Skip to main content
Supplement ScienceSupplementScience

Relative Risk, Absolute Risk & Confidence Intervals — in Plain English

This content is for informational purposes only and does not constitute medical advice. Statements about dietary supplements have not been evaluated by the FDA and are not intended to diagnose, treat, cure, or prevent any disease. Individual results may vary — consult your healthcare provider before starting any supplement. Full disclaimer

Relative risk compares the chance of an outcome between two groups (for example, 20% lower), while absolute risk is the...

Relative risk compares the chance of an outcome between two groups (for example, 20% lower), while absolute risk is the actual change in your odds (for example, from 2% to 1.6%). A confidence interval shows the range the true effect likely falls within — if it crosses 'no effect,' the result isn't statistically significant. Large relative numbers can hide tiny absolute differences.

Key Takeaways

  • Relative risk compares two groups; absolute risk is the actual change in your odds — and they can sound very different.
  • A 20% relative reduction can mean a real change of well under 1 percentage point.
  • A confidence interval shows the plausible range for the true effect; crossing 'no effect' means it isn't statistically significant.
  • Narrow confidence intervals come from larger, more precise studies; wide ones signal uncertainty.
  • A low p-value means a result probably isn't chance — not that it's large or clinically meaningful.

Get the free evidence-based Relative Risk, Absolute Risk & Confidence Intervals — in Plain English guide — delivered in 60 seconds.

No spam. Unsubscribe anytime.

Why the Numbers Trip People Up

Study headlines love big percentages. But the same result can sound dramatic or trivial depending on how it's expressed. Knowing three terms — relative risk, absolute risk, and the confidence interval — lets you see what a finding really means.

Relative vs Absolute Risk

Suppose a supplement lowers the yearly chance of some outcome from 2% to 1.6%.

  • Relative risk reduction: the drop from 2% to 1.6% is a 20% *relative* reduction. Impressive-sounding.
  • Absolute risk reduction: the actual change is 2% minus 1.6% = 0.4 percentage points. Out of 1,000 people, that's 4 fewer affected.

Both describe the same result. Marketing tends to quote the bigger relative number; the absolute number tells you what it means for one person.

Number Needed to Treat

A related idea: if 0.4% of people benefit, you'd give the supplement to about 250 people for one of them to avoid the outcome. That figure — the number needed to treat — turns statistics into something tangible.

Confidence Intervals

No study measures the 'true' effect exactly; it estimates it. A 95% confidence interval is the range in which the true value most plausibly sits.

  • A narrow interval means a precise estimate (often from a large study).
  • A wide interval means a lot of uncertainty (often a small study).
  • If the interval crosses the point of 'no effect' (a relative risk of 1.0, or a difference of 0), the result is not statistically significant — the data can't rule out 'no real effect.'

P-values, Briefly

A p-value below 0.05 is the common threshold for calling a result 'statistically significant.' It does *not* tell you the effect is large or important — only that it's unlikely to be pure chance. A statistically significant result with a tiny absolute benefit may not matter in real life.

The Takeaway for Reading Studies

Ask for the absolute numbers, check whether the confidence interval crosses 'no effect,' and treat any single small study cautiously. NCCIH and MedlinePlus both publish plain-language primers on these statistics [1][2].

Frequently Asked Questions

Why do ads use relative risk instead of absolute risk?

Because relative numbers sound bigger. 'Cuts risk 20%' is more persuasive than 'lowers risk by 0.4 percentage points,' even when they describe the identical result. Asking for the absolute change keeps things honest.

What does a 95% confidence interval mean?

Roughly, it's the range in which the true effect most plausibly lies. If you repeated the study many times, about 95% of such intervals would contain the real value. A wide interval signals more uncertainty.

What does it mean if a confidence interval 'crosses 1.0'?

For a relative risk or ratio, 1.0 means 'no difference.' If the interval includes 1.0, the study can't rule out no effect, so the result is not statistically significant — even if the central estimate looks like a benefit.

Is a 'statistically significant' result always important?

No. Significance only means the result is unlikely to be chance. With a very large study, even a trivial difference can be statistically significant. Always check the absolute size of the effect to judge whether it matters in practice.

References

  1. National Center for Complementary and Integrative Health (NCCIH) (2024). Know the Science. NIH National Center for Complementary and Integrative Health.
  2. MedlinePlus, U.S. National Library of Medicine (2024). Evaluating Health Information. MedlinePlus (NIH National Library of Medicine).