The latest projections from the National Cancer Registry show that the number of new cancer cases being diagnosed each year in Ireland is expected to double by 2040. To understand such a dramatic rise we need to look a bit more deeply at how the figures are calculated. Three elements determine the projection of trends in cancer incidence: population change, cancer trends, and changes in exposure to risk factors.
An ageing population is the main factor driving an increase in cancer numbers. This is because the rate of ageing is much bigger than any changes in the rate of cancer.
The NCR projections use models of population change provided by the Central Statistics Office. These models make a number of assumptions about mortality, migration and fertility. Only the first of these is likely to have any major impact on future cancer numbers, as the others affect the younger population, where cancer cases are low. With the ageing of the population, and improved life expectancy, the median age of cancer patients at diagnosis will increase.
The projections indicate that almost 50% of people with invasive cancers will be aged over 70. As a result there will not only be more cancer patients, but they will be older on average.
The projections are also based on the assumption that current trends in cancer incidence will continue. For each cancer, we examined the incidence rates from 1994 to 2010 to find the current trend.
Non-melanoma skin cancer rates, for example, have been increasing since 2001. The projections show that this increase will continue in the future, and skin cancers will in fact be the most rapidly increasing cancers in the future. The primary risk factor for non-melanoma cancer of the skin is UV exposure, and evidence suggests that this steady increase is mainly due to recreational UV exposure, such as sunbeds.
But it doesn’t always follow that what is happening now will predict what will happen in the future. The rapid increase in prostate cancer cases in the 1990s and early 2000s, for example, was due to the widespread use of Prostate Specific Antigen (PSA) testing. This is a blood test which measures the level of the PSA protein in the bloodstream, and was introduced as a way to help with the diagnosis of prostate cancer. But this trend began to level off around 2004, when PSA testing was found to be less useful that previously thought, because many people with raised PSA levels do not have prostate cancer and some prostate cancers identified through testing are very slow growing and may not require treatment. Because of the large variations in trend for prostate cancer in the past two decades, the projections aren’t regarded as very reliable.
Another example of something that can affect cancer rates and therefore projected trends is screening. The introduction of screening for breast and cervical cancers led to a considerable increase in the number of cases that were diagnosed. Screening for colorectal cancer recently begun in Ireland and novel methods of screening for other cancers will probably appear in the next decade. While some screening may eventually bring about a reduction in health service costs, screening will initially increase case numbers and costs above what might be expected from natural increase.
Four risk factors have been shown to determine the majority of the attributable risk of cancer (excluding non-melanoma skin cancer) in the UK: tobacco, diet (including food energy balance, obesity and physical activity), alcohol and reproductive factors. Three of these can be considered lifestyle-related. There are no equivalent calculations for Ireland, but given the similarity in lifestyle between the populations of Ireland and the UK, it is reasonable to assume that the distribution and prevalence of attributable risks due to these major factors are similar.
The link between exposure to these risk factors and cancer incidence can perhaps be most clearly seen with tobacco smoking. Smoking has been implicated as a causal factor in many cancers, and is the cause of the overwhelming majority of lung cancer cases. Smoking prevalence in Ireland is high, although decreasing slowly, and more rapidly in males than females. Overall there has been a fall of about 9% in smoking prevalence for males and of about 8% for females since 1986.
The estimated increase in tobacco-related cancers in Ireland and between 2010 and 2040 is 110-115% in females and 83-91% in males. However, given the gradual decrease in female smoking rates in the 1990s, there is likely to be an eventual levelling-off in the number of female lung cancer cases. Men have much higher smoking rates overall, so although it is falling for both genders, it will still take longer for male rates to catch up.
What all this shows is that projections of cancer cases based on existing trends are limited in their ability to predict the future cancer burden. However, the overall conclusions from these projections are clear. Although we cannot modify the effects of demographic change, the majority of attributable cancer risk is due to a small number of well understood and potentially modifiable behaviours: UV exposure, smoking, alcohol consumption, diet and exercise.