In medical research, there are a number of ways to talk about risk of a disease, event, or outcome. One of these ways is to compare the risk of an event occurring in one group with the risk of the same event occurring in a different group.
Let’s say you’re comparing two groups of men: those whose fathers had colon cancer and those whose fathers did not. The risk of contracting the disease is greater in the first group because colon cancer is hereditary.
This type of assessment is called relative risk (often abbreviated as RR). Relative risk helps you understand how likely a disease is to occur between two groups with different demographic characteristics or different exposures.
What is relative risk?
Relative risk is a helpful way of thinking about the risk of a disease outcome in one group compared with or relative to another. This type of analysis is often used in epidemiological studies of diseases having a strong association with certain environmental exposures.
In these cases, the groups may vary by their exposure to a risk factor like the carcinogen asbestos, which is still found in many older buildings, with groups that have not been exposed. Relative risk may also be used to understand the different disease risk that exists between demographic groups, such as those defined by biological sex, ethnic background, or genetic makeup.
Relative risk is a type of statistical analysis used to assess data in various types of studies, including:
How to calculate and interpret relative risk
Relative risk is also known as the risk ratio or rate ratio. That’s because it’s the ratio of the probability of a particular event happening in one group to the probability of the same event happening in a different group.
Often, these groups are defined by exposures, such as cigarette smoke, alcohol, or asbestos. The comparison is conducted between an exposed group and an unexposed group.
A relative risk of 1 means that there’s no risk difference between the two groups. A relative risk of more than 1 means there’s a higher risk of the disease or event occurring in that group, and a relative risk of less than 1 means the risk is lower.
Here’s a real-world example. In the preprint “Epidemiological analysis of Lung cancer in Erbil province of Iraqi Kurdistan: Incidence, Survival, Relative Risk Ratio, and Treatment Regimes in males and females,” the authors use a table to show the relative risk of lung cancer between females and males as a function of age.
Here’s how they calculate relative risk:
Probability of lung cancer in females = number of females in age group / all females in study
Probability of lung cancer in males = number of males in age group / all males in study
Relative risk = (Probability of lung cancer in females) / (Probability of lung cancer in males)
In the 20-year-old age group, for example, relative risk is calculated as:
- Probability of lung cancer in females = 13 / 336
- Probability of lung cancer in males = 12 / 254
- Relative risk = (13 / 336) ÷ (12 / 254) = 0.82
This means that females in the 20-year-old age group within this study population are 0.82 times as likely to have lung cancer as males in this age group. Or, in other words, females in this age group have an 18% lower risk of having lung cancer compared to males in this age group (1 − 0.82 = 0.18, or 18%).
Overall, the table suggests that females are at lower risk of developing lung cancer than males up to age 40. After age 40, the trend reverses and females begin to show higher risk for lung cancer than males.
Other types of risk assessment
Relative risk is sometimes confused with other measures used to quantify risk, such as the absolute risk or odds ratio.
Relative risk is a measurement of risk of an event in one group in comparison with another distinct group. Absolute risk, however, refers simply to the likelihood the event will occur within the total at-risk population.
Absolute risk is calculated by dividing the number of people who actually experience the event by the total number of people exposed to the risk in question.
For example, if 1 in 10 individuals exposed to cigarette smoke develops lung cancer, then the absolute risk of a cigarette smoker developing lung cancer is 1/10 or 10%.
Likewise, if 1 in 100 non-smokers develops lung cancer, we can say the absolute risk of a non-smoker developing lung cancer is 1/100 or 1%. In this case, the relative risk of lung cancer in smokers vs. non-smokers can be calculated as:
Relative Risk = (1/10) ÷ (1/100) = 10
This means that although both groups have their own absolute risks, a smoker is 10 times more likely to develop lung cancer than a non-smoker.
The difference between the risk ratio and odds ratio is more subtle.
- “Risk” refers to the likelihood of an event occurring relative to all other outcomes.
- “Odds” refers to the ratio of the likelihood of an event occurring to the likelihood of that event not occurring.
The odds ratio, then, is the ratio of odds of an event in one group to the odds of the event in another group. You can find a deeper explanation of risk difference, risk ratio, and odds ratio by Dr. Chittaranjan Andrade at the University of Pittsburgh.
As with relative risk, odds ratios of 1 indicate no difference in the chances of an outcome between groups. Odds ratios of greater or less than 1 mean more or less likelihood, respectively.
The odds ratio and relative risk can be used somewhat interchangeably when the outcome is rare (<10%), but an odds ratio will tend to exaggerate the likelihood of an event when that event is more common (>10%). In such cases, relative risk tends to give a more accurate representation of risk.