Incidence, survival and mortality

Incidence means how many people get a particular type of cancer every year. It is often written as number of cancer cases per 100,000 people in the general population. These calculations are done for every type of cancer. 

For example, in the UK in 2008 there were 

• 47,693 women diagnosed with breast cancer - about 153 cases per 100,000 women
• 22,846 men diagnosed with lung cancer - about 76 cases per 100,000 men
• 2,138 men diagnosed with testicular cancer - about 7 cases per 100,000 men

Non melanoma skin cancer is the most commonly diagnosed cancer overall, with at least than 98,000 new cases registered each year. Nearly all these people are cured. Non melanoma skin cancer is often excluded from cancer incidence statistics because it is usually much less serious than other types of cancer.

Other ways of looking at incidence include

• Incidence in different age groups
• Incidence in different countries

Overall, prostate cancer is the most common cancer in men -  24 out of every 100 male cancers (24%) are prostate cancer. But other cancers are more common in younger age groups. For men aged 25-49, testicular cancer is the most common (15%). Prostate cancer only accounts for 3% of male cancers in this age group.

If you look at cancers diagnosed in children (under 15 years old) then leukaemia is the most common cancer. 1 in 3 (33%) of cancers in this age group are leukaemia.

Breast cancer is the most common in women - 31 out of every 100 cancers (31%) diagnosed in women are breast cancer. But in women aged 15-24, melanoma is the most common accounting for 17% of cancers, with only a few women diagnosed with breast cancer in this age group.

Although there are variations for specific cancers, the number of cancers diagnosed generally increases with age. Nearly two thirds of cancers (63%) are diagnosed in people aged 65 and over. 

This is interesting because it could shed some light on the causes of different cancers. Breast cancer incidence is often quoted as an example of this. The incidence of breast cancer in Japan is much lower than in the USA or the UK. But the incidence of breast cancer in Japanese women living in the USA is the same as for the general population of American women. To doctors, this implies that the causes of breast cancer are probably more closely related to life style and the environment than they are to our genetic make up.

Survival statistics for cancer are usually written as '5 year survival' or '10 year survival'. These statistics probably cause more confusion than any others.

What 5 year survival means is that X% of patients were alive 5 years after they were diagnosed. It does not mean that these people lived for exactly 5 years and then died. It doesn't mean they were all cured either. Some of them will be cured. Some will have already had a recurrence of their cancer, but still be alive. Some will get a recurrence after the 5 year period.

5 year and 10 year time periods are used largely for convenience. Research studies often follow people up for 5 or 10 years. So figures for 5 and 10 years are often available to be quoted.

For some types of cancer the chances of the cancer coming back after 5 years is quite small and 5 year survival is sometimes used to mean 'cure' when actually it does not. For almost all types of cancer, the chances of recurrence after 5 years are much lower than they are after 2 years. So the more time passes, the less likely it is that your cancer will come back.

Everyone who has that particular type of cancer unless the statistics say otherwise. You may come across 5 year survival figures by stage of cancer. That may be simplified into 'local disease' (cancer that has stayed in the area where is was diagnosed) and 'metastatic disease' (cancer that has spread to another part of the body). For breast cancer, the figures are sometimes divided up by whether the cancer has spread to the lymph nodes or not.

Straight forward 5 year survival figures are very general. They include everyone with that type of cancer, at all stages and grades. If you have an early stage cancer, then the percentage quoted as surviving will be too low. If you have an advanced cancer, then the figures may be a little optimistic. So be wary of taking this type of statistic too literally. The figures quoted are true. But they may not apply to you.

Sometimes disease free survival figures are used. This does mean everyone with that type of cancer who is alive and well (without a recurrence of their cancer) 5 years after diagnosis.

Mortality statistics mean the number of people who have died from a particular type of cancer in a year. On their own, these figures don't mean much. They have to be looked at alongside incidence figures and other statistics.

As with incidence statistics, you have to think about what mortality figures actually mean. They don't take account of stage, grade or any specific sub type of a cancer. They are used as a general guide to what is happening in diagnosis and treatment of diseases.

Remember - statistics are general. You may read that 47,693 women were diagnosed with breast cancer in the UK in 2008 and 12,047 women died from it that same year. Don't assume that these are the same women. Almost all of the 12,047 who died would have been diagnosed some years before. In a lot of cases, many years before.

Changes in mortality figures over time are difficult to interpret. The incidence of a cancer may be falling, so less people die from it. Or treatment may be improving so more people are cured. Treatment improvement may not mean cure, but it may mean more people are living longer after they are diagnosed so this will also make the mortality figures fall in the short term.

The last type of statistics to cover are those about risk of getting any particular type of cancer. This is often written as 'life time risk'. Using the available figures, statisticians work out the risk of any one of us getting a certain type of cancer at some point during our lives.

This is sometimes written as a percentage. And sometimes as 1 in X. For example, the calculated life time risk of lung cancer for a man in the UK is 1 in 14. This means that out of every 14 men in the UK, one will get lung cancer at some point in his life (and 13 won't). 

To change this to a percentage, you divide it into 100. 100 divided by 14 equals 7. So the percentage life time risk of lung cancer for a man in the UK is 7%. But this is the overall risk. It will be higher or lower, depending on whether an individual smokes or not. And because it is 'lifetime risk', it is a cumulative risk. This means that the risk adds up as you get older. So the risk for an average 45 year old is not 1 in 14 - it will be much lower.

Again this doesn't mean a lot for any one person. These figures are produced to help people who study the incidence of disease (epidemiologists) and people who work out how much money we need to spend on health care (health economists) do their jobs. They are important because they tell us the most important health problems of our time. And from that, we can work out the areas where we need to concentrate spending on research, screening, treatment and training.