Cancer Mortality in Broad Italian Geographical Areas, 1975-1977

Death certification rates from various cancers or groups of cancers in broad Italian geographical areas (North/Center/South) over the period 1975-1977 were analyzed. In both sexes, there was a clear North/South gradient, with considerably higher rates in the North for total cancer mortality as well as for most common neoplasms. The geographical variation was more marked in males (North/South ratio for total cancer mortality = 1.75 at all ages and 1.70 truncated 35-64 years) than in females (ratio = 1.48 at all ages and 1.28 truncated 35-64). Although, in general terms, the present results confirm previous analyses of cancer mortality in Italy, a few interesting tendencies should be noted. First, the geographical differences in the mid-late 1970's were much more marked for tobacco-related cancers (a factor