Abstract

A comparative analysis of cancer prevalence in France, Spain and Italy is presented as part of the EUROPREVAL project. The three countries are culturally and sociologically relatively homogeneous compared with Europe as a whole. However, in all three countries, the cancer registries (CRs) providing the data for prevalence calculation cover only small fractions of the populations, and have been operating for relatively short periods. This leads to problems of representativity and to prevalence underestimates as surviving cases diagnosed before operation of the CR are not recorded.

Partial prevalences obtained directly from CR data were therefore corrected using a completeness index obtained by modelling to provide estimates of the complete prevalence. For CRs operating for only 5 years, only approximately half the prevalence was observed. Thus, due to the rather recent start of most of southern European CRs, the role of correction is very important.

The prevalence of all cancers was highest in Italy for women and in France for men, while lowest in Spain. Differences in the age structures of the populations were the major cause of these discrepancies and after age adjustment only the prevalence of stomach cancer remained highest in Italy, although differences in incidence also contributed to the prevalence differences. Survival varied little between the three countries and differences in incidence are more important determinants of prevalence. Prevalence of cancer in the elderly represents an increasing load for the community, particularly for France, Italy and Spain due to the ageing population in these countries.

Elderly patients with cancer frequently suffer from problems of co-morbidity and disability factors, thus placing a burden on the local medical system where this proportion is high. Prevalent cases diagnosed 1–5 years before the prevalence date formed approximately one-third of the total prevalence, with higher proportions for melanoma, and prostate cancer in males and breast and colorectal cancer in females, and lower proportions for uterine cancer. This subset of the prevalent population consists of those probably on intensive follow-up, or being treated for cancer recurrence or sequelae to primary therapy.

Received 18 June 2001; revised 17 December 2001; accepted 9 January 2002

Introduction

France, Italy and Spain present certain characteristics and difficulties in common with regard to the estimation of cancer prevalence from cancer registry (CR) data, and the interpretation of those estimates.

First, in none of these countries does cancer registration cover the whole population, and in those participating in the EUROCARE/EUROPREVAL projects it was below 10% in 1992 [1]. In France, the situation was particularly unfavourable as cancer registration covered only 3% of the population at that time, and this included specialised CRs that collect data on certain types of cancer only. Because of this, particular caution is required when extrapolating prevalence data to the whole country.

Secondly, although some CRs were established in these southern European countries at the end of the 1970s, ∼20 years later than CRs in several northern European countries, the great majority of CRs in France, Italy and Spain were set up in the mid-1980s. The implication of this is that the majority of surviving cancer patients in these populations were diagnosed before the start of registration and hence cannot be included in estimates of cancer prevalence calculated directly from CR data. Extensive statistical modelling is therefore required to estimate the numbers of these prevalent but unregistered cases, with consequently greater uncertainty in total prevalence figures compared with those obtained from CRs operating for 20–30 years, where the great majority of prevalence cases are registered.

France, Italy and Spain are reasonably homogeneous in terms of lifestyle, diet, cancer incidence patterns and cancer survival patterns; they are additionally characterised by rapidly increasing life expectancy at birth and expanding health care systems. In these respects they constitute a distinct subgroup within which cancer prevalence generally varies less widely than among other European countries [2].

A limited amount of cancer prevalence data has been published previously for Italy and France [38]; to the best of our knowledge nothing has been published on cancer prevalence in Spain. The recent systematic comparison of prevalence in the areas covered by Italian CRs used the same methodologies as in the present study [3]. The aim of this paper is to provide a unified presentation of cancer prevalence in France, Italy and Spain, whose data are directly comparable with those of other papers on prevalence in Europe authored by the EUROPREVAL study group. As in all these analyses, we considered the following 11 cancer sites/categories: stomach, colon, rectum, breast, uterine, prostate and lung (including pleurae), melanoma, Hodgkin’s lymphoma, leukaemias, and all malignant cancers combined.

Methods and data sources

Sixteen CRs from France, Italy and Spain provided incidence and survival data allowing direct calculation of cancer prevalence. Table 1 shows the populations, periods of diagnosis, and total and category-specific numbers of cases contributed, for each participating registry grouped by country. All registries assured follow-up for life status until 31 December 1992, which was the index date for the prevalence estimate. The numbers of cases shown in Table 1 include patients who did not survive until the reference date and therefore were not included in the prevalence calculation. Table 2 shows the age distributions of the populations covered by the participating CRs. The populations covered by the Italian CRs were older than those in France and Spain, being characterised by both the lowest proportion under 45 years and the highest proportion 65 years and over. The French population was the youngest.

A comprehensive account of the methods used to calculate the prevalence, including the problems to be overcome, corrections applied and choices adopted, is given by Capocaccia et al. [9]. We briefly summarise the methods, noting particular difficulties pertaining to the CR data from the three countries of this study.

The basic procedure used to estimate the prevalence of a given cancer is to allocate all the incident cases recorded by the CR, and alive at the index date, to cells within a two-dimensional matrix defined by age at the index date and the number of years passed since diagnosis (duration). The numbers thus provided are divided by the corresponding population size to give the cancer site-, age- and duration-specific prevalence as a proportion of the population at the index date. This simple procedure would give the complete prevalence if all the incident cases alive at the prevalence date were recorded by the registry. In practice, CR data were incomplete for several reasons, and the observed prevalence calculated by this direct method had to be corrected to allow for the various sources of incompleteness.

Some cases had data missing (e.g. sex, diagnosis date, microscopic diagnosis), although on average not more than 1% of all records in each CR were incomplete. For these records imputation procedures were used to provide as complete a record as possible. For cases lost to follow-up, life status at the index date was uncertain. Such cases were dealt with by assuming their survival to be the same as that of the subset of patients of the same age and diagnosis period who were not lost to follow-up. The number of lost cases thus estimated alive at the prevalence date was added to the number of prevalent cases counted directly.

Cases known from death certificate only or discovered at autopsy, for which the date of diagnosis was unknown, were not considered in the present prevalence calculations. Such cases are a potential source of prevalence underestimation, which in most cases is likely to be small. This problem is discussed by Capocaccia et al. [9].

Definitions of prevalence differ in relation to how multiple primary cancers are considered. We present data on person prevalence. With person prevalence, only the first primary malignant cancer diagnosed in each person is considered; it is a measure of the number of people with a diagnosis of cancer. The tumour prevalence considers all the primary malignancies diagnosed and present in a population; it is a count of the number of tumours irrespective of whether they are the first or subsequent cancers diagnosed in a given individual.

As noted, prevalence calculated from the data of CRs operating for a short period is necessarily an underestimate. This was the case for all CRs considered in the present study and the partial prevalence data they provided had to be corrected by applying a completeness index, R, that depends on the length of registration (L), age class, and other factors affecting incidence and survival such as cancer site and sex.

The method we used to calculate R [10] has been extensively validated on data from the Connecticut Tumor Registry, which has been operating since 1935 [11]. R was defined as follows: R = NO(m)/NT(m), where NO(m) and NT(m) are model-based estimates of observed and total prevalence, respectively. These estimates were derived from mathematical expressions relating prevalence to incidence and survival. The prevalence measured by a registry, NO, was divided by the R obtained as above, to furnish an estimate of the total prevalence, NT = NO/R.

Overall incidence and survival data for the three countries were used to estimate a single set of parameters from which Rs were calculated. However, because R is highly sensitive to relative survival, a death ratio, specific for each registry, was used to increase the precision of estimates of R.

