What is cancer?

Understanding statistics in cancer research

On this page is brief information about making sense of statistics in research papers. There is information about

Finding research papers

Many people with cancer look for information about treatments for themselves. Many people use the internet to try to find out about new treatments. But it is important to understand that a single research paper won't give you the whole picture about research. You need to read in the context of all the other relevant research for that particular type of cancer.

When a new treatment finishes all the research stages, it will probably have a large number of published papers about it. Some of these papers will show that its a useful new treatment that might help to slow cancer growth or cure it. But there might be some papers that show that it didn't work better than the current treatment.

The results that seem to contradict the other research papers may have happened by chance. Or there may have been problems with the group of people who took part. Or there may have been difficulties with giving the treatment.

What doctors and researchers are looking for is evidence that, on balance, the treatment is better than the existing treatment. Sometimes gather together all the results of all the trials and do a . This compares the results of all the relevant trials to give a broader picture. It gives a clearer idea than a single research paper of whether a treatment is helpful or not.

Unfortunately, some trials never produce results. This might be because the research team couldn’t find enough people to take part. Or because the trial was stopped early for some reason. This means they don’t have enough information to get reliable results.

Understanding statistical tests

During research studies, scientists use statistical tests such as 'T tests' and 'chi squared' tests. These tests compare the results of the different treatments used in the studies. They give an idea of whether there really was a difference in how well the treatments worked or whether the difference happened by chance.

To show how confident the researchers are that the results did not happen by chance, they use confidence intervals. For example, a 95% confidence means that the researchers are pretty sure that the result has not happened by chance.

You might find reading the results of research helpful if you understand statistics. If not, don't worry. You can read the discussion at the end of the paper you are looking at to get an idea of what the study found. The discussion usually explains the claims in the research paper more clearly.

You may want to ask your specialist doctor or nurse to explain the research results to you. You can also look on our clinical trials database, for plain English summaries of trial results.

Questions to ask when reading research reports

There are a few important points to think about when looking at research papers.

Here are a few questions you might ask:

What type of cancer is being investigated? If it is not the type you have, then the treatment is unlikely to be helpful to you
What stage of cancer do the people have? Is it a treatment for early stage or ?
Is the treatment used in combination with another treatment?
How many people took part in the study? The bigger the numbers, the more likely the results are accurate. The fewer there are, the more likely the results happened by chance. The most accurate studies use thousands of people, often in many different countries
What was the aim of the trial? Phase 3 clinical trials usually compare a new treatment with the current . Earlier trials look more at what happens to the drug inside the body and what the side effects are
Was it a controlled trial? This means comparing the new treatment with the standard treatment. People are allocated to one or other treatment at random to prevent
How much did the new treatment help? Sometimes a new treatment seems really promising when you read the reports. But when you look in more detail, people only lived a few weeks longer
Did the trial look at the people's ? For a treatment to be useful, the benefits need to outweigh any inconvenience or side effects. If the side effects are too severe, your quality of life could be worse than it would be without the treatment. This is important if the treatment is only going to slow down or shrink the cancer for a while
Who carried out the research? Is it a reputable, well known organisation? The quality of research can vary a great deal
Where was the study reported? Was it in a well known and respected journal? Check with your doctor or nurse if you aren't sure

It is fine to look for information about a new treatment if you know the answer to these questions and decide it is still for you. But it is important to understand clearly what you are likely to gain. The questions above are also important when looking at alternative treatments. These are treatments used instead of conventional cancer treatments. Many research studies used to promote alternative treatments are case studies. They are the story of one or 2 people who have had the new treatment and there is no comparison with other people. So, we don't know whether the treatment worked or whether people would have stayed well anyway. And some case studies may not publish the cases where the treatment didn't work.