tap to skip
Heard something confusing about health online?
Check it against the evidence — explained in plain English.
Works for a broad topic (seed oils, vitamin D) or a specific claim (does turmeric help arthritis) — paste a caption or quote and we'll turn it into a question.
The last 5 things people searched on Miss Informed
I'm a medical student who grew up in a surge of health information. Every scroll through social media brings another health claim. If you exist outside the scientific field, it's hard to know who to trust — influencers cite articles to back up their claims, but what do those studies actually say?
I created Miss Informed for people who feel lost in that gap. Scientific papers are dense, full of jargon, and hard to decode. Miss Informed uses AI, but for one job only — reading a single paper and telling you what it says, never weighing in across studies or handing you a verdict on a topic. That distinction matters: most AI-generated health answers get shaky when they're asked to synthesize dozens of papers into one tidy conclusion, quietly blending studies that don't actually agree. Here, every summary comes with the exact quote it's drawn from and a link to the original paper, so you're never taking the AI's word for it.
Search a claim or explore a topic, and you'll find papers from open-access sources, along with what they actually say, what it means for you, and a direct quote from the paper. A link to the original source is always included.
My goal is simple: ordinary people shouldn't feel helpless in the face of scientific literature. You should always feel equipped to do your own research — and confident that you can understand it.
Every card on this site carries a small label — "peer-reviewed", "meta-analysis", "preprint" — that's shorthand for something specific about how trustworthy or how strong a piece of evidence is. If you didn't spend years in a lab or a journal's editorial office, there's no reason you'd already know what any of that means, and we'd rather explain it than leave you guessing. None of these labels are meant to gatekeep — they're here so you can judge the evidence yourself instead of taking our word for it. Click any label on a result card and you'll land right back here, at the relevant entry.
Before a journal publishes a study, it sends the manuscript to other independent scientists working in the same field — reviewers who weren't involved in the work — to check it over. They question the methods, look for errors or missing data, and judge whether the conclusions are actually supported by the results, then tell the journal to reject it, send it back for changes, or approve it. It usually takes months. It isn't a guarantee of truth — reviewers can still miss things, and peer-reviewed papers do occasionally turn out to be wrong — but it means qualified people scrutinised the work before it reached the public record. That's exactly why peer-reviewed research carries more weight than a claim with no such process behind it, and why you'll see it flagged clearly on every card here.
A systematic review gathers every study that's been done on a question and summarises what they show, using a documented, repeatable method for finding and judging them — not just whichever studies the author happened to notice. A meta-analysis goes one step further and combines their results into a single overall statistic. Both sit near the top of the evidence hierarchy, because the conclusion is drawn from many studies at once rather than resting on just one.
A study where participants are randomly assigned to either receive the treatment being tested or not — often given a placebo (a fake treatment that looks the same) instead. Random assignment means the two groups end up similar in every other way, on average, so if one group does better, the treatment itself is the most likely explanation, not some other difference between who ended up in which group. That's what makes a well-run RCT one of the strongest ways to show a treatment actually causes an effect, rather than just being associated with one.
Researchers watch what happens to people who are already living their lives — comparing, say, people who happen to drink coffee against people who don't — rather than assigning anyone to a group. This is useful for spotting real-world patterns, especially when a randomised trial would be impractical or unethical (you can't randomly assign people to smoke for 20 years). The tradeoff is that some other unmeasured difference between the groups could explain the result instead of the thing being studied, so an observational study is generally considered weaker evidence of cause-and-effect than an RCT.
An expert's summary of the existing research on a topic, written in their own words. Unlike a systematic review, a plain review doesn't have to follow a documented, repeatable method for finding or weighing the studies it covers — it's a useful overview and often a good place to start, but it depends more heavily on which studies the author chose to focus on and how they read them.
A study made public before it's been through peer review — often shared early so other researchers can see the findings quickly. It might turn out to be perfectly good science, but nobody outside the author's own team has formally checked the methods or conclusions yet, so it's worth treating with a bit more caution than a peer-reviewed paper making the same claim.
A formal withdrawal of a published paper by the journal or its own authors, usually because a serious error, data problem, or misconduct was discovered after publication. A retracted paper is no longer considered part of the reliable evidence base — Miss Informed excludes retracted papers from results outright, rather than just ranking them lower, so you shouldn't run into one here at all.