Statistically Significant
Probably not a chance finding — which is different from being large enough to matter.
What it means
That a difference found in a study is unlikely to have arisen purely by chance. Conventionally this means a p value below 0.05 — less than a one in twenty probability of seeing that result if there were really no effect.
What it does not mean is important. A very large study can find a statistically significant difference that is far too small to matter to anyone — a drug that lowers blood pressure by 1mmHg, for instance.
"Clinically significant" is the question that actually matters, and it is a different question entirely.
In practice
In research papers, and in press coverage of them.
Frequently used in headlines as though it meant "proven" or "important", which it does not.
Why it matters
Ask how big the effect is, not just whether it was significant. Those are two different questions and only the first tells you anything useful.
Look for confidence intervals, which show the range the true effect probably sits within — much more informative than a p value.
Be wary of a single significant result. Findings that hold up are ones that replicate.
Clinically reviewed by Dr Mohammad Zubair Khan, GMC 7563469
Last reviewed
August 30, 2026