A headline saying a treatment “cuts risk by 25%” can sound clear, but you need the starting risk to know what it means for real people. Relative risk compares two groups; absolute risk gives the actual chance in each group. Read both numbers together before deciding how large a result feels to you.
What is the difference between relative and absolute risk?
Absolute risk is the number of people in one group who have an outcome over a stated time. If 4 of 100 people have an outcome over five years, the absolute risk is 4%, or 4 in 100.
Relative risk compares that risk with the risk in another group. It is a ratio: the risk in the group receiving the treatment, test, or exposure divided by the risk in the comparison group. A relative risk of 1 means the two groups had the same measured risk.
Neither number replaces the other. Relative risk makes comparisons quick. Absolute risk tells you how many people were affected, which is often easier to use in a personal care conversation. A 2021 clinical review explains that relative measures can make an effect appear larger, while absolute effects show the risk people face without treatment and the change with treatment. European Urology Focus review
Can you show me the same result both ways?
Imagine a study follows 1,000 people in each group for one year. This is a made-up example to show the math, not a prediction of your own risk.
| Group | People with the outcome | Absolute risk |
|---|---|---|
| Comparison group | 8 of 1,000 | 0.8% |
| Treatment group | 6 of 1,000 | 0.6% |
The absolute difference is 0.8% minus 0.6%, or 0.2 percentage points. Put another way, there were 2 fewer outcomes for every 1,000 people over the year.
The relative risk is 0.6% divided by 0.8%, which equals 0.75. You may also see this described as a 25% relative risk reduction: 1 minus 0.75 equals 0.25, or 25%.
Both statements describe the same figures. “25% lower” is the relative description; “2 fewer per 1,000 over one year” is the absolute description. The baseline risk—the comparison group’s starting risk—sets the scale. See Relative Risk and Absolute Risk for another way a health claim can leave out that scale.
What does a relative risk of 0.75 mean?
A relative risk of 0.75 means the measured risk in one group was 75% of the risk in the comparison group. That is a relative reduction of 25%. It does not, by itself, tell you the actual number of people helped or harmed. You still need the two event counts and the follow-up period.
What do relative risks of 2.5 and 1.2 mean?
| Relative risk | Plain-language meaning | Example if comparison risk is 4 in 1,000 |
|---|---|---|
| 1 | The groups had the same measured risk. | 4 in 1,000 in each group |
| 1.2 | The measured risk was 1.2 times as high, or 20% higher. | About 4.8 in 1,000 |
| 2.5 | The measured risk was 2.5 times as high, or 150% higher. | 10 in 1,000 |
The example matters because the same relative risk can describe very different absolute changes. A relative risk of 2.5 takes a 4-in-1,000 risk to 10-in-1,000, a rise of 6 per 1,000. If the comparison risk were 40 in 1,000, it would become 100 in 1,000, a rise of 60 per 1,000.
How do I calculate absolute risk and relative risk?
Start with the number of people who had the outcome, then divide by the total number of people in that group.
- Absolute risk: people with the outcome ÷ all people in that group.
- Relative risk: absolute risk in the group of interest ÷ absolute risk in the comparison group.
- Absolute risk difference: comparison-group risk minus group-of-interest risk, when you are looking for a reduction.
For the 6-versus-8 example: 6 ÷ 1,000 = 0.006, or 0.6%; 8 ÷ 1,000 = 0.008, or 0.8%; and 0.006 ÷ 0.008 = 0.75. The comparison must be fair: the groups, outcome, and time period must match.
A relative-risk calculation is most useful when you can also see the raw counts, the percentage in each group, and how long people were followed. If a report gives a relative number without those details, put it on an open-questions list for your next visit.
What does research show about reporting both numbers?
A 2022 JAMA Internal Medicine systematic review combined 21 randomized clinical trials. A randomized trial assigns participants to comparison groups by chance; this review included adults older than 18, compared statins with placebo or usual care, and included trials planned to last at least two years. Across those trials, the reported absolute risk reductions were 0.8 percentage points for death from any cause, 1.3 percentage points for heart attack, and 0.4 percentage points for stroke. The related relative reductions were 9%, 29%, and 14%.
That example shows why the two formats belong together: relative changes may sound larger than the corresponding absolute changes. The review also reported differences among the trials, which lowers confidence that one average figure will fit every person or situation.
A separate analysis of 316 randomized medication trials published in six major medical journals reviewed trials from June 2008 through September 2010. Of 157 trials with statistically significant positive results, 69, or 44%, reported results in their abstracts only in relative terms. The authors concluded that articles should include absolute numbers as well. Journal of General Internal Medicine analysis
For any study, the result belongs to the people studied. Age, health history, and the care compared can affect whether its baseline risk resembles yours. Who Was Studied Matters can help you spot those details.
What should I look for in a health headline?
- What happened? Ask for the exact outcome, not only a broad phrase such as “better health.”
- Out of how many? Ask for the number and percentage with the outcome in each group.
- Over what time? A result over six months means something different from one over ten years.
- Compared with what? Find out whether the comparison was placebo, usual care, another treatment, or a different group of people.
- Who was included? The study size and the people enrolled help show how closely it may fit your situation.
Relative and absolute risk describe measured results, not a guarantee about what will happen to one person. They are a starting point for a clearer conversation about benefits, harms, and alternatives.
Questions to take to a clinician
- What was the absolute risk in each group, and over what follow-up time?
- What does this relative risk mean in people with a baseline risk like mine?
- How many people were in the study, and who were they?
- What was the comparison group receiving?
- What benefits and harms should I compare before making a decision?
MedlinePlus recommends writing down questions and bringing a list of medicines, vitamins, and supplements to an appointment. Johns Hopkins Medicine also advises setting an agenda and asking for a plain-language explanation when something is unclear.
By Health Evidence Literacy Guide Editorial Team
This article is for general information purposes only and does not constitute medical advice. Consult your doctor or qualified healthcare provider before making changes to your health routine.
Leave a Reply