Making health decisions often comes down to a calculation of risk. It is the measurement of the likelihood of a negative outcome, like a disease or a fracture.

Risk is expressed in two ways: absolute risk and relative risk. You need to understand what each describes and how they are different to interpret information about it. That's especially true when parsing claims from pharmaceutical companies about the ability of their drugs to reduce risk.

Today you'll gain an essential skill for cutting through the misleading statistics used by Big Pharma to get the truth about the efficacy of drugs and other health interventions.

All About Risk And How It’s Assessed

Risk is the likelihood of a negative outcome. The health choices we make, whether eating a particular diet, doing an exercise regimen, or taking a drug, typically aim to reduce the risk of certain undesirable health outcomes.

To decide which choices to make, we often compare how much each choice reduces risk. We call that change, “risk reduction”.¹

The most headline-grabbing result of a study is often its risk reduction. This is most often calculated by comparing the outcomes of two different groups of participants in a study.

Usually, risk reduction compares one group that tried an intervention and one that did not, or two groups that performed a particular behavior (i.e. walking, eating vegetables, etc.) at different frequencies or for different durations.

Examining risk and the potential for risk reduction gives us the ability to improve our choices and our health. However, some ways of calculating risk can be deceptive, so it's important to understand the different ways that risk is calculated and expressed.

Synopsis

By measuring the risk of a particular negative outcome, and how it changes based on our choices, we can learn how to reduce it.

Absolute Risk and Number Needed To Treat

Let's say that 100 people tried to walk the length of a balance beam. We'll use their likelihood of falling to demonstrate the types of risk and risk reduction.

If 6 of those 100 people fell while walking the beam, their absolute risk of falling was 6%. Anytime we're looking at the likelihood of a single outcome, that risk is an Absolute Risk (AR).

In our imaginary study, another 100 people walked across the beam, but these people carried a balancing pole. Only one of those people fell off the beam, so that group's Absolute Risk of falling was only 1%.

The balancing pole reduced the risk of falling. To calculate the Absolute Risk Reduction (ARR) achieved by using the balancing pole, we simply take the difference between those two numbers. 6% minus 1% equals 5%, so the ARR is 5%.

That means for every hundred people that cross the balance beam using a balancing pole, five falls will be prevented.

Since the goal here is preventing falls from the balance beam, the question arises of how many balancing poles have to be used to successfully prevent a single fall. That figure is called the Number Needed to Treat, or NNT.¹

The NNT describes how many participants need to receive the intervention for one of them to achieve the desired outcome. The NNT is calculated by dividing 100 by the Absolute Risk Reduction. 100 divided by 5 equals 20, so you'd have to give 20 people crossing the balancing beam a balancing pole to prevent one fall.

These figures– Absolute Risk Reduction and Number Needed to Treat– offer the most accurate and clear assessment of an intervention's efficacy. Those are the numbers to use when deciding whether a particular treatment or intervention is worth the effort, expense, and potential side effects.

Synopsis

Absolute Risk Reduction (ARR) is the difference between the risk of a particular outcome with no intervention and the risk with an intervention. The number of people who need to receive an intervention before someone avoids the outcome is called the number needed to treat. ARR is the most honest and accurate measurement of risk reduction.

Relative Risk And Relative Risk Reduction

Relative Risk (RR), as the name implies, always describes the relationship between two different outcomes.

We calculate the Relative Risk by dividing the Absolute Risk in the intervention group (balancing pole) by the Absolute Risk of the control group (no balancing pole). One divided by 6 equals approximately 0.17.

A relative risk below 1 shows a reduction in risk, above 1 shows an increase in risk, and exactly 1 shows no change in risk.

Relative Risk Reduction (RRR) describes the amount of risk reduction in relation to the starting risk. We find the RRR by dividing the change in Absolute Risk by the Absolute Risk in the intervention group.

So to find the RRR achieved by using the balancing pole, we divide the ARR of using the pole (5%) by the AR when not using a balancing pole (6%). The resulting number is .83, which when expressed as a percentage is an 83% Relative Risk Reduction

An 83% RRR sounds much larger than a 5% ARR, but in fact, they both describe the same results.

Let's look again at the figures from our example:

• Control group absolute risk: 6%
• Intervention group absolute risk: 1%
• Absolute Risk Reduction: 5%
• Number Needed to Treat: 20
• Relative Risk: .17
• Relative Risk Reduction: 83%

Without an understanding of where these numbers came from, we may be assuming that an intervention is reducing risk far more than it actually is.

Synopsis

Relative Risk Reduction does not accurately represent the reduction of risk. Instead of measuring the relationship between the starting and ending risk, it measures the relationship between the starting risk and the change in risk. This figure is often quite large, even when the Absolute Risk Reduction is very small.

Big Pharma Uses Relative Risk To Overstate Efficacy

Big Pharma often uses Relative Risk Reduction to overstate the efficacy of their drugs. Instead of advertising the reduction in Absolute Risk– which is often quite modest– they promote the Relative Risk Reduction– which gives the impression of a larger benefit.

Let's take for example the osteoporosis drug Evenity (Romosozumab). According to drug maker Amgen's studies on fracture risk in women taking Evenity, the drug has an ARR of 1.3% and an RRR 73%.²

Amgen chose to trumpet the 73% RRR, making it sound as though the drug reduced participants' chance of a spinal fracture by 73 percentage points. In fact, the ARR was only 1.3%, dropping from 1.8% to .5%.²

In addition to being deceptive, these outsized figures distract from fundamental issues with osteoporosis drugs, such as their grisly side effects.

Evenity's side effects can include an increased risk of death from cardiac events, osteonecrosis of the jaw, and atypical femur fractures, just to name a few. A 1.3% decrease in absolute risk of a spinal fracture isn't worth increasing the risk of these other life-altering outcomes.

Synopsis

Big Pharma company Amgen advertises its osteoporosis drug Evenity using its Relative Risk Reduction of 73%, but its Absolute Risk Reduction (the more useful and honest figure) is only 1.3%. This is deceptive, and the large figure distracts from the increased risk of side effects like osteonecrosis of the jaw, atypical femur fractures, or dangerous cardiac events.

What This Means To You

Pay close attention to claims of miraculous results and massive reductions in risk– all too often they prove to be manipulations of statistical data. Big Pharma companies are especially guilty of this.

Studies can be complex and confusing, but pursuing a more independent path to stronger bones requires learning about the science behind the choices we make. The Save Institute created the Osteoporosis Reversal Program so that Savers could learn everything they need to know to improve their bone health. It's easy to read and easy to follow, thanks to a digestible course format broken down into sections and lessons.

Take your fight for strong bones and a healthy and independent future to the next level. You have what it takes and you're not alone!

References

¹ https://www.ncbi.nlm.nih.gov/books/NBK63647/

² https://www.ema.europa.eu/en/documents/assessment-report/evenity-epar-public-assessment-report_en.pdf