Cancer Research UK on Google+ Cancer Research UK on Facebook Cancer Research UK on Twitter

Looking at statistics in research

Nurse and patients talking about cancer   

This page is a brief guide to making sense of statistics in research papers. There is information about

 

Finding and choosing research papers

More and more people with cancer are looking for information about treatments for themselves. Many people use the internet to try to find out about new treatments. But it is important to think about what you read. Cancer research is going on all over the world. A single research paper in isolation won't give you the whole picture about research into that particular type of cancer. It needs to be read in the context of all the other relevant research.

Each time a new treatment makes it through all the stages of research and clinical trials, it will have a large number of published research papers about it. Some of these papers will show it to be a useful new treatment that may contribute to slowing down, or curing a cancer. But there are likely to be one or two papers that show that it didn't work better than the existing treatment it was compared to. These results that seem opposite to the others may have happened by chance. Or there may have been problems with the patient group that was selected. 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 an improvement. Sometimes statisticians gather together all the results of all the trials and do a meta analysis. This is a process that compares the results of all the relevant trials to give a broader picture.

The point that we are trying to make is that reading any one of these papers is unlikely to give you the broader picture. You probably can't rely on research results that come from just one study.

 

Understanding statistical tests

During research studies, statisticians and researchers use statistical tests such as T tests and chi squared tests. These tests mathematically compare the results of the different treatments used in the studies. The tests give an idea of whether there really was a difference in how well the treatments worked or whether the difference could have happened by chance. 

To show how confident the researchers are that the results did not happen by chance, they use confidence intervals. For example, if a result is declared with 95% confidence, then it means the researchers are pretty sure that the result is OK and has not happened by chance. 

If you understand statistics you may find reading the results of research helpful. 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. This part 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. 

 

Questions to ask when reading research reports

There are a few important points you may like to bear in mind 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 advanced cancer?
  • Is the treatment being studied used in combination with another treatment?
  • How many patients were involved 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 compare a new treatment with the current standard treatment. Earlier phase trials look more at things such as what happens to the drug inside the body and what the side effects are. You can read about phases of trials
  • Was it a controlled trial? This means the treatment was compared with standard treatment (which might be no treatment). Patients were allocated one or other treatment at random to prevent bias. For example, a treatment may seem better because the doctors used it for all the healthier patients and gave the other treatment to all the patients who were more ill
  • 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 peoples' quality of life? 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 made worse than it would be without the treatment. This is especially 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 are not sure

It is fine to look for information about a new treatment if you know the answer to all 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 two patients who have been given the treatment and there is no comparison with other patients. So, in a case study we don't know whether the treatment worked or whether the patients would have stayed well anyway.

 

More information about research trials

Find detailed information about clinical trials and what it means to take part in our section about trials and research.

You can also find details about UK clinical trials on our clinical trials database.

Rate this page:
Submit rating

 

Rated 4 out of 5 based on 8 votes
Rate this page
Rate this page for no comments box
Please enter feedback to continue submitting
Send feedback
Question about cancer? Contact our information nurse team

No Error

Updated: 21 November 2014