Absolute Risk vs Relative Risk in Drug Side Effects: How to Interpret the Numbers

When you hear a drug reduces your risk of a heart attack by 50%, it sounds amazing. But what if your risk was only 2% to begin with? That 50% drop means you’re now at 1%-not a miracle, just a small improvement. This is the gap between absolute risk and relative risk, and understanding it could change how you decide whether to take a medication.

What Absolute Risk Really Means

Absolute risk tells you the actual chance something will happen to you. If 1 in 100 people get a rare side effect from a drug, that’s a 1% absolute risk. It doesn’t compare you to anyone else. It just says: out of 100 people like you, one might have this problem.

Let’s say a drug lowers your chance of a stroke from 4% to 3%. The absolute risk reduction is 1 percentage point. That’s it. No multiplication. No fancy percentages. Just a 1% drop in real-world likelihood.

Doctors use absolute risk to answer the question: “How likely is this to affect ME?” It’s personal. It’s concrete. If you’re one of 100 people, and only one gets a side effect, you can picture that. One person out of a classroom. One car in a parking lot of 100. That’s how your brain understands risk.

What Relative Risk Is Trying to Sell You

Relative risk compares two groups: people who took the drug versus those who didn’t. It’s a ratio. If your stroke risk drops from 4% to 3%, the relative risk reduction is 25%. How? Because 1% (the difference) divided by 4% (the original risk) equals 0.25, or 25%.

Pharmaceutical ads love this number. 25% sounds better than 1%. But here’s the trick: relative risk doesn’t tell you the starting point. A 50% reduction in risk sounds huge-until you learn the original risk was 0.02%. Then the absolute drop is 0.01%. That’s like going from 2 in 10,000 to 1 in 10,000. You’re still almost certainly not going to have the problem.

This is why you see headlines like “New Drug Cuts Heart Attack Risk by Half!” But if you’re a healthy 30-year-old with a 2% lifetime risk of a heart attack, cutting that in half still leaves you at 1%. That’s not a life-saving breakthrough. It’s a small statistical shift.

Why the Difference Matters in Real Life

Imagine two drugs for high blood pressure. Drug A reduces your chance of a heart attack from 10% to 8%. That’s a 20% relative risk reduction. Drug B reduces it from 2% to 1%. That’s a 50% relative risk reduction. Which one sounds better?

Drug B’s 50% sounds amazing. But Drug A’s 2% absolute reduction means 2 out of every 100 people taking it will avoid a heart attack. Drug B? Only 1 out of every 100 people benefits. The relative risk makes Drug B look superior. The absolute risk tells you Drug A helps more people in real terms.

Same thing with side effects. A drug might increase your risk of nausea by 200%. Sounds scary. But if the baseline risk was 2%, now it’s 6%. That’s 4 extra people out of 100 feeling sick. Not trivial-but not a disaster either. The 200% sounds like a red flag. The absolute increase? Manageable.

The Number Needed to Treat (NNT): The Real Cost of Benefit

There’s a number doctors use to cut through the noise: Number Needed to Treat, or NNT. It tells you how many people need to take a drug for one person to benefit.

If a drug reduces heart attack risk from 4% to 3%, the absolute risk reduction is 1%. So NNT = 1 ÷ 0.01 = 100. That means 100 people must take the drug for one person to avoid a heart attack. The other 99 won’t benefit-but might still get side effects.

Compare that to a drug with an NNT of 5. That’s way better. Five people take it, one benefits. That’s a strong signal. But if the NNT is 500? You’re treating 500 people to help one. That’s a weak benefit, even if the relative risk looks good.

NNT is the most honest metric. It doesn’t hide behind percentages. It just says: how many people are you asking to take a pill for one to win?

How Ads Trick You (And Why It’s Legal)

In 2021, a study found 78% of direct-to-consumer drug ads in the U.S. highlighted relative risk reduction-without ever mentioning the absolute risk. That’s not a mistake. It’s strategy.

Think about it: a 90% relative risk reduction sounds like a miracle cure. But if the original risk was 0.01%, the absolute drop is 0.009%. You’re talking about going from 1 in 10,000 to 1 in 100,000. That’s not a cure. It’s a tiny statistical tweak.

Meanwhile, side effects? Those are often reported as absolute risks. “1 in 50 people get dizziness.” That’s clear. But the benefit? “Reduces risk by 70%!”-with no context. That’s not misleading. It’s legal. And it works. Drugs marketed with relative risk reductions see 23% higher initial prescriptions.

But here’s the catch: when patients later learn the real numbers, adherence drops by 15%. People feel tricked. And they stop taking the drug.

How to Read the Numbers Yourself

You don’t need a statistics degree. Here’s how to decode what you’re hearing:

1. Ask for the baseline risk. “What’s my chance of this problem without the drug?”
2. Ask for the absolute change. “How much does this lower my risk?”
3. Ask for the NNT. “How many people need to take this for one to benefit?”
4. Compare side effects the same way. “How many people get this side effect?”

Example: Your doctor says, “This statin cuts your heart attack risk in half.”

You reply: “What’s my current risk?”

They say: “About 2% over the next 10 years.”

You say: “So it drops to 1%?”

They say: “Yes.”

You think: “So out of 100 people like me, 2 would have a heart attack. Now only 1 will. That means 98 people take this drug every day for 10 years, and only one avoids a heart attack. And 20% of people get muscle pain. Is that worth it for me?”

Now you’re making a real decision-not a marketing one.

What Experts Say

Dr. Steve Woloshin and Dr. Lisa Schwartz from Dartmouth have spent decades studying this. They say: “Absolute numbers matter.” If you only hear relative risk, you’re getting half the story. A 50% reduction sounds powerful. But if the absolute risk was 0.1%, you’re talking about 0.05% improvement. That’s not a treatment. That’s noise.

The FDA now recommends including both absolute and relative risks in patient materials. But enforcement is weak. Most patient leaflets still bury the absolute numbers.

One study found 60% of doctors can’t correctly convert relative risk to absolute terms. If they’re confused, how are patients supposed to understand?

Visual Aids Help-A Lot

Try this: imagine 100 little people. Each is you. Now color them based on risk.

Before the drug: 4 out of 100 are red (heart attack risk). After the drug: 3 out of 100 are red. One person changed color. That’s it.

Now imagine the same 100 people with a side effect: 5 are yellow (nausea). After the drug: 7 are yellow. Two more turned yellow.

That’s the whole story. No percentages. No ratios. Just 100 people. You see the trade-off. That’s why pictograms and risk ladders work better than charts. Your brain gets it.

When Relative Risk Is Actually Useful

Relative risk isn’t useless. It’s great for spotting patterns. If smokers have 20 times the risk of lung cancer, that’s a powerful signal. It helps researchers find causes. But for you? For your decision? Absolute risk is what matters.

Think of it like weather forecasts. “There’s a 20% chance of rain” is absolute. “It’s twice as likely to rain today as yesterday” is relative. Which one tells you whether to bring an umbrella? The 20%.

What You Should Do Next

Next time a doctor or ad talks about risk reduction:

• Don’t accept “cut in half” or “reduced by 70%” as the full answer.
• Ask: “What’s my starting risk?”
• Ask: “How many people need to take this for one to benefit?”
• Ask: “What’s the absolute increase in side effects?”

Write it down. Compare it. Don’t let numbers scare you or sell you. Understand them.

Medication decisions aren’t about percentages. They’re about people. You’re one person. You deserve to know what that one person’s real chance is-not what the math looks like on a billboard.