We updated to 2015 our forecasts and analyses of the effects of variation components on cancer incidence trends since 1981 in Piedmont and its Local Health Authorities (LHA).
Trends are represented through histograms showing the following different components: changes in risk, in the age structure of populations, or in their absolute dimension .
Incidence forecasts by LHA were obtained by applying the mortality/incidence ratio method with correction for birth-cohort, thus accounting for risk patterns, as observed in incidence data, and trends in survival, by their effects on mortality [2,3].
Estimates account for population ageing by modelling the effect of age and birth cohort (five-year classes, upper category 90 or more years). Mortality from uterine cancers was corrected by accounting for the often important proportion of cases with unspecified (uterine) site; the specific cause of death was assigned based on results of the cohort of incident cases in Torino. Forecasts to 2015 are based on population demography predictions (central prediction) by the Istituto nazionale di statistica (ISTAT), with area correction for sub-province level populations.
The statistical methods already adopted for our previous 2012 estimates were now applied to the new (up to 2010) incidence data for Torino residents and to the regional mortality data up to 2010. The reference year was set to 1981 as it was a census year. Our analyses, spanning over thirty years, show how changes in time are attributable to distinct components:
- the “true” risk of developing cancer at any particular age;
- population ageing, that it the growth of the proportion of individuals belonging to the age classes with the highest incidence of neoplasms;
- the overall dimension of the population and its variations.
The graphs represent as cumulative bars such components, and display the resulting cancer burden and its variation over time between 1981 and 2015.
The components effect on cancer incidence was calculated by applying a model which enabled us to discriminate:
- differences among age-standardized incidence rates (representing true changes in risk);
- the proportion of residents by age class (which contributes to differences due to variation in the age structure of the population, in Piedmont generally represented by ageing);
- the overall dimension of the population (which determines the overall number of cases).
The net difference across components is given by the difference between crude rates.
Based on our data, interpretation of the role of population dimension is straightforward.
During the observation period, Piedmont and most of its LHAs had a basically stable population, with slight if any growth in absolute dimension. A substantial immigration flux occurred but it involved mainly young individuals, with little relevance for the present occurrence of malignancies.
Some displacements were more important, mainly relocation of families and persons from Torino to its province or emigration from the Verbano Cusio Ossola area towards other Italian regions; the effects of such circumstances can be identified in various graphs and will not be commented further.
Lastly, it should be remarked that population ageing causes the highest risk sub-group of the population to grow, by increasing the absolute number of elderlies.
An example of guided reading of a histogram is provided to ease data interpretation.
- C16 Stomach
- C18-21 Colon-rectum
- C25 Pancreas
- C03-06, C10-15, C30-32 Upper airways digestive tract
- C33-34 Lung
- C50 Breast
- C53 Cervix uterine
- C54 Corpus uteri
- C56 Ovary
- C61 Prostate
- C71-72 Brain
- C81, C82-85, C96 Lymphomas
- C91-92 Leukemia
- All tumors
 Bashir S. Estève J. Analysing the difference due to risk and demographic factors for incidence or mortality. Int J Epidemiol. 2000; 29: 878-84
 Ferlay J, Shin HR, Bray F, Forman D, Mathers C and Parkin DM. GLOBOCAN 2008, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 10. Lyon, France: International Agency for Research on Cancer; 2010.
Available at: http://globocan.iarc.fr
 Colonna M, Grosclaude P, Faivre J, Revzani A, Arveux P, Chaplain G, Tretarre B, Launoy G, Lesec'h JM, Raverdy N, Schaffer P, Buémi A, Ménégoz F, Black RJ. Cancer registry data based estimation of regional cancer incidence: application to breast and colorectal cancer in French administrative regions. J Epidemiol Community Health. 1999; 53: 558-64