Few things create more unnecessary food anxiety than a headline announcing that a common kitchen ingredient "increases disease risk by 25%." To most readers, "25% higher risk" sounds as if one out of every four people who eat that food will develop the condition. In reality, that headline is almost certainly reporting relative risk—a ratio between two groups—while omitting the baseline absolute risk.
Without knowing the starting baseline probability (absolute risk), a relative percentage change is mathematically incomplete. A 25% relative increase on top of a rare baseline event of 4 in 1,000 changes the absolute rate to 5 in 1,000—an absolute risk difference of just 1 person per 1,000 (0.1 percentage points).
Drawing on the National Center for Complementary and Integrative Health (NCCIH) Know the Science resources and epidemiological definitions from the Linus Pauling Institute, this guide walks through how to calculate and interpret both numbers whenever you read nutrition news.
What Absolute Risk and Relative Risk Actually Measure
Absolute risk is the actual probability that an event occurs in a specific population over a defined time period. For example, if researchers follow 1,000 people with low intake of a food for 10 years and 20 of them develop a condition, the absolute risk in that group is 20 out of 1,000, or 2.0%.
Relative risk (RR)—or a hazard ratio (HR) in time-to-event cohort studies—compares the absolute risk in one group against the absolute risk in another group by dividing the two rates. If a second group of 1,000 people with high intake of that food experiences 25 cases over 10 years (an absolute risk of 2.5%), dividing 2.5% by 2.0% yields a relative risk of 1.25, which gets reported as a "25% relative increase" even though the absolute risk difference between the two groups is only 0.5 percentage points (5 cases per 1,000 people over 10 years).
- Absolute Risk: Cases divided by total people in a group over a stated time period (e.g., 20 per 1,000 over 10 years = 2.0%).
- Absolute Risk Difference: The simple subtraction between two absolute risks (2.5% minus 2.0% = 0.5 percentage points, or 5 in 1,000).
- Relative Risk (RR): The ratio of the two risks (2.5% / 2.0% = 1.25, or a 25% relative increase).
Checking Confidence Intervals and Extreme Intake Quartiles
Beyond distinguishing relative from absolute risk, two additional details in a scientific paper tell you how much weight to give a reported risk ratio: the 95% confidence interval (CI) and the definition of the exposure groups. A 95% confidence interval shows the range of values within which the true population estimate is likely to fall; for a relative risk or hazard ratio, if the 95% confidence interval crosses 1.00 (where 1.00 means zero difference between groups), the finding is typically not statistically significant.
Furthermore, nutrition studies often compare the highest intake quantile (such as the top 20% of consumers) against the lowest intake quantile (the bottom 20%). Occasionally, "high intake" in an extreme group means consuming five servings a day of an ultra-processed item, which has little bearing on using a modest culinary portion once or twice a week.
- Relative Risk = 1.00: Indicates identical risk in both groups (null association).
- 95% Confidence Interval: A narrow interval far from 1.00 indicates a more precise estimate; a wide interval crossing 1.00 indicates high uncertainty.
- Modest Relative Risks in Observational Data: Small observational relative risks (such as RR 1.05 to 1.15) can easily reflect residual lifestyle confounding rather than a true biological effect.
A 3-Step Checklist Before Changing Your Groceries Over a Headline
When you encounter a dramatic percentage claim about a food, beverage, or culinary spice, pausing to ask three concrete questions turns sensational numbers into calm, practical perspective.
Overall dietary patterns—eating plenty of vegetables, legumes, whole grains, nuts, and seeds while keeping added sugars, sodium, and saturated fats in check—consistently matter far more to long-term health than minor relative risk fluctuations attached to any single ingredient.
- Ask for the baseline absolute numbers ("Out of 1,000 people over 10 years, how many cases occurred in each group?").
- Check the actual serving amounts compared between the highest and lowest intake groups.
- Look at whether the finding comes from a single observational study or a consistent body of randomized trials and systematic reviews.
Sources & Classical References
- NCCIH — Know the Science — Primary NIH educational resource on evaluating health news stories, risk statistics, and scientific journal articles.
- Linus Pauling Institute Micronutrient Information Center — Glossary — Reference definitions for relative risk, odds ratio, confidence intervals, and epidemiological study designs.
