Are statistics reliable? Many people do not trust them. They think that statistics can be made to ‘say’ whatever anyone wants. So really, it is statisticians that they don’t trust!
To some extent, it is true that you can manipulate statistics to fit what you want to say. But only if you are deliberately trying to change the figures to suit yourself. For example, if you are trying to sell a product, you will pick the statistics that show your product to be the best thing since sliced bread! You won’t print the statistics that said half the people who had bought it thought it was rubbish!
Researchers and doctors are not trying to fiddle the figures. They are genuinely interested in finding answers to the questions they need to ask. For example, which treatment works best?
But the answers you get from statistics still depend on the questions you ask. Will the answers to these questions be the same?
- Which treatment works best?
- Which treatment is most cost effective?
They won’t necessarily be the same answer. The more cost effective treatment may put more cancers in remission per 100 spent just because it is cheaper. It is important to carefully read the question being asked when you are looking at statistics of any sort.
What can statistics tell you?
Understanding statistics can be very important if you are trying to find out about new ‘conventional’ treatments or looking into various alternative therapies. Statistics can tell you whether one treatment is likely to work better than another for a particular type of cancer.
Understanding statistics a little better can help you to understand more about what your doctor tells you too. There are many different statistics collected on cancer and treatments, but you need to know what they really mean if they are to be of any interest to you.
What statistics can’t tell you
Statistics can’t tell you what will happen to you. They are general pieces of information that apply to a ‘population’ of people from which the statistics were calculated. This could be tens, hundreds or thousands of people. Statistics may be able to tell you what the chances of something happening to you are. If 65% of people with the same type of cancer as you responded to a treatment, then there is a two out of three chance that you will too. But no one can say definitely one way or the other.
Statistics cannot fit your situation exactly. They are usually much more general than people think they are. A patient may ask her doctor, “What are the survival statistics for someone with a grade 3, stage 2 breast cancer who has had a lumpectomy, 6 weeks of radiotherapy, 6 months of CMF chemotherapy and has been on “chemotherapy X” for 2 years?” Statistics in that sort of detail just don’t exist.
Individual research studies will have produced statistics on bits of that scenario. For example, researchers have shown that pre-menopausal women who have a course of adjuvant chemotherapy have a lower chance of their cancer coming back.
To produce the sort of detail the patient above asked for, you would have to find several hundred women in that exact situation to test the chemotherapy on, and then follow their progress for some years, to see how they did. And even then, you would only get generalized statistics, because they would probably have several different types of breast cancer. And they may well have had radiotherapy in different ways, or at different doses. And are they estrogen receptor positive? Oh, and how old are they – pre or post menopausal? As you can see, no situation is as straight forward as it might appear!
You and your situation are unique. As there is no one else quite like you, then no statistics will be able to give you cast iron answers to the questions you want to ask about your cancer.