The mortality hazard ratio is a relative measure of how much more (or less) strongly a factor influences the probability of death or an event. In statistics it is called the hazard ratio (HR) and is derived from the Cox proportional hazards model.
The HR answers the question: how many times higher is the "instantaneous" risk in one group than in another — for example, in smokers versus non-smokers.
How to read a hazard ratio
HR = 1 — the risk in the two groups is the same, no difference. HR above 1 — the risk is increased: HR = 1.6 means a 60% higher "instantaneous" hazard of the event. HR below 1 — the factor is protective: HR = 0.2 corresponds to an 80% reduction in risk.
Importantly, this is a relative, not an absolute value: HR = 2 does not mean "twice the chance of dying" at a given moment. Significance is confirmed by the 95% confidence interval — if it does not cross 1, the effect is statistically significant.
The Cox model and biomarkers
The Cox model makes it possible to estimate the contribution of each factor (age, blood pressure, inflammation, lipids) to the risk of death while holding the others constant. It is on such coefficients that prognostic scales are built.
For example, a simple index of two cytokines — interleukin-6 and the soluble TNF receptor — predicts 10-year mortality in older adults with an HR of about 1.6 per standard deviation of inflammation.
Where you encounter this coefficient
Hazard ratios lie at the heart of "aging clocks": the PhenoAge scale converts biomarkers into an estimate of the risk of death and then turns that into a biological age. So an abstract HR becomes an understandable "age" figure and helps measure longevity.
Reference ranges
How to interpret the hazard ratio (HR) of a mortality factor:
| HR = 1 | risk unchanged |
| HR = 1.5 | risk 50% higher |
| HR = 2.0 | risk twice as high |
| HR = 0.7 | risk 30% lower |
| Significance | 95% CI does not include 1 |
Calculate using this marker
Calculators where Mortality risk ratio is used directly:
Frequently asked questions
Are HR and relative risk the same thing?
Close, but not identical. Relative risk compares the cumulative probability by the end of follow-up, whereas the hazard ratio compares the instantaneous risk at each point in time. In practice, for rare events their values are similar.
What does HR = 1.6 mean for mortality?
That the "instantaneous" risk of death in this group is 60% higher than in the comparison group, all else being equal. It is a relative value: the absolute number of additional deaths depends on the baseline risk in the population.
Related terms
Sources
- Wikipedia — Hazard ratio
- Varadhan R, et al. Inflammatory index predicts 10-year mortality, J Gerontol A (2013)