Table 3 shows calculated Rs by cancer site and registry, and for men and women, all ages combined, together with the time in years that each CR has been operating. R values may be between 1, when all prevalent cases are observed (i.e. the CR has been operating for a very long time) and theoretically 0, when no prevalent cases are observed by the CR. R values also vary within a registry according to the type of cancer. They are higher for cancers that mainly affect the elderly, and lower for cancers characterised by long survival. As is evident from Table 3, the observation periods (L) for the CRs considered in this study were all relatively short (from 5 to 15 years) so the corresponding Rs were in general considerably much lower than 1 and around 0.5 for CRs operating only for 5 years, the latter figure implying that only half of the prevalent cases had been registered by the CR and that the remainder had to be estimated by modelling. For the Varese CR with the longest observation period in southern Europe (15 years) the completeness of prevalence estimates ranged from 72% for corpus uteri cancer to 97% for prostate cancer. For the Mallorca CR with one of the shortest observation periods (5 years) the completeness of estimates ranged from 32% for corpus uteri to 66% for prostate cancer.

Validation of Rs

We considered CRs operating for at least 15 years (Calvados, Parma and Varese). For these registries the observed 15-year prevalence (cases diagnosed 1978–1992) and the observed 5-year prevalence (cases diagnosed 1988–1992) were determined. The empirical ratios of 5-year to 15-year prevalence were then compared with the corresponding theoretical values obtained by modelling. Figure 1 shows the empirical and calculated 5-year/15-year indices according to sex, as a function of age for selected cancer sites only. The calculated values were generally very close to observed ones and it is therefore reasonable to conclude that the use of R provides reliable 15-year prevalence estimates for CRs with observation periods as short as 5 years. However, for cancers in patients under 40 years of age, for which the observed ratios were unstable due to the small number of cases, the discrepancy between empirical and calculated ratios was considerable, particularly for leukaemia and cancer of the colon. The match between theoretical and empirical completeness ratios was also good for those cancer sites not illustrated in Figure 1.

Results

Registry-specific prevalence data were grouped by country. The resulting figures (per 100 000 of the population) are shown by sex and age class in Table 4. The prevalence of all cancers was highest in Italy for women and in France for men. Prevalence was lowest in the populations covered by the Spanish registries, and was particularly low for women in the oldest age class.

Although the incidence of all cancers combined was higher in men than in women [12], prevalence was higher for women due to the fact that survival is generally better in women [1]. Note, however, that in the oldest segment of the population, prevalence was greater in men than in women.

For all ages combined, breast cancer was the leading cancer among women, with the highest prevalence in Italy (1070 per 100 000 women). The leading malignancy in men was prostate cancer with the highest prevalence in France. The prevalence proportions of colon and rectal cancer were highest in Italy and lowest in Spain, for both men and women. The prevalence of male lung cancer was quite high in all three countries (187, 165 and 135 per 100 000 in Italy, France and Spain, respectively), and the most important cause of intercountry variation in prevalence for this cancer was incidence, since survival is poor.

Table 5 shows prevalence according to time since diagnosis. For all cancers combined, the 1-year prevalence (persons diagnosed in the year up to the index date) represented 17% of the male total and 10% of the female total, with no major variation between the three countries. The 1-year prevalence as a percentage of the total was highest for lung cancer (26–33%), which has a poor prognosis, and low for uterine cancer and Hodgkin’s disease (<10%). The 1- to 5-year prevalence formed one-quarter to one-third of the total prevalence in the three countries, with slightly higher levels for breast, colon and rectal cancers in females, and levels well above one-third for melanoma and prostate cancer in males; values tended to be slightly lower for cancers of the uterus. For all cancers combined the 5- to 10-year prevalence constituted 18.3–23.5% of the total prevalence in both sexes. Again for all cancers combined, the >10-year prevalence was 25% in men and 40% in women; however, the proportion of long-term survivors varied markedly with cancer site. In spite of the declining incidence trend in recent decades, the long-term prevalence of stomach cancer was also relatively high due in part to higher incidence in the past.

Cancer prevalence by cancer site in the 65+ and 45–64 years of age classes is shown in Table 6 for both sexes combined. People aged 65 years or more accounted for nearly 58% of the total prevalence in Italy, 55% in France and 54% in Spain. The proportion was much higher for cancers of the digestive tract, corpus uteri and prostate. More than 90% of prevalence cases of prostate cancer were in the over 65 years of age class in all three countries.

Discussion

The assumptions and methodologies for calculating prevalence in the CR populations in the three subject countries of this study were the same as those used for the other European populations involved in EUROPREVAL [9]. However, since southern European CRs have shorter observation periods than most CRs in Europe, and cover only small proportions of the populations of the countries in which they operate, the reliability of our estimates must be assessed critically. Comparison of empirical with theoretical partial R ratios (Figure 1) indicated that R values were, in general, consistent and that they do not constitute a major source of error for our prevalence estimates, even for CRs operating only for 5 years.

With regard to the problem of extent of coverage, comparisons with previously published prevalence data may help to place the present estimates in context. The ITAPREVAL study [3] reported cancer prevalence in 1992 in 10 Italian CR areas. The present results are based on nine Italian CRs, with the Veneto CR left out. The Veneto CR covers a large area in north-eastern Italy. Overall prevalence estimates calculated with and without Veneto CR data were satisfactorily similar for all cancers combined, and for breast cancer, colon cancer, rectal cancer and leukaemia. The prevalences of these cancers in Veneto were similar to those in other Italian registry areas. For cancers of the stomach and lung the prevalence estimated in this study was ∼20% higher than in the ITAPREVAL study, which is directly attributable to the absence of data for Veneto, an area with lower prevalence for these cancers. For melanoma and cancer of the corpus uteri, prevalence was rather high in the Veneto CR, and thus the overall prevalence for the Italian CRs found in this study was about 20% lower than the overall figure provided by ITAPREVAL. As noted, for cervical cancer and Hodgkin’s disease the Rs were inaccurate and were not used for these cancers instead of total prevalence estimates, and the 15-year prevalence, which proved to be more stable, is given Table 4.

With regard to France, meaningful comparison of our results with published national estimates obtained by modelling [7, 8] is impossible since, in the modelled estimates, patients over 79 years of age were ignored, and the youngest age considered varied: 0 years for haematopoietic cancers in children, 20 years for most cancer sites, and 40 years for prostate cancer. No cancer prevalence estimates are available for Spain. We ex-pect that inaccuracies in overall prevalence estimates for France and Spain will be similar to or greater than that for Italy, since coverage in those countries is lower.

The prevalence of all cancers combined was highest in Italy for women and in France for men, and lowest in Spain for both sexes. The main determinants of cancer prevalence are incidence and survival rates, and population structure; however, survival did not vary greatly between the three countries. The main factor determining prevalence differences between France and Italy was the difference in age structure of the two populations (Table 4); cancer incidence rates for these two countries were similar and survival was only slightly higher in France. When the prevalence for all cancers combined was adjusted to the age structure of the world population [2], it emerged that France had the highest prevalence. The main determinant of lower overall cancer prevalence in Spain was lower incidence rates [2].

For most cancer sites, crude (non-age-adjusted) prevalence was highest in Italy due to the fact that the population was older in Italy compared with the other two countries. By contrast, crude prevalence in Spain was generally the lowest because of the lower cancer incidence in that country [2]. For breast cancer the crude prevalence was higher in Italy than France and Spain, while after age standardisation France had the highest prevalence. This pattern occurred because breast cancer incidence was slightly higher in France and the population older in Italy.

The only cancer site for which age-standardised prevalence was lower in Italy was cervical cancer, due to the particularly low incidence in that country [2]. France had both the highest crude and age-adjusted prevalence of this cancer. Considering cervical cancer by age class, the prevalence was higher in Spain than in Italy in women up to 64 years of age, but much lower among Spanish women of 65 years of age and above. However, the age-standardised prevalence of this cancer was higher in Spain than in Italy. Cohort effects can explain the prevalence pattern of cervical cancer in Spain. Older Spanish women are monogamous and not promiscuous, and are thus characterised by low rates of cervical cancer, accounting for the low prevalence among them. Young cohorts have higher rates of cervical cancer, even though screening is in place.

While total prevalence refers to all living persons with a diagnosis of cancer, whose specific health needs vary markedly, prevalence according to time elapsed between diagnosis and reference date results in subgroups that are much more homogeneous with regard to their requirement for health resources. The distribution of prevalence by time since diagnosis (Table 5) varied markedly with cancer site, while between-country variation was small. The 1-year prevalence was highest for lung cancer, consistent with the fact that most people with this disease die within a year of diagnosis. For all cancers combined, about 17% of the total male prevalence and 10% of the total female prevalence consisted of people diagnosed within 1 year of the index date. This subgroup consisted mainly of patients undergoing primary treatment for their disease.

The 1- to 5-year prevalence formed one-quarter to one-third of the total prevalence in the three countries. Prevalent cases diagnosed 1–5 years before the index date consist of both cured cases and those in whom the disease is progressing. They are likely to be receiving either intensive follow-up to identify recurrences and sequelae, or palliative treatments for incurable disease. The 5- to 10-year prevalence formed 18.3–23.5% of the total prevalence for all cancers in both sexes. Some of these patients can be considered cured and in general they make modest demands on health service resources in relation to their cancer diagnosis. The final category of cases, those diagnosed >10 years before the index date, make the lowest demands on the health system for their cancer diagnosis. Some may, however, suffer disabilities and impairments arising from their treatment and may still make more demands on health resources than the age-matched general population. This long-term or ‘cured’ prevalence is an important indicator of the success of cancer treatment; it constituted approximately one-quarter of the total prevalence in men and 40% of the total in women. The proportion of long-term survivors, however, varied markedly with cancer site, being highest in cancers of the uterus and lowest for prostate cancer. In spite of the declining incidence trend in recent decades, long-term prevalence of stomach cancer was also reasonably high due to higher incidence in the past.

Age is known to have a major influence on disease stage at diagnosis, the accuracy of diagnosis and the effectiveness of treatment, and hence the prognosis. Cancer prevalence among the elderly is also increasing due to population ageing, particularly for France, Spain and Italy. In addition, elderly cancer patients more frequently have co-morbidities and disabilities, so they make increased demands on health service resources. For these reasons cancer prevalence among the elderly is an important indicator of the health demand.

Although there were clear differences between France, Spain and Italy in terms of cancer prevalence, these differences were less marked than those characterising Europe as a whole [2]. Thus, for colon cancer prevalence varied from 120 per 100 000 in Spain to 210 per 100 000 in Italy, and from 45 (Poland) to 240 (Sweden) for Europe as a whole; for breast cancer the range was 617 (Spain) to 1070 (Italy) for the southern European countries and 380 (Slovakia) to 1210 (Sweden) for Europe. The corresponding ranges for all cancers combined were 1860 (Spain) to 2600 (Italy), and 1170 (Poland) to 3050 (Sweden). Since survival did not differ greatly in the three countries examined in this study, it follows that cancer incidence and population structure were the main factors responsible for variations in cancer prevalence. By contrast, survival differences play an important role in determining differences in cancer prevalence throughout Europe as a whole. Thus the prominence of the 10-year-and-over prevalence in the southern European populations analysed in this paper is an indication of a successful struggle against cancer.

If increased prevalence is mainly due to rising incidence, it is a sign that we are losing the battle against cancer; if it is due to improved survival and a decline in mortality for competing causes, the message is definitely positive. What is always clear, however, is that increasing prevalence indicates increasing demand for health care, and the requirement that more resources be devoted to that care. As life expectancy increases in these countries, the prevalence of cancer is also expected to increase, and is likely to become a major economic burden. The only way to directly combat increasing prevalence is to prevent cancer. This implies that the campaign against smoking should be redoubled, that information on healthy eating should be actively promulgated and that research on the relationship between cancer and dietary and other environmental factors should be stepped up. Screening programmes for breast and cervical cancer should be extended as these have helped reduce mortality in the former and reduce the incidence of the latter. Although in Italy and Spain the healthy ‘Mediterranean diet’ (rich in fruit and vegetables, and apparently protective against cancer) is still widespread, and also characterises much of the southern part of France, the high prevalence of colorectal cancer in Italy and France indicates that even in these countries, diet needs to be considered as a means of reducing cancer incidence.

Acknowledgements

This study was supported by the EC-BIOMED-II programme (Contract No. BMH4 98 3899). The authors thank Don Ward for help with the English.

Members of the EUROPREVAL Working Group: Austria: W. Oberaigner (Cancer Registry of Tyrol). Denmark: H. Storm, G. Engholm (Danish Cancer Society ‘Institute of Cancer Epidemiology’). Estonia: T. Aareleid (Estonian Cancer Registry). Finland: T. Hakulinen (Finnish Cancer Registry). France: G. Hédelin, (Bas-Rhin Cancer Registry); H. Lefevre (Calvados Digestive Cancer Registry); J. Mace-Lesec’h (Calvados General Cancer Registry); J. Faivre (Côte d’Or Digestive Cancer Registry); G. Chaplain (Côte d’Or Gynaecologic Cancer Registry); P.M. Carli (Côte d’Or Malignant Haemopathies Registry); P. Arveux (Doubs Cancer Registry) J. Estève (University of Lyon); M. Colonna (Isère Cancer Registry); N. Raverdy, P. Jun (Somme Cancer Registry). Germany: J. Michaelis (German Registry of Childhood Malignancies); H. Ziegler, C. Stegmaier (Saarland Cancer Registry). Iceland: H. Tulinius (Icelandic Cancer Registry). Italy: R. Capocaccia (Project Leader); I. Corazziari, R. De Angelis, S. Francisci, S. Hartley, F. Valente, A. Verdecchia, A. Zappone (National Institute of Health, Rome); F. Berrino, G. Gatta, A. Micheli, E. Mugno, M. Sant (National Institute for the Study and Cure of Tumors, Milan); P. Crosignani (Lombardy Cancer Registry); E. Conti, V. Ramazzotti (Latina Cancer Registry); M. Vercelli, C. Casella, A. Puppo (Liguria Cancer Registry – NCI, Univ. Genova); M. Federico, (Modena Cancer Registry); M. Ponz De Leon (Modena Colorectal Cancer Registry); V. De Lisi, (Parma Cancer Registry); R. Zanetti, S. Rosso (Piedmont Cancer Registry); C. Magnani (Piedmont Childhood Cancer Registry); L. Gafà, R. Tumino (Ragusa Cancer Registry); F. Falcini (Romagna Cancer Registry); E. Paci, E. Crocetti (Tuscany Cancer Registry); S. Guzzinati, P. Zambon (Venetian Cancer Registry). Poland: J. Rachtan (Cracow Cancer Registry; M. Bielska-Lasota (Warsaw Cancer Registry. Slovakia: I. Plesko (National Cancer Registry of Slovakia). Slovenia: V. Pompe-Kirn (Cancer Registry of Slovenia). Spain: I. Izarzugaza (Basque Country Cancer Registry); A. Izquierdo (Girona Cancer Registry); I. Garau (Mallorca Cancer Registry); E. Ardanaz, C. Moreno (Navarra Cancer Registry); J. Galceran (Tarragona Cancer Registry); V. Moreno (Catalan Institute of Oncologia). Sweden: T. Möller, H. Anderson (Southern Swedish Regional Tumour Registry). Switzerland: J. Torhorst (Basel Cancer Registry); C. Bouchardy, J.M. Lutz, M. Usel (Geneva Cancer Registry); J.E. Dowd (W.H.O., Geneva). The Netherlands: J.W.W. Coebergh, M. Janssen-Heijnen (Eindhoven Cancer Registry); R.A.M. Damuhis (Rotterdam Cancer Registry). Scotland: R. Black, V. Harris, D. Stockton (Scottish Cancer Intelligence Unit). United Kingdom: T.W. Davies, (East Anglian Cancer Registry); M.P. Coleman, S. Harris (London School of Hygiene and Tropical Medicine); E.M.I. Williams (The Merseyside and Cheshire Cancer Registry); D. Forman, R. Iddenden (Northern and Yorkshire Cancer Registry and Information Service & Centre for Cancer Research); M.J. Quinn (Office for National Statistics); M. Roche (Oxford Cancer Intelligence Unit); J. Smith (South and West Cancer Intelligence Unit); H. Moller (Thames Cancer Registry); P. Silcocks (Trent Cancer Registry); G. Lawrence, K. Hemmings (West Midlands Cancer Intelligence Unit).

+

Correspondience to: Dr A. Verdecchia, Laboratorio di Epidemiologia e Biostatistica Istituto Superiore di Sanità, Viale Regina Elena, 299, 00161 Rome, Italy. Tel: +39-06-4990-2230; Fax: +39-06-4938-7069; E-mail: verdeck@iss.it

Figure 1. Comparison of observed and modelled 5-year/15-year indices for estimating prevalence in Calvados, Parma and Varese in 1992 for selected cancers.

Figure 1. Comparison of observed and modelled 5-year/15-year indices for estimating prevalence in Calvados, Parma and Varese in 1992 for selected cancers.

Table 1.

 Registry, population covered, period of diagnosis, number of cases by cancer type, and percentage lost to follow-up

Registry Population Period All cancers Lost (%) ICD9 151 ICD9 153 ICD9 154 ICD9 162 ICD9 172 ICD9 174 ICD9 180 ICD9 182 ICD9 185 Hodgkin’s disease Leukaemia 
France                
Somme   549 205 1982–92 20 360 10.0  874 1505 1079 2410 234 2503 491  450 1632 163  598 
Calvados   625 111 1978–92 – – 1407 1969 1563 – – – – – – – – 
Cote d’Ora   499 328 1978–92 – – 1244 2392 1646 – – 2430 378  448 – 159  966 
Italy                
Florence 1 182 239 1985–92 44 373  0.7 5149 4129 2389 5663 721 5511 501 1087 2249 319  992 
Genoa   678 696 1986–92 24 550  0.2 1430 2292 1126 3760 363 3342 355  474 1134 174  519 
Latina   479 421 1983–92 10 878  0.5  820  765  491 1811 156 1311 231  286  377 108  391 
Modenab   606 023 1988–92 14 566  1.1 1085 1468  852 2066 197 1812 149  342  734  83  321 
Parma   391 807 1978–92 27 262  0.3 3502 2249 1313 3633 330 3392 386  718 1177 211   676 
Ragusa   290 706 1981–92   8338  0.2  669  593  409  985 114 1116 212  306  406  86  291 
Romagna   426 438 1986–92 15 626  0.1 1984 1375  642 2199 265 1808 257  358  826  99  416 
Turin   955 827 1985–92 31 888  0.0 1874 2812 1494 4552 489 4542 553  784 1387 262  736 
Varese   798 797 1976–92 45 803  0.9 4167 3757 2062 6523 660 6811 665 1207 1843 430 1174 
Spain                
Basque. 2 097 400 1986–92 40 807  0.0 3412 2833 2097 4976 638 4711 567 1003 1839 375 1011 
Mallorcac   585 606 1982–92 10 426  1.5  384 1321  950 1294 145 1040 242  248  539  66  231 
Navarra   521 673 1985–92 12 522  0.0 1113  939  636 1205 178 1604 127  328  882 109  294 
Tarragona   553 309 1985–92 12 384  0.6  681 1080  656 1307 182 1537 253  389  732  89  349 
Registry Population Period All cancers Lost (%) ICD9 151 ICD9 153 ICD9 154 ICD9 162 ICD9 172 ICD9 174 ICD9 180 ICD9 182 ICD9 185 Hodgkin’s disease Leukaemia 
France                
Somme   549 205 1982–92 20 360 10.0  874 1505 1079 2410 234 2503 491  450 1632 163  598 
Calvados   625 111 1978–92 – – 1407 1969 1563 – – – – – – – – 
Cote d’Ora   499 328 1978–92 – – 1244 2392 1646 – – 2430 378  448 – 159  966 
Italy                
Florence 1 182 239 1985–92 44 373  0.7 5149 4129 2389 5663 721 5511 501 1087 2249 319  992 
Genoa   678 696 1986–92 24 550  0.2 1430 2292 1126 3760 363 3342 355  474 1134 174  519 
Latina   479 421 1983–92 10 878  0.5  820  765  491 1811 156 1311 231  286  377 108  391 
Modenab   606 023 1988–92 14 566  1.1 1085 1468  852 2066 197 1812 149  342  734  83  321 
Parma   391 807 1978–92 27 262  0.3 3502 2249 1313 3633 330 3392 386  718 1177 211   676 
Ragusa   290 706 1981–92   8338  0.2  669  593  409  985 114 1116 212  306  406  86  291 
Romagna   426 438 1986–92 15 626  0.1 1984 1375  642 2199 265 1808 257  358  826  99  416 
Turin   955 827 1985–92 31 888  0.0 1874 2812 1494 4552 489 4542 553  784 1387 262  736 
Varese   798 797 1976–92 45 803  0.9 4167 3757 2062 6523 660 6811 665 1207 1843 430 1174 
Spain                
Basque. 2 097 400 1986–92 40 807  0.0 3412 2833 2097 4976 638 4711 567 1003 1839 375 1011 
Mallorcac   585 606 1982–92 10 426  1.5  384 1321  950 1294 145 1040 242  248  539  66  231 
Navarra   521 673 1985–92 12 522  0.0 1113  939  636 1205 178 1604 127  328  882 109  294 
Tarragona   553 309 1985–92 12 384  0.6  681 1080  656 1307 182 1537 253  389  732  89  349 

a The area is covered by three specialised registries operating for different periods of time: digestive (ICD9 151, 153 and 154) (1978–92); haematological for leukaemia and lymphomas (1980–92); and gynaecological (ICD9 174, 180 and 182) (1982–92). Data for this area were obtained from these three registries.

bSpecialised colorectal cancer registries 1983–87, general cancer registries 1988–92.

cSpecialised colorectal cancer registries 1982–87, general cancer registries 1988–92.

ICD9, International classification of Diseases Version 9. –, Cancer sites not available for French specialized CR.

Table 2.

 Age distribution (as percentage of total population) in France, Italy and Spain

 Men (age group in years)  Women (age group in years) 
 0–44 45–64 65+  0–44 45–64 65+ 
France 69 20 11  64 20 16 
Italy 58 27 15  53 26 21 
Spain 63 24 13  58 24 18 
 Men (age group in years)  Women (age group in years) 
 0–44 45–64 65+  0–44 45–64 65+ 
France 69 20 11  64 20 16 
Italy 58 27 15  53 26 21 
Spain 63 24 13  58 24 18 
Table 3.

 Cancer-specific completeness indices for all ages combined by cancer registry. The operating period (L) of each registry is also given

Registry L (years) Stomach ICD9 151 Colon ICD9 153 Rectum ICD9 154 Lung ICD9 162 Melanoma ICD9 172 Prostate ICD9 185 Leukaemia ICD9 204–8  
(A) Men          
France          
Somme 11 0.64 0.80 0.80 0.77 0.84 0.91 0.87  
Calvados 15 0.74 0.89 0.89 – – – –  
Cote d’Or 11 0.65 0.80 0.80 – – – 0.86  
Italy          
Florence  8 0.56 0.69 0.69 0.73 0.70 0.82 0.71  
Genoa  7 0.52 0.65 0.65 0.70 0.65 0.78 0.70  
Latina 10 0.65 0.77 0.77 0.77 0.80 0.90 0.80  
Modena  5 0.44 0.54 0.54 0.64 0.50 0.67 0.59  
Parma 15 0.75 0.89 0.88 0.86 0.91 0.96 0.91  
Ragusa 12 0.68 0.83 0.83 0.82 0.83 0.92 0.87  
Romagna  7 0.52 0.65 0.65 0.70 0.63 0.77 0.71  
Turin  8 0.57 0.70 0.70 0.72 0.70 0.82 0.74  
Varese 15 0.76 0.89 0.89 0.86 0.91 0.97 0.91  
Spain          
Basque C.  7 0.54 0.66 0.66 0.67 0.66 0.78 0.71  
Mallorca  5 0.46 0.80a 0.80a 0.60 0.52 0.66 0.59  
Navarra  8 0.58 0.69 0.71 0.70 0.69 0.82 0.77  
Tarragona  8 0.57 0.70 0.70 0.70 0.70 0.82 0.76  
Registry L (years) Stomach ICD9 151 Colon ICD9 153 Rectum ICD9 154 Lung ICD9 162 Melanoma ICD9 172 Breast ICD9 174 Corpus uteri ICD9 182 Leukaemia ICD9 204–8 
(B) Women          
France          
Somme 11 0.69 0.77 0.74 0.82 0.65 0.77 0.56 0.88 
Calvados 15 0.79 0.88 0.84 – – – – – 
Cote d’Or 11 0.69 0.77 0.74 – – 0.76 0.55 0.88 
Italy          
Florence  8 0.61 0.66 0.62 0.79 0.53 0.63 0.47 0.75 
Genoa  7 0.57 0.61 0.58 0.76 0.49 0.57 0.39 0.70 
Latina 10 0.70 0.75 0.72 0.84 0.64 0.73 0.58 0.82 
Modena  5 0.49 0.50 0.48 0.70 0.36 0.45 0.33 0.60 
Parma 15 0.79 0.87 0.83 0.92 0.78 0.87 0.73 0.92 
Ragusa 12 0.74 0.81 0.76 0.88 0.70 0.79 0.64 0.87 
Romagna  7 0.57 0.61 0.59 0.76 0.49 0.57 0.42 0.72 
Turin  8 0.62 0.67 0.63 0.79 0.52 0.63 0.47 0.76 
Varese 15 0.80 0.88 0.83 0.92 0.78 0.87 0.72 0.92 
Spain          
Basque C.  7 0.58 0.62 0.59 0.74 0.50 0.59 0.45 0.73 
Mallorca  5 0.50 0.78a 0.74a 0.66 0.37 0.46 0.32 0.59 
Navarra  8 0.61 0.66 0.63 0.76 0.54 0.64 0.48 0.77 
Tarragona  8 0.63 0.67 0.63 0.76 0.54 0.63 0.49 0.79 
Registry L (years) Stomach ICD9 151 Colon ICD9 153 Rectum ICD9 154 Lung ICD9 162 Melanoma ICD9 172 Prostate ICD9 185 Leukaemia ICD9 204–8  
(A) Men          
France          
Somme 11 0.64 0.80 0.80 0.77 0.84 0.91 0.87  
Calvados 15 0.74 0.89 0.89 – – – –  
Cote d’Or 11 0.65 0.80 0.80 – – – 0.86  
Italy          
Florence  8 0.56 0.69 0.69 0.73 0.70 0.82 0.71  
Genoa  7 0.52 0.65 0.65 0.70 0.65 0.78 0.70  
Latina 10 0.65 0.77 0.77 0.77 0.80 0.90 0.80  
Modena  5 0.44 0.54 0.54 0.64 0.50 0.67 0.59  
Parma 15 0.75 0.89 0.88 0.86 0.91 0.96 0.91  
Ragusa 12 0.68 0.83 0.83 0.82 0.83 0.92 0.87  
Romagna  7 0.52 0.65 0.65 0.70 0.63 0.77 0.71  
Turin  8 0.57 0.70 0.70 0.72 0.70 0.82 0.74  
Varese 15 0.76 0.89 0.89 0.86 0.91 0.97 0.91  
Spain          
Basque C.  7 0.54 0.66 0.66 0.67 0.66 0.78 0.71  
Mallorca  5 0.46 0.80a 0.80a 0.60 0.52 0.66 0.59  
Navarra  8 0.58 0.69 0.71 0.70 0.69 0.82 0.77  
Tarragona  8 0.57 0.70 0.70 0.70 0.70 0.82 0.76  
Registry L (years) Stomach ICD9 151 Colon ICD9 153 Rectum ICD9 154 Lung ICD9 162 Melanoma ICD9 172 Breast ICD9 174 Corpus uteri ICD9 182 Leukaemia ICD9 204–8 
(B) Women          
France          
Somme 11 0.69 0.77 0.74 0.82 0.65 0.77 0.56 0.88 
Calvados 15 0.79 0.88 0.84 – – – – – 
Cote d’Or 11 0.69 0.77 0.74 – – 0.76 0.55 0.88 
Italy          
Florence  8 0.61 0.66 0.62 0.79 0.53 0.63 0.47 0.75 
Genoa  7 0.57 0.61 0.58 0.76 0.49 0.57 0.39 0.70 
Latina 10 0.70 0.75 0.72 0.84 0.64 0.73 0.58 0.82 
Modena  5 0.49 0.50 0.48 0.70 0.36 0.45 0.33 0.60 
Parma 15 0.79 0.87 0.83 0.92 0.78 0.87 0.73 0.92 
Ragusa 12 0.74 0.81 0.76 0.88 0.70 0.79 0.64 0.87 
Romagna  7 0.57 0.61 0.59 0.76 0.49 0.57 0.42 0.72 
Turin  8 0.62 0.67 0.63 0.79 0.52 0.63 0.47 0.76 
Varese 15 0.80 0.88 0.83 0.92 0.78 0.87 0.72 0.92 
Spain          
Basque C.  7 0.58 0.62 0.59 0.74 0.50 0.59 0.45 0.73 
Mallorca  5 0.50 0.78a 0.74a 0.66 0.37 0.46 0.32 0.59 
Navarra  8 0.61 0.66 0.63 0.76 0.54 0.64 0.48 0.77 
Tarragona  8 0.63 0.67 0.63 0.76 0.54 0.63 0.49 0.79 

aL = 11 years.

Table 4.

 Total prevalence per 100 000 by country, age class (years), sex and cancer site

 France (age group in years)  Italy (age group in years)  Spain (age group in years) 
 0–44  45–64 65+ All ages  0–44  45–64  65+  All ages  0–44  45–64  65+  All ages 
Men               
 Stomach   2.1   59.6    487.8   67.7    6.9  184.1  738.7  162.8    8.2  183.0  543.8  112.9 
 Colon   5.2  188.3   1128.0  166.7    7.7  225.4  956.1  206.5    8.4  146.8  720.3  126.0 
 Rectum   6.1  144.1    788.1  120.6    3.6  141.3  572.7  124.8    5.1  130.6  497.9   93.3 
 Lung   9.8  528.7   1233.2  252.5    5.2  280.4  737.7  187.3    8.6  236.6  667.2  140.4 
 Melanoma   9.9   43.2     66.6   23.0   14.4   80.1  104.4   45.3    8.2   46.7  108.0   29.1 
 Prostate   0.5  106.9   2324.7  289.4    0.1   74.4 1274.4  209.1    0.1   63.1 1189.7  157.9 
 Hodgkin’s diseasea  24.0   35.5     16.4   25.4   31.3   39.3   23.2   32.2   26.3   41.9   21.2   29.3 
 Leukaemia  22.9   65.3    285.4   61.6   20.6   45.0  125.1   42.6   19.6   43.1  136.6   39.1 
 All cancers 284.2 3597.1 11 202.9 2192.2  309.4 2444.1 8513.3 2096.5  300.2 2508.9 8407.4 1784.8 
Women               
 Stomach   0.6   28.3    194.1   37.0    5.4   83.2  432.9  114.5    4.2   58.8 264.2   59.8 
 Colon   6.4  175.5    760.5  160.3    8.9  209.4  754.5  216.4    9.5  128.9  500.6  118.2 
 Rectum   4.4  105.6    495.0  102.8    4.4  116.3  368.5  109.4    2.8   84.3  289.6   69.0 
 Lung   1.2   38.7     69.5   19.5    2.6   42.2   99.8   33.2    1.4   19.4   31.2   10.4 
 Melanoma  12.1  117.9    170.3   58.4   24.3  133.6  156.3   80.6   22.2   92.9  160.6   61.3 
 Breast 106.2 1835.0   2654.4  857.8  107.9 1705.5 2714.3 1070.7   82.3 1163.8 1824.2  617.3 
 Cervixa  26.6  236.4    282.1  109.1   24.7  178.7  220.0  105.9   25.5  194.4  161.7   86.5 
 Corpus uteri   1.6  196.7    844.8  176.4    7.1  271.4  865.8  254.9    4.7  271.7  708.4  182.3 
 Hodgkin’s diseasea  21.8   26.7      2.9   19.7   35.6   24.3   19.3   29.2   20.1   17.6   13.1   18.4 
 Leukaemia  15.0   62.6    176.9   50.6   17.8   32.1   80.0   34.5   11.9   26.9   74.0   25.6 
 All cancers 382.8 4214.3   8926.1 2512.9  492.9 4053.2 8349.0 3062.4  400.8 2978.4 6127.1 1937.8 
 France (age group in years)  Italy (age group in years)  Spain (age group in years) 
 0–44  45–64 65+ All ages  0–44  45–64  65+  All ages  0–44  45–64  65+  All ages 
Men               
 Stomach   2.1   59.6    487.8   67.7    6.9  184.1  738.7  162.8    8.2  183.0  543.8  112.9 
 Colon   5.2  188.3   1128.0  166.7    7.7  225.4  956.1  206.5    8.4  146.8  720.3  126.0 
 Rectum   6.1  144.1    788.1  120.6    3.6  141.3  572.7  124.8    5.1  130.6  497.9   93.3 
 Lung   9.8  528.7   1233.2  252.5    5.2  280.4  737.7  187.3    8.6  236.6  667.2  140.4 
 Melanoma   9.9   43.2     66.6   23.0   14.4   80.1  104.4   45.3    8.2   46.7  108.0   29.1 
 Prostate   0.5  106.9   2324.7  289.4    0.1   74.4 1274.4  209.1    0.1   63.1 1189.7  157.9 
 Hodgkin’s diseasea  24.0   35.5     16.4   25.4   31.3   39.3   23.2   32.2   26.3   41.9   21.2   29.3 
 Leukaemia  22.9   65.3    285.4   61.6   20.6   45.0  125.1   42.6   19.6   43.1  136.6   39.1 
 All cancers 284.2 3597.1 11 202.9 2192.2  309.4 2444.1 8513.3 2096.5  300.2 2508.9 8407.4 1784.8 
Women               
 Stomach   0.6   28.3    194.1   37.0    5.4   83.2  432.9  114.5    4.2   58.8 264.2   59.8 
 Colon   6.4  175.5    760.5  160.3    8.9  209.4  754.5  216.4    9.5  128.9  500.6  118.2 
 Rectum   4.4  105.6    495.0  102.8    4.4  116.3  368.5  109.4    2.8   84.3  289.6   69.0 
 Lung   1.2   38.7     69.5   19.5    2.6   42.2   99.8   33.2    1.4   19.4   31.2   10.4 
 Melanoma  12.1  117.9    170.3   58.4   24.3  133.6  156.3   80.6   22.2   92.9  160.6   61.3 
 Breast 106.2 1835.0   2654.4  857.8  107.9 1705.5 2714.3 1070.7   82.3 1163.8 1824.2  617.3 
 Cervixa  26.6  236.4    282.1  109.1   24.7  178.7  220.0  105.9   25.5  194.4  161.7   86.5 
 Corpus uteri   1.6  196.7    844.8  176.4    7.1  271.4  865.8  254.9    4.7  271.7  708.4  182.3 
 Hodgkin’s diseasea  21.8   26.7      2.9   19.7   35.6   24.3   19.3   29.2   20.1   17.6   13.1   18.4 
 Leukaemia  15.0   62.6    176.9   50.6   17.8   32.1   80.0   34.5   11.9   26.9   74.0   25.6 
 All cancers 382.8 4214.3   8926.1 2512.9  492.9 4053.2 8349.0 3062.4  400.8 2978.4 6127.1 1937.8 

a15-year prevalence.

Table 5.

 Prevalence according to time since diagnosis for each cancer site, by country and sex (per 100 000 and as percentage of total prevalence for each site)

Cancer site Time since diagnosis (years) Men  Women 
 France  (%) Italy  (%) Spain  (%)  France  (%) Italy  (%) Spain  (%) 
Stomach   1  10.0 14.7 24.7 15.2 17.3 15.3  7.6 20.4 17.6 15.3 10.5 17.6 
   1–5  19.2 28.4 47.2 29.0 31.1 27.5  10.5 28.3 34.6 30.2 16.5 27.6 
   5–10  11.2 16.5 31.5 19.3 25.1 22.2  5.7 15.4 26.4 23.1 14.6 24.3 
 >10  27.3 40.4 59.5 36.5 39.4 34.9  13.3 35.9 35.9 31.3 18.2 30.5 
Colon   1  27.2 16.3 34.2 16.6 22.6 17.9  21.0 13.1 31.4 14.5 18.3 15.5 
   1–5  65.2 39.1 76.6 37.1 48.1 38.2  57.6 36.0 74.8 34.6 42.7 36.1 
   5–10  32.8 19.7 49.7 24.1 28.3 22.4  36.7 22.9 56.6 26.2 29.2 24.7 
 >10  41.4 24.9 45.9 22.2 27.1 21.5  44.9 28.0 53.6 24.8 28.0 23.7 
Rectum   1  18.7 15.5 20.6 16.5 16.0 17.1  12.2 11.9 15.2 13.9 9.1 13.2 
   1–5  48.2 40.0 45.6 36.6 34.5 36.9  35.2 34.2 37.5 34.2 24.7 35.8 
   5–10  23.4 19.4 29.8 23.9 22.8 24.4  20.0 19.4 25.3 23.2 15.9 23.1 
 >10  30.3 25.1 28.8 23.0 20.1 21.5  35.4 34.5 31.4 28.7 19.3 28.0 
Lung   1  76.1 30.1 59.5 31.8 40.1 28.6  6.4 32.8 11.0 33.2 2.7 26.1 
   1–5  83.5 33.1 56.4 30.1 40.7 29.0  5.0 25.5 11.5 34.6 3.7 35.2 
   5–10  29.1 11.5 28.8 15.4 24.6 17.5  4.3 21.8 5.8 17.4 2.2 21.1 
 >10  63.9 25.3 42.5 22.7 35.0 25.0  3.9 19.9 4.9 14.9 1.8 17.6 
Melanoma   1  4.5 19.5 6.8 15.0 3.9 13.6  6.4 11.0 8.5 10.5 5.6 9.1 
   1–5  10.1 43.8 17.5 38.7 11.5 39.6  17.4 29.9 22.1 27.4 18.3 29.9 
   5–10  4.1 17.9 11.6 25.5 7.7 26.6  10.7 18.3 19.8 24.6 15.4 25.2 
 >10  4.3 18.8 9.4 20.8 5.9 20.2  23.9 40.9 30.2 37.5 21.9 35.7 
Breast   1  – – – – – –  92.1 10.7 114.3 10.7 71.4 11.6 
   1–5  – – – – – –  311.7 36.3 360.8 33.7 216.9 35.1 
   5–10  – – – – – –  219.9 25.6 293.0 27.4 165.6 26.8 
 >10  – – – – – –  234.1 27.3 302.6 28.3 163.4 26.5 
Cervixa   1  – – – – – –  10.2 9.4 9.6 9.0 8.2 9.4 
   1–5  – – – – – –  24.6 22.5 28.5 26.9 27.1 31.4 
   5–10  – – – – – –  41.9 38.4 34.5 32.5 26.6 30.7 
  10–15  – – – – – –  32.4 29.7 33.4 31.6 24.6 28.5 
Corpus uteri   1  – – – – – –  13.4 7.6 19.1 7.5 15.4 8.4 
   1–5  – – – – – –  35.2 19.9 60.2 23.6 46.5 25.5 
   5–10  – – – – – –  42.6 24.2 61.7 24.2 43.4 23.8 
 >10  – – – – – –  85.3 48.3 114.0 44.7 77.0 42.3 
Prostate   1  55.9 19.3 45.5 21.8 33.2 21.0  – – – – – – 
   1–5  130.5 45.1 95.4 45.6 72.9 46.1  – – – – – – 
   5–10  69.7 24.1 46.6 22.3 36.0 22.8  – – – – – – 
 >10  33.3 11.5 21.6 10.3 15.8 10.0  – – – – – – 
Hodgkin’s diseasea   1  3.3 13.1 2.9 9.1 2.2 7.4  1.1 5.7 3.0 10.1 1.7 9.4 
   1–5  6.3 24.6 10.6 32.8 10.5 35.8  5.0 25.5 8.2 28.2 6.0 32.8 
   5–10  7.0 27.7 10.4 32.3 9.1 31.1  6.3 32.1 9.8 33.4 5.5 30.0 
  10–15  8.8 34.6 8.3 25.8 7.5 25.7  7.2 36.7 8.3 28.3 5.1 27.8 
Leukaemia   1  8.8 14.3 7.9 18.4 6.8 17.3  6.5 12.9 6.1 17.7 4.5 17.6 
   1–5  22.5 36.5 16.7 39.2 16.2 41.4  15.1 29.8 13.0 37.8 10.3 40.1 
   5–10  18.0 29.2 9.8 23.0 9.5 24.2  18.0 35.7 8.7 25.3 6.9 26.9 
 >10  12.3 20.0 8.3 19.4 6.7 17.2  10.9 21.6 6.6 19.3 4.0 15.5 
All cancers   1  392.2 17.9 351.3 16.8 283.1 15.9  241.3 9.6 325.0 10.6 203.5 10.5 
   1–5  825.1 37.6 740.3 35.3 600.4 33.6  699.7 27.8 842.7 27.5 543.5 28.0 
   5–10  401.5 18.3 477.3 22.8 419.2 23.5  525.3 20.9 688.7 22.5 439.2 22.7 
 >10  573.3 26.2 527.5 25.2 482.1 27.0  1046.7 41.7 1205.9 39.4 751.6 38.8 
Cancer site Time since diagnosis (years) Men  Women 
 France  (%) Italy  (%) Spain  (%)  France  (%) Italy  (%) Spain  (%) 
Stomach   1  10.0 14.7 24.7 15.2 17.3 15.3  7.6 20.4 17.6 15.3 10.5 17.6 
   1–5  19.2 28.4 47.2 29.0 31.1 27.5  10.5 28.3 34.6 30.2 16.5 27.6 
   5–10  11.2 16.5 31.5 19.3 25.1 22.2  5.7 15.4 26.4 23.1 14.6 24.3 
 >10  27.3 40.4 59.5 36.5 39.4 34.9  13.3 35.9 35.9 31.3 18.2 30.5 
Colon   1  27.2 16.3 34.2 16.6 22.6 17.9  21.0 13.1 31.4 14.5 18.3 15.5 
   1–5  65.2 39.1 76.6 37.1 48.1 38.2  57.6 36.0 74.8 34.6 42.7 36.1 
   5–10  32.8 19.7 49.7 24.1 28.3 22.4  36.7 22.9 56.6 26.2 29.2 24.7 
 >10  41.4 24.9 45.9 22.2 27.1 21.5  44.9 28.0 53.6 24.8 28.0 23.7 
Rectum   1  18.7 15.5 20.6 16.5 16.0 17.1  12.2 11.9 15.2 13.9 9.1 13.2 
   1–5  48.2 40.0 45.6 36.6 34.5 36.9  35.2 34.2 37.5 34.2 24.7 35.8 
   5–10  23.4 19.4 29.8 23.9 22.8 24.4  20.0 19.4 25.3 23.2 15.9 23.1 
 >10  30.3 25.1 28.8 23.0 20.1 21.5  35.4 34.5 31.4 28.7 19.3 28.0 
Lung   1  76.1 30.1 59.5 31.8 40.1 28.6  6.4 32.8 11.0 33.2 2.7 26.1 
   1–5  83.5 33.1 56.4 30.1 40.7 29.0  5.0 25.5 11.5 34.6 3.7 35.2 
   5–10  29.1 11.5 28.8 15.4 24.6 17.5  4.3 21.8 5.8 17.4 2.2 21.1 
 >10  63.9 25.3 42.5 22.7 35.0 25.0  3.9 19.9 4.9 14.9 1.8 17.6 
Melanoma   1  4.5 19.5 6.8 15.0 3.9 13.6  6.4 11.0 8.5 10.5 5.6 9.1 
   1–5  10.1 43.8 17.5 38.7 11.5 39.6  17.4 29.9 22.1 27.4 18.3 29.9 
   5–10  4.1 17.9 11.6 25.5 7.7 26.6  10.7 18.3 19.8 24.6 15.4 25.2 
 >10  4.3 18.8 9.4 20.8 5.9 20.2  23.9 40.9 30.2 37.5 21.9 35.7 
Breast   1  – – – – – –  92.1 10.7 114.3 10.7 71.4 11.6 
   1–5  – – – – – –  311.7 36.3 360.8 33.7 216.9 35.1 
   5–10  – – – – – –  219.9 25.6 293.0 27.4 165.6 26.8 
 >10  – – – – – –  234.1 27.3 302.6 28.3 163.4 26.5 
Cervixa   1  – – – – – –  10.2 9.4 9.6 9.0 8.2 9.4 
   1–5  – – – – – –  24.6 22.5 28.5 26.9 27.1 31.4 
   5–10  – – – – – –  41.9 38.4 34.5 32.5 26.6 30.7 
  10–15  – – – – – –  32.4 29.7 33.4 31.6 24.6 28.5 
Corpus uteri   1  – – – – – –  13.4 7.6 19.1 7.5 15.4 8.4 
   1–5  – – – – – –  35.2 19.9 60.2 23.6 46.5 25.5 
   5–10  – – – – – –  42.6 24.2 61.7 24.2 43.4 23.8 
 >10  – – – – – –  85.3 48.3 114.0 44.7 77.0 42.3 
Prostate   1  55.9 19.3 45.5 21.8 33.2 21.0  – – – – – – 
   1–5  130.5 45.1 95.4 45.6 72.9 46.1  – – – – – – 
   5–10  69.7 24.1 46.6 22.3 36.0 22.8  – – – – – – 
 >10  33.3 11.5 21.6 10.3 15.8 10.0  – – – – – – 
Hodgkin’s diseasea   1  3.3 13.1 2.9 9.1 2.2 7.4  1.1 5.7 3.0 10.1 1.7 9.4 
   1–5  6.3 24.6 10.6 32.8 10.5 35.8  5.0 25.5 8.2 28.2 6.0 32.8 
   5–10  7.0 27.7 10.4 32.3 9.1 31.1  6.3 32.1 9.8 33.4 5.5 30.0 
  10–15  8.8 34.6 8.3 25.8 7.5 25.7  7.2 36.7 8.3 28.3 5.1 27.8 
Leukaemia   1  8.8 14.3 7.9 18.4 6.8 17.3  6.5 12.9 6.1 17.7 4.5 17.6 
   1–5  22.5 36.5 16.7 39.2 16.2 41.4  15.1 29.8 13.0 37.8 10.3 40.1 
   5–10  18.0 29.2 9.8 23.0 9.5 24.2  18.0 35.7 8.7 25.3 6.9 26.9 
 >10  12.3 20.0 8.3 19.4 6.7 17.2  10.9 21.6 6.6 19.3 4.0 15.5 
All cancers   1  392.2 17.9 351.3 16.8 283.1 15.9  241.3 9.6 325.0 10.6 203.5 10.5 
   1–5  825.1 37.6 740.3 35.3 600.4 33.6  699.7 27.8 842.7 27.5 543.5 28.0 
   5–10  401.5 18.3 477.3 22.8 419.2 23.5  525.3 20.9 688.7 22.5 439.2 22.7 
 >10  573.3 26.2 527.5 25.2 482.1 27.0  1046.7 41.7 1205.9 39.4 751.6 38.8 

a Percentages refer to 15-year prevalence.

Table 6.

 Proportions of prevalence in age classes 65+ and 45–64 years

 Age 65+ (years)  Age 45–64 (years) 
 France Italy Spain  France Italy Spain 
Stomach 81.8 72.0 63.5  16.5 25.5 32.0 
Colon 75.8 70.6 69.6  21.8 27.3 25.8 
Rectum 74.9 68.9 66.7  21.9 29.2 30.2 
Lung 56.4 59.1 56.7  40.8 38.9 39.0 
Melanoma 43.4 38.1 44.0  38.8 45.0 35.2 
Breast (female) 50.2 52.5 49.1  41.9 42.2 42.8 
Corpus uteri 77.6 70.3 64.6  21.8 28.2 33.8 
Cervix 41.9 43.0 31.1  42.4 44.7 51.0 
Prostate 92.6 90.5 90.8   7.2  9.5  9.2 
Hodgkin’s disease  5.2 12.2 10.0  27.3 27.3 28.5 
Leukaemia 54.9 45.6 44.5  22.6 26.5 24.7 
All cancers 55.0 57.9 54.5  24.2 33.5 33.7 
 Age 65+ (years)  Age 45–64 (years) 
 France Italy Spain  France Italy Spain 
Stomach 81.8 72.0 63.5  16.5 25.5 32.0 
Colon 75.8 70.6 69.6  21.8 27.3 25.8 
Rectum 74.9 68.9 66.7  21.9 29.2 30.2 
Lung 56.4 59.1 56.7  40.8 38.9 39.0 
Melanoma 43.4 38.1 44.0  38.8 45.0 35.2 
Breast (female) 50.2 52.5 49.1  41.9 42.2 42.8 
Corpus uteri 77.6 70.3 64.6  21.8 28.2 33.8 
Cervix 41.9 43.0 31.1  42.4 44.7 51.0 
Prostate 92.6 90.5 90.8   7.2  9.5  9.2 
Hodgkin’s disease  5.2 12.2 10.0  27.3 27.3 28.5 
Leukaemia 54.9 45.6 44.5  22.6 26.5 24.7 
All cancers 55.0 57.9 54.5  24.2 33.5 33.7 

References

1.
Berrino F, Capocaccia R, Estève J et al. (eds). Survival of Cancer Patients in Europe 1985–89: The EUROCARE-2 Study. IARC Scientific Publications No. 151; IARC: Lyon
1999
.
2.
Micheli A, Mugno E, Krogh V et al. Cancer prevalence in European registry areas.
Ann Oncol
 
2002
;
13
:
840
–865.
3.
Micheli A (ed). Cancer prevalence in Italy: the ITAPREVAL study.
Tumori
 
1999
,
85
:
307
–412.
4.
Capocaccia R, Verdecchia A, Micheli A et al. Breast-cancer incidence and prevalence estimated from survival and mortality.
Cancer Causes Control
 
1990
;
1
:
23
–30.
5.
Capocaccia R, De Angelis R, Frova L et al. Estimation and projections of colorectal cancer trends in Italy.
Int J Epidemiol
  1997,
26
:
924
–932.
6.
Capocaccia R, De Angelis R, Frova L et al. Estimation and projections of stomach cancer trends in Italy.
Cancer Causes Control
 
1995
;
6
:
339
–346.
7.
Colonna M, Hédelin G, Estève J et al. National cancer prevalence estimation in France.
Int J Cancer
 
2000
,
87
:
301
–304.
8.
Colonna M, Grosclaude P, Launoy G et al. Estimation of colorectal cancer prevalence in France.
Eur J Cancer
 
2001
;
37
:
93
–96.
9.
Capocaccia R, Colonna M, Corazziari I et al. Measuring cancer prevalence in Europe: the EUROPREVAL Project.
Ann Oncol
 
2002
;
13
:
831
–839.
10.
Capocaccia R, De Angelis R. Estimating the completeness of prevalence based on cancer registry data.
Stat Med
 
1997
;
16
:
425
–440.
11.
Merrill RM, Capocaccia R, Feuer EJ, Mariotto A. Cancer prevalence estimates based on tumour registry data in the Surveillance, Epidemiology, and End Results (SEER) Program.
Int J Epidemiol
 
2000
;
29
:
197
–207.
12.
Parkin DM, Whelan SL, Ferlay J et al. (eds). Cancer Incidence in Five Continents, Vol. VII. IARC Scientific Publications, No. 143; IARC: Lyon
1997
.

Author notes

1Istituto Superiore di Sanità, Rome; 2Istituto Nazionale Tumori, Milan, Italy; 3Registre de Tumeur d’Isère, Grenoble, France; 4Institut Catala d’Oncologia, Barcelona; 5Registro de Cancer de Euskadi, Bilbao, Spain; 6Registro Tumori Toscano, Florence, Italy