut group has slightly more breakouts that week. Nineteen other types show nothing. Guess which peanut makes the headline.
This is called p-hacking, and it's a real enough problem that the American Statistical Association put out a formal warning about it. The more specific and oddly precise a claim sounds ("eating blueberries at 8am improves memory"), the more skeptical I get.
5. Is that percentage as scary as it sounds?
This is the one that gets me the most, because it's everywhere, even in respectable outlets.
"Eating processed meat raises your risk of Disease X by 50%!" sounds alarming. Now here are the real numbers behind a claim shaped like that:
Without processed meat: 1 in 10,000 people get Disease X With processed meat: 1.5 in 10,000 people get it
That 50% jump is technically true. It's also 50% of almost nothing. Researchers who study this exact issue have found that reporting the scary percentage without the real numbers behind it is one of the most common ways health reporting misleads people, and it happens more than it should, even in serious medical journals.
Always ask: what's my actual chance of this happening, in real numbers?
Before you go
Real science is slow and cautious. It says "may be linked to," not "cures." It shows you its numbers instead of hiding them behind a scary percentage.
Bad health journalism is loud, certain, and usually trying to sell you something.
Next time a headline makes your chest tighten or makes your hand reach for your card, run it through these five questions first. I promise, most of the panic won't survive the fact-check.
Talk soon, Cassie
P.S. If a headline had you fooled this week, hit reply and tell me. I read every single one, and honestly, it's usually the most interesting part of my inbox.