The Risk Paradox: How Relative and Absolute Risk Shape the Stories We Tell in Medicine

In medicine, numbers don’t just inform — they shape narratives, drive policy, and influence public perception. Among the most commonly used tools in clinical research are relative risk (RR) and absolute risk (AR). Both are statistically valid, but depending on which one gets emphasized, the story can sound very different.

Let’s explore this paradox by comparing three influential studies — on statins and cholesterol, COVID-19 vaccines and cardiovascular risks, and herpes zoster vaccination and dementia. Each tells a compelling story, but not always in the same statistical language.

📉 Case 1: Statins, LDL Cholesterol, and the Relative Risk Mirage

In their critical 2022 paper, Diamond and Leaverton questioned the way relative risk reduction was used to promote statins and dietary interventions aimed at lowering LDL cholesterol [1]. While statin trials often report 20–30% RR reductions in myocardial infarction (MI), the absolute risk reduction in many low-risk populations is less than 1–2% over five years.

RR sounds impressive — a 30% risk reduction!AR reveals the nuance — it might mean going from a 3% to a 2% risk of heart attack.

The authors argue that such framing can mislead both clinicians and patients, inflating the perceived benefit of preventive therapies and supporting mass-medication strategies that may not be necessary for all.

💉 Case 2: COVID-19 mRNA Vaccines and Cardiovascular Events

A 2025 meta-analysis by Karimi et al. assessed the potential link between mRNA COVID-19 vaccines (Pfizer-BioNTech and Moderna) and cardiovascular events like coronary artery disease (CAD), myocardial infarction (MI), and stroke [2].

• The relative risk (OR) of CAD after the second dose of Pfizer's vaccine was reported as 3.44 — a more than threefold increase.

• But the absolute risk remained very low — events were rare, occurring in fewer than 1 in 10,000 people.

  
  

Here, public messaging flipped the script: RR was downplayed, and AR was emphasized to reassure the public of vaccine safety.

This is statistically fair — after all, absolute risk is what most individuals care about. But it’s a stark contrast to the way RR is typically celebrated when showcasing benefits.

🧠 Case 3: Herpes Zoster Vaccine and Dementia Prevention

Then comes the 2025 Nature study by Eyting et al., which made waves by showing that herpes zoster vaccination is associated with a ~25% relative reduction in dementia risk [3]. The finding was hailed as a breakthrough in dementia prevention — and rightfully so, given the lack of other interventions with even modest protective effects.

But what about absolute risk? The AR reduction was just 0.8% over seven years — small, but meaningful at a population level.

Unlike the COVID-19 vaccine debate, RR was front and center, used to justify broader use of the zoster vaccine in older adults.

🎯 Same Tools, Different Stories

Let’s compare them side by side:

Study Context	Relative Risk (RR)	Absolute Risk (AR)	Narrative Framing
Statins & LDL	RR ↓ MI by ~25%	AR ↓ often <2%	RR criticized as overstating benefit
COVID-19 Vaccines	RR ↑ CAD (OR = 3.44)	AR ↑ very slightly	AR emphasized to downplay RR
Zoster Vaccine & Dementia	RR ↓ dementia by ~25%	AR ↓ by 0.8% over 7 years	RR emphasized as a breakthrough

🧠 What’s the Takeaway?

The scientific validity of both RR and AR is not in question — but their interpretation and communication can be highly selective.

• RR is excellent for detecting patterns and making headlines.

• AR is essential for understanding clinical significance at the individual level.

• But when RR is used to promote benefits and AR is used to dismiss harms, we risk introducing bias into public health narratives.

🔍 Final Thought

In a data-driven world, how we frame risk matters as much as the numbers themselves. The next time you hear about a risk “doubling” or being “cut in half,” ask: Compared to what? And by how much, really?

Clarity, transparency, and consistency in risk communication are more than statistical virtues — they are public health imperatives.

📚 References

1. Diamond DM, Leaverton PE. Historical Review of the Use of Relative Risk Statistics in the Portrayal of the Purported Hazards of High LDL Cholesterol and the Benefits of Lipid-Lowering Therapy. Expert Review of Clinical Pharmacology. 2022.

2. Karimi R, Norozirad M, Esmaeili F, Mansourian M, Marateb HR. COVID‑19 Vaccination and Cardiovascular Events: A Systematic Review and Bayesian Multivariate Meta‑Analysis. Int J Prev Med. 2025;16:14.

3. Eyting M, et al. A Natural Experiment on the Effect of Herpes Zoster Vaccination on Dementia. Nature. 2025