How to interpret the value of ‘Hazard Ratio” in practice?

The Hazard ratio (HR) is one of the measures that in clinical research are most often difficult to interpret for students and researchers. In this post we will try to explain this measure in terms of its practical use.

You should know what the Hazard Ratio is, but we will repeat it again.

Let’s take as an example a study in which I want to evaluate the survival of patients exposed to smoke compared to those exposed to high-calorie diet.

In this case the Hazard Ratio is the ratio between the mortality rate in the group of patients exposed to smoke and the mortality rate in the mortality rate in the group of subjects exposed to high-calorie diet.
If you don’t like the definition “mortality rate”, you can interpret it as a “speed of death”, even if it’s not quite the same thing.

I’m not going to explain how to calculate the Hazard Ratio because this article is not about it. We are interested here in its interpretation and the way it should be reported in the literature.

Let’s go ahead.

Let’s say for example that you have estimated the hazard ratio between the experimental and the control groups using a statistical model (a classic example: a Cox model) and its value, let’s say, is 2.2.

How can we report and interpret the value of 2.2 in terms of practical use?

  • The mortality rate in smokers is 2.2 times higher of that in the high-calorie diet group.
  • The mortality rate of those exposed to smoking is 220% of that exposed to high-calorie food.
  • Exposure to smoking increases the mortality rate by 220% compared to exposure of a high-calorie diet.
  • At each time point of an observation, for every 100 deaths due to high-calorie diet, there will be 220 deaths for the reason of smoking.


Now let’s take a HR less than 1. Let’s say that in your experiment the calculated Hazard Ratio is equal to 0.65.
This is how you can interpret and report it.

  • The mortality rate in a group of smokers drops by 35% compared to the group of high-calorie diet.
  • The mortality rate among smokers is 0.65 times of that among patients with a high-calorie diet.
  • mortality rate of smokers is 65% of that of gluttons.
  • At each time point of an observation, for every 100 deaths due to high-calorie food consumption, there will be 65 due to smoke.

That’s all.
In the description or comments to your results add this “practical use” and it will easier to a reader to understand it.

Leave a Reply

Your email address will not be published. Required fields are marked *