Making sense of relative risk
How do you make sense of a recent study that found that a cholesterol-lowering strategy can reduce your risk of heart attack by 20%? What actual benefit can you expect?
Get ready for some math. Stay with me.
- Say that a study compares 100 people taking a drug to lower cholesterol with a group taking a placebo, a drug or intervention that has no therapeutic effect.
- In the drug group, four people, or 4%, have a heart attack over the course of the study compared with eight people in the placebo group.
- To find out the risk difference between the two groups, simply, subtract the two percentages: 8% minus 4% equals 4%. That makes the absolute risk reduction 4%.
- But if we wanted to find out the probability of a heart attack between the two groups we would use a formula called relative risk. All we need to do to find the relative risk is to turn the outcome into a fraction: 4/8. Because 4 is half of 8 we can then say the drug treatment may reduce the risk of heart attack by 50%.
Why report relative risk instead of absolute risk?
Many studies present research findings in terms of the change in relative risk rather than absolute risk. They do this because it makes the findings seem more substantial.
Let’s take for example the widely-publicized analysis in The Lancet that looked at data from 13 studies. The analysis found that 4.9% of people taking statins, a cholesterol lowering drug, developed diabetes compared to 4.5% who did not take a statin.
- In absolute terms, that is just a 0.4% increase in the risk of developing diabetes while taking a statin, clearly a very small risk.
- But if we divide the 4.5% with the 4.9%, we get a 9% greater chance of getting diabetes while on a statin.
Relative risk can be used to “dramatize” the benefits or risks of a therapy. But it also helps us make sense of the numbers. It draws attention to the fact that small differences in large populations can be very significant in the real world.
An important note
Having a risk factor doesn’t mean you are doomed to develop that disease.
Having high low-density lipoprotein (LDL “bad”) cholesterol doesn’t mean you will have a heart attack. It only means that your chances of developing coronary artery disease are greater than for someone who doesn’t have that risk factor, if all other things—age, sex, weight, and blood pressure, for instance—are equal.
Likewise, taking steps to reduce that risk lowers your risk. But it does not eliminate it.
Calculating individual risk for heart attack and stroke
Individual risk is a completely different issue that is determined by the collection of risk factors for heart disease that you possess. It’s always best to avoid thinking that the results of any one study has a direct effect on your individual risk of developing heart disease (or anything else).
The Framingham Risk Calculator is a tool that calculates a person’s 10-year risk of a heart attack.
You can use the risk calculator if you are between the ages of 45 to 75 and already don’t have the following:
- Heart Disease
- Very high level of LDL cholesterol (190 mg/dL or higher)
If you are considered at high risk for heart attack or stroke you are automatically a candidate for a statin drug and don’t need to calculate your risk.
The risk calculator takes into account the following factors:
- Total cholesterol level
- High-density lipoprotein (HDL “good”) cholesterol level
- Systolic blood pressure (the top number)
- Presence of diabetes
- Whether you smoke
- Whether you are treated for high blood pressure
The heart risk calculator is available at this website.
If you exceed the risk threshold
If your estimated 10-year risk is 7.5% or higher, the guidelines recommend that you talk with your doctor about any lifestyle factors (smoking, exercise habits, diet, body weight) that may increase your risk. You should also discuss the benefit—and any potential risks—of taking a statin drug.
Because this latest assessment tool gives added weight to age, it’s especially important to weigh the pros and cons of beginning drug therapy for primary prevention if you are over 60—especially if your LDL cholesterol level is favorable (below 130 mg/dL).
The risk calculator is better calibrated for people who are middle-aged, but it can also underestimate risk. That is because the calculator doesn’t take into account a family history of premature heart disease or stroke.
If your 10-year risk is 5% to 7.5% and there is uncertainty about whether to initiate statin therapy, the guidelines suggest that you and your doctor consider several additional factors:
- Family history (if your father had a heart attack or stroke before age 50 or your mother before age 60, that may mean you are at an additional risk)
- LDL cholesterol level over 160 mg/dL
- High levels of C-reactive protein (a marker for inflammation linked to coronary risk)
- Peripheral artery disease (indicated by an ankle brachial index test that compares blood pressure in the ankle and the brachial artery in the upper arm)
- High score from a coronary calcium scan
The coronary calcium scan detects buildup of calcium on artery walls. Calcium buildup is a marker for cardiovascular risk. Many experts consider it the best tiebreaker in making a decision about starting on a statin. It may be especially useful when age is the main factor that raises your risk score.
For some patients, additional factors may be considered:
- Elevated triglycerides
- Low HDL levels
- Other blood markers of heart disease risk should also be considered
People 75 and older
Make sure to take into account your:
- Specific risk profile
- Preferences about drug therapy
- Other medications you are taking
- Risk of side effects (which are increased in older people)
You and your doctor should discuss statin therapy on an individual basis.