Compare Absolute and Relative Risk
Introduction 🎉
Welcome to Decode How Risk Gets Framed! In the courses before this one, you explored how randomness behaves, why small samples can lead us astray, and how to read test results for rare events. This course takes a different turn: instead of the underlying facts, you will focus on how risk information is presented — and why the same number can feel alarming or reassuring depending on the words wrapped around it.
This first lesson digs into one of the most common sources of confusion in health news, advertising, and policy debates: the gap between absolute risk and relative risk. In this lesson, you will learn to:
- Distinguish absolute risk from relative risk by recognizing when a claim states a standalone probability versus when it describes how one group's risk compares to another's.
- Translate between the two framings using the relationship to turn a proportional change into a concrete number.
- Judge whether a scary-sounding headline deserves attention by always hunting for the baseline and measuring the actual absolute difference it implies.
A Tale of Two Headlines 📰
Imagine reading these two statements about the same medical finding:
- Headline A: "New study finds treatment cuts risk of heart attack in half!"
- Headline B: "New study finds treatment lowers heart attack risk from 2 in 1,000 to 1 in 1,000."
Both describe exactly the same result, yet Headline A sounds dramatic while Headline B sounds modest. The difference is not in the science; it is in the framing. Headline A uses relative risk language ("cuts in half"), and Headline B uses absolute risk language ("from 2 in 1,000 to 1 in 1,000"). Learning to move between these two framings is the core skill you will build in this lesson.
A quick visual makes the equivalence easier to see:
![[Chart showing that cutting risk in half means going from 2 to 1 heart attack per 1,000 people]](https://k3-production-bucket.s3.us-east-1.amazonaws.com/uploads/9EveH9nzgozShCcmd_image.png)
What Is Absolute Risk? 🎯
Absolute risk is the straightforward probability that something happens. It answers the question: Out of a given number of people (or trials), how many experience the outcome?
You can express it as a fraction, a decimal, or a percentage. For example, if 30 out of every 10,000 people develop a certain condition in a year, the absolute risk is:
The key feature of absolute risk is that it does not require a comparison group. It does not compare one group to another or describe a proportional change. Instead, once the relevant population, denominator, and time window are defined, it tells you how common the outcome is. For example, " of people develop this condition in one year" is meaningful in a way that "50% higher risk" is not. You still need context about who is being measured and over what period, but you do not need another group’s risk to interpret the basic probability.

