HTA Glossary

Hazard ratio (HR)

A hazard ratio (HR) compares the rate at which an event, such as death or disease progression, occurs in one group with the rate in another group during the follow-up of a study. A hazard ratio of 1 means no difference; a value below 1 means the event occurs at a lower rate in the treatment group.

By HTA Course · Last reviewed

In more detail

HR explained

A hazard ratio of 0.75 for death means that, at any moment during follow-up, patients in the treatment group who are still alive die at a rate 25% lower than those in the control group. It does not mean that 25% fewer patients die in total, or that patients live 25% longer.

Hazard ratios are usually estimated with a Cox proportional hazards model. The model assumes that the ratio stays constant over time. When survival curves cross or separate late, that assumption is unlikely to hold and a single hazard ratio can mislead.

A hazard ratio is reported with a confidence interval. If the 95% interval includes 1, the difference is not statistically significant at the conventional 5% level. In HTA, hazard ratios feed indirect comparisons and network meta-analyses, and economic models apply them to a baseline survival curve.

Go further

Learn HTA from A to Z

The HTA Course puts this term in context, with worked examples and interactive lessons across 15 modules and 79 self-paced lessons, and a certificate of completion.