Look for controls
The idea of a “control” is fundamental to research, and ensures that the results of scientific experiments are meaningful.
In our fictional sausage study of 5,000 people with cancer, a good control group would be 5,000 people of similar ages who don’t have cancer.
And it would help if we knew more about their lifestyles as well as their diet. For example, we need to know if they smoked, drank alcohol, were overweight or took lots of exercise, as all of these things can also affect the risk of cancer.
There’s more about this kind of study on our Healthy Living pages, as well as a discussion of how to make sense of mixed messages about cancer risks.
Control groups are also very important in clinical trials of new cancer treatments, to work out whether they are more effective than current therapies.
If seven out of ten people taking a new drug survive cancer for five years, this is only meaningful if we know what happens to people who don’t take the drug. There’s more about clinical trials on our website.
In the lab, scientists use controls to check that their experiments are working properly and are free of contamination. For example, if they are testing whether a chemical dissolved in water can kill cancer cells, it’s important to also test the water alone.
Read our other top tips: