A hazard ratio is a time-to-event effect measure that compares the instantaneous event rate between two groups over a follow-up period.
What it measures
Hazard ratios appear when researchers study events such as diagnosis, relapse, hospitalization, cardiovascular events, or death. A ratio of 1 means no difference in event rate between groups. Values below or above 1 indicate lower or higher event rates in the group being compared.
The metric is about event timing across follow-up, not simply the final percentage of people who had an event.
Why interpretation is tricky
A hazard ratio is often mistaken for an absolute risk change. It is not. A lower hazard can still translate to a small absolute difference when the baseline event rate is low.
The proportional hazards assumption also matters. If group differences change sharply over time, a single ratio may hide a more complicated pattern.
Where it shows up
Hazard ratios are common in large observational studies, survival analyses, and some long randomized controlled trials. They are less common in short supplement trials that measure continuous outcomes such as mood or sleep.
They should be read with a confidence interval. An interval crossing 1 means the data do not clearly separate higher from lower hazard under the model.
How it affects evidence
For recommendation ranking, a hazard ratio is useful only when converted into a decision-relevant frame: baseline risk, absolute risk difference, population match, and study design quality.
A meta-analysis can pool hazard ratios when studies define the event and follow-up similarly.
Safety note
Hazard ratios in population studies do not tell a user whether a supplement is safe with their medications, conditions, or pregnancy status.