Before You Believe a Big Percentage in a Headline
By PiedmontPrimaryCare.com Wellness Team | Last updated: September 24, 2026
When a headline says something “cuts risk by 50%,” the number alone does not tell you how much your own chances change. Relative risk compares the chance in two groups, often as a percentage. Absolute risk is how often something actually happens, such as 1 person in 1,000. Looking at both gives you a fuller picture of a claim.
This guide gives you a short process, a few plain-language questions, and a checklist you can print and bring to a health visit. It does not tell you what to do about your health. It helps you ask better questions.
If you are having symptoms right now that worry you, do not wait on a news story. Contact your health care provider, or call your local emergency number in an emergency.
A Quick Map: What to Do When a Statistic Catches Your Eye
- Pause. Do not share or act on the number yet.
- Find the “how many.” Look for the actual counts, not just the percentage.
- Ask “compared with what?” Find the starting point the percentage is measured against.
- Check the study basics. Who was studied, how many, for how long, and who paid.
- Write your questions down. Bring them to your provider before you rely on the claim.
What Is the Difference Between Absolute Risk and Relative Risk?
The Association of Health Care Journalists (AHCJ), an association for health reporters, explains these terms in its health journalism glossary.
- Absolute risk is the chance that something will happen within a given amount of time, stated in raw numbers. Example: 2 out of 1,000 people.
- Relative risk is a ratio of likelihood. It compares the chance in one group to the chance in another. It is often shown as a percentage, such as “50% lower,” or as a multiple, such as “three times more likely.”
AHCJ says the two are distinct and that ideally both are given. It also says that when the starting rate is low, a large relative increase can still mean a small increase in absolute terms. It cautions that headlines should not use relative risk in a way that suggests a far greater risk than the absolute numbers do.
Can You Show Me an Example With Made-Up Numbers?
The Wellness Team invented these numbers to show the math. They are not real data about any condition, product, or treatment.
Example A. Imagine a made-up event that happens to 2 out of every 1,000 people in one group. In a second group it happens to 1 out of every 1,000 people.
- Relative view: “The second group has 50% less of the event.” That sounds big.
- Absolute view: The difference is 1 person out of 1,000. That is 0.1 percentage points.
Example B. Now imagine the event happens to 2 out of 100 people in one group and 4 out of 100 in another.
- Relative view: “Twice as likely,” or “100% higher.”
- Absolute view: 2 more people out of every 100.
Both examples use the same kind of comparison. The absolute numbers tell you how big the change is in real life. The starting point matters too. A “doubling” of a very rare event is still a rare event.
What Is the Difference Between a Percent and a Percentage Point?
This is our own plain-language arithmetic, not a point drawn from the sources. A percentage point is the plain gap between two percentages. Going from 2% to 4% is a rise of 2 percentage points. It is also a 100% relative increase. The two terms are easy to mix up, so when you read a number, check which one is being described.
What Should I Check in a Health News Story?
MedlinePlus, a service of the National Library of Medicine, says some news stories about medical research may not include all the facts you need. It suggests asking these questions:
- Did the research involve people or animals?
- If people, how many were in the study, and who were they?
- How long was the study?
- What type of study was it?
- Who paid for the research?
MedlinePlus also says to talk with your health care provider before you rely on information you find. Your provider can tell you whether it is good for you. MedlinePlus has more on evaluating health information and links to a page on understanding medical research.
What Extra Questions Help With Statistics?
These questions are our own framework for reading numbers. They build on the source questions above and are not official guidance.
- Compared with what? Is the comparison group described?
- Over what time? One year and ten years can give different numbers.
- Out of how many? Can you find the real counts?
- Who was in the study? Do those people look like you or the person you are asking about?
- Is only a percentage given? If so, look for the absolute numbers before drawing conclusions.
- What is missing? Does the story mention downsides, limits, or things researchers are unsure about?
What Can I Ask My Provider About a Statistic?
The Agency for Healthcare Research and Quality (AHRQ) publishes The 10 Questions You Should Know. Several of them fit well when a news number leads you to a health choice. AHRQ's list includes these:
- What is the test for?
- Are there any side effects?
- Why do I need this treatment?
- Are there any alternatives?
- What are the possible complications?
You can add one of your own: “This story said a number changed by a certain percent. What does that mean for someone in my situation?” Your provider knows your health history. A news story does not.
Printable Checklist: Reading a Health Statistic
Copy this list, print it, or keep it on your phone.
- ☐ I wrote down the headline and where I saw it.
- ☐ I found whether the story says “percent” or “percentage points.”
- ☐ I looked for the real counts, such as “__ out of __.”
- ☐ I found the comparison group and the time period.
- ☐ I checked whether the study involved people or animals.
- ☐ I noted how many people were studied, who they were, and how long the study ran.
- ☐ I noted the type of study and who paid for it.
- ☐ I looked for the same information from another reliable source.
- ☐ I wrote down my questions for my provider.
- ☐ I did not change any medicine or health routine because of a news story.
What Are the Limits of This Guide?
This guide explains how to read numbers. It cannot tell you whether a specific study is good, whether a claim is true, or what is right for your health. The study details MedlinePlus asks about, such as who was studied, how many, and for how long, help you judge whether a result may apply to you. Statistics are one piece of the picture.
Where This Information Comes From and Who We Are
The definitions of absolute and relative risk, and the point about low starting rates, come from the AHCJ Health Journalism Glossary, as we reviewed it on September 24, 2026. The news-story questions come from MedlinePlus (National Library of Medicine), last updated February 2024 on the page we reviewed. The provider questions come from AHRQ's “The 10 Questions You Should Know,” last reviewed November 2020. The example numbers, the percent versus percentage point explanation, and the extra questions are the Wellness Team's own.
PiedmontPrimaryCare.com is an independent wellness information website. It is not a medical practice and does not provide health care. This domain was previously used by a healthcare provider that is no longer associated with this website, and the site is not affiliated with any medical practice using a similar name. This article was not written or reviewed by a doctor or other clinician. You can read more on our About page and our How We Work page, which explains how the site makes money.
Educational disclaimer: This article is general information, not medical advice. It does not diagnose, treat, or recommend any product or plan. Talk with your health care provider before making decisions about your health, and never start, stop, or change a medicine because of a news story. In an emergency, call your local emergency number.