Absolute vs. relative risk: how to read a health headline

Absolute risk is the starting chance

Absolute risk describes the chance of an event over a specified time. Relative risk compares that chance between groups. NIH explains why looking only at a relative percentage can make a benefit seem larger than the underlying numbers suggest. Both descriptions can be correct; they answer different questions. NIH: understanding absolute and relative health risks

“Four out of 200 people over five years” is an absolute-risk statement. “Half the risk of the comparison group” is relative. Neither tells you much unless you know what happened to those people: a symptom, a diagnosis, a hospital admission, or something else.

Put the same headline beside two different baselines

These are invented arithmetic examples, not results from a treatment trial. Each compares two groups of 200 people over the same five years.

For an absolute difference, subtract the risks. For a relative reduction, divide that difference by the comparison risk. “One percentage point” and “50 percent” can therefore describe the same change in Example A. These calculations follow the distinction explained in Smart Health Choices. Irwig et al.: risk differences and relative risk reduction

The table does not tell you which decision to make. A small absolute difference can matter when the outcome is serious and the intervention has little downside. A larger difference still needs to be weighed against harms and inconvenience.

Keep the time frame and outcome unchanged

A one-year chance is not directly comparable with a lifetime chance. A diagnosis is not the same outcome as dying from that disease. The National Cancer Institute's risk charts deliberately separate time frames and distinguish diagnosis from death; they also explain that their estimates do not account for every personal risk factor. NCI: risk charts, time frames, and their limitations

When comparing two reports, write the event and time period above the numbers. If they differ, stop before doing the subtraction. “Five fewer hospital admissions per 1,000 over a year” is more interpretable than a large percentage without a time horizon.

Do not turn a group estimate into a personal forecast

Your starting risk may differ from the study average. Age, medical history, and other characteristics affect how relevant the result is; NIH advises checking which people the statistic describes. An article aimed at men over 40 still covers people with very different histories. NIH: understanding absolute and relative health risks

Arithmetic also cannot establish what caused a difference. Ask whether the underlying study tested an intervention or merely observed different groups. Do not treat a reported association as a guarantee that changing one behavior will reproduce the percentage.

Use a short question at your next appointment

Try: “For people like me, how many would have this outcome with and without the option, over the same period—and what are the downsides?” Ask how uncertain the estimate is and whether the difference is likely to matter in your situation.

Bring the original link, not only a screenshot of its headline. Leave room for “we do not have a reliable number” rather than forcing an exact personal prediction. Our doctor-visit preparation guide can help keep that discussion focused.

The useful habit is not to dismiss every large percentage. It is to put the missing starting number back beside it before deciding how much weight it deserves.