What a Medical Reviewer Actually Does

Somewhere on the internet right now, an article about blood pressure is telling readers something that was true in 2011. It is well written. It ranks on Google. It carries a little badge that says "medically reviewed." And it is wrong.

That badge is supposed to mean something. Often it does. Sometimes it means a doctor glanced at the piece for eleven minutes between clinics, found nothing actively dangerous, and signed it off. The difference between those two versions of "reviewed" is invisible to readers — which is exactly why it's worth explaining what a proper medical review involves.

Three different questions, usually asked as one

When a clinician reviews a piece of health content properly, they are answering three separate questions that tend to get blurred together.

Is it accurate? This is the obvious one: do the claims match the evidence? But accuracy is the easiest of the three, because outright factual errors are rare in professionally written content. Writers are good at not saying false things.

Is it safe? This is harder, and it's where most content fails. A piece can be entirely accurate and still unsafe — usually through omission. An article about headaches that never mentions the symptoms that need urgent assessment is accurate in everything it says and dangerous in what it doesn't. Safety review is largely the work of asking: who could read this, act reasonably on it, and come to harm? That question requires clinical experience, because you need a mental library of the people who actually turn up in emergency departments having read something on the internet.

Is it current? Medicine moves. Guidelines are rewritten, drug safety alerts are issued, thresholds shift. The blood pressure article from 2011 wasn't wrong when it was written. Content review has to include a currency check against present guidance — in the UK, that usually means NICE, the relevant royal college or specialist society, and the MHRA's safety updates.

A reviewer who checks only the first question has done a fact-check, not a medical review.

What actually changes between draft and publication

The edits a clinical reviewer makes are rarely dramatic. They are small, specific, and load-bearing. A few examples of the genre:

A sentence saying a medicine "has few side effects" becomes a sentence naming the common ones and the rare serious one. "Few side effects" is marketing; the named list is medicine.

A phrase like "studies show" gets traced back to its source. Sometimes the study exists and says what's claimed. Often it exists and says something narrower — a result in one population, at one dose, over twelve weeks — that the article has quietly generalised to everyone, forever.

A confident claim acquires its missing denominator. "Doubles the risk" means very little until you know the risk doubled from what to what. Doubling a one-in-ten-thousand risk and doubling a one-in-ten risk are different conversations, and readers deserve to know which one they're having.

And almost always, a safety-netting line gets added: the specific circumstances in which the reader should stop reading articles and speak to a clinician.

Why the reviewer's background matters

Not all clinical eyes see the same things. A doctor who has worked in emergency medicine reads content with a particular bias: towards the presentations that go wrong, the symptoms that are usually benign but occasionally aren't, and the predictable ways people misjudge their own risk. Someone with a pharmacology-heavy background reads medication content differently again — interactions, contraindications, and the gap between a drug's licence and the way it's discussed online.

The point isn't that one background is better. It's that "medically reviewed" should prompt the question: reviewed by whom, with what relevant expertise, against what standard? The publishers who take this seriously — and there are good ones — can answer that question. The ones who can't are buying a badge.

What good review looks like from the outside

Readers can't audit the review process, but there are visible signs of one. Content that names its sources, and whose sources are guidelines and primary literature rather than other websites. Dates of publication and review. Specific numbers instead of vague reassurance. The presence of a "when to seek help" section that feels written rather than pasted. Language that distinguishes what is known, what is likely, and what is still argued about.

None of these guarantee quality. Their absence, though, tells you a lot.

Practical takeaways

• "Medically reviewed" spans everything from rigorous clinical scrutiny to a rubber stamp — the badge alone tells you little.
• A proper review asks three questions: is it accurate, is it safe, and is it current. Most weak content fails the second and third, not the first.
• The most important edits are usually additions: named side effects, absolute risks, sources, and safety-netting.
• Visible markers of real review include dated revisions, guideline-level sources, and specific numbers in place of reassurance.

What this doesn't mean

It doesn't mean unreviewed content is always wrong, or that a reviewer's sign-off makes content infallible. Reviewers miss things; evidence changes after publication. Review reduces risk; it does not abolish it.

A closing thought

The best compliment a medical reviewer can receive is that readers never noticed they existed. The work, done well, is invisible — a set of errors that never reached anyone, a dangerous omission that became a paragraph, a 2011 fact quietly retired. Health content is one of the few kinds of writing where the editing can matter more than the prose. It deserves to be done by people who know where the bodies are buried — sometimes literally.

Further reading and sources

• General Medical Council — Good Medical Practice (duties around communication and honesty)
• Patient Information Forum — PIF TICK criteria for trustworthy health information
• International Committee of Medical Journal Editors (ICMJE) — recommendations on editorial standards
• NICE — guidance library (the currency benchmark for UK content)
• MHRA — Drug Safety Updates