Absolute risk vs relative risk in cancer headlines

The headline says a treatment “cuts risk in half.” You screenshot it, send it to the family group chat, and wait for the next oncology visit with one question stuck in your head: half of what — and for people like you?

This is not clinical evidence and not medical advice. It does not replace your care team. Discuss anything here with your clinician before you act on it.

Same story, two different numbers

NIH’s News in Health walks through an illustrative example (not a real cancer-drug trial result). Imagine a disease that affects 2 in 100 middle-aged men over their lifetimes. A medicine lowers that chance to 1 in 100. The starting risk was 2 in 100. The ending risk is 1 in 100. The difference — 1 person in 100, or 1 percentage point — is the absolute change.

You can also say the risk was cut by 50%, because 1 is half of 2. That 50% is the relative change. Looking only at the relative number, the medicine can sound highly effective. NIH notes that relative risk often sounds much greater than the absolute change, which is why the absolute figure is the one to keep in view.

Those two ways of talking about the same comparison have names. Absolute risk is the chance that something happens over a stated period. Relative risk — NCI also calls it a risk ratio — compares the chance in one group with the chance in another. In cancer research, risk ratios show up in forward-looking studies such as cohort studies and clinical trials. A ratio of one means no difference between groups. A ratio above or below one usually means higher or lower risk, or that two treatments do not have the same effects.

NCI’s dictionary also gives a concrete absolute-risk example for cancer: a 35-year-old woman with no known breast-cancer risk factors has about a 13% lifetime absolute risk of developing breast cancer — roughly 1 in 8 — which also means about an 87% chance she will never have it. Absolute risk always comes with a time window and a starting point. Without those, “risk” is fog.

Why headlines lean on the relative percent

Press releases and study summaries often lead with the relative percent because it is short and dramatic. “Cuts risk in half” fits a title. “From 2 in 100 down to 1 in 100” takes more words and sounds smaller — even when both sentences describe the same result.

Framing shapes how numbers feel even when the math is identical. NIH points out that “more than 20% of Americans will eventually die of cancer” lands differently from “nearly 80% of Americans will not die of cancer,” and that the same idea can feel clearer as “more than 1 in 5.” None of that means the relative number is false. It means it is incomplete until you also see the starting risk.

Ask for the baseline, not only the percent off

When a clinician, paper, or news brief says “risk reduction,” try to get four pieces on the table: the starting risk without the intervention, the ending risk with it, the time window (five years? ten? lifetime?), and who was in the study — age, sex, disease stage, prior treatments. Absolute risk is the pair of starting and ending chances. Relative risk is the comparison between them.

AHRQ’s health-literacy guidance for patient materials says the same thing in clinic language: give absolute risk — for example, a decrease from 6 out of 100 people to 3 out of 100 — rather than only a relative reduction of 50%, especially when the numbers are small. Keep the denominator the same when you compare. Say both sides: how many will have the outcome, and how many will not. Prefer a time span that means something in real life, such as ten years rather than an abstract lifetime, when that is how the evidence is reported.

Back to NIH’s illustrative numbers: another way to say a 1-in-100 absolute difference is that you would need to treat 100 people with the medicine to prevent one additional person from getting the disease. Researchers sometimes call that idea “number needed to treat.” It can help, but it is still just the flip side of the absolute difference. Ask for the 2-in-100 and 1-in-100 style figures first. Then, if someone offers a “treat X to help one” number, you already know what it came from.

Group averages are not a personal forecast

Even a carefully reported absolute risk is usually a group estimate. MedlinePlus puts it plainly for genetic and other health statistics: the numbers are not personalized; they describe groups. A genetics professional or other clinician can help interpret what they mean for one person once family history, medical history, and other factors are in the room.

NCI experts making the same point about breast-cancer risk tools say the models present population averages. If a tool gives a 5% risk over a set period, that means about 5 out of 100 people with similar listed factors may develop the cancer in that window — not that the tool can name which five, and not a guarantee either way for any one person. Their advice matches what you already need for trial headlines: talk with a health care provider about what the estimate means for you.

MedlinePlus’s page on health statistics adds a useful habit: numbers on a chart can look straightforward when they are not. Be critical. Consider the source. Ask questions until the figure is clear enough to discuss with your team.

Questions that keep the conversation concrete

NIH’s “Wise Choices” list for talking with a doctor about health risks is a good starter sheet for a portal message or visit note:

What are my personal health risks — and are those over a lifetime or a shorter period? What are my options for treatments, tests, or other changes? What will this test or therapy tell us about my risks? What are possible benefits and side effects, and what are the chances those might affect me?

Add two risk-math questions: What was the risk in people like me without this option, and with it? If someone quotes a percent reduction, can we write starting and ending numbers as “X out of 100” over the same time window?

Where Triangle’s cancer evidence library fits

Triangle’s cancer evidence library helps you explore public cancer research — study records, labels, and other materials you can open yourself. It is not a second opinion, not a personalized risk calculator, and not a claim that Triangle reviewed the science for your case. Use it to find the source behind a headline. Then bring absolute numbers, the time window, and your history back to your clinician.

The next time a headline says “cuts risk in half,” you do not have to settle for the relative percent alone. Ask for the starting risk, the ending risk, and the clock. Absolute and relative are two views of one comparison. Absolute is usually the view that keeps the size of the change visible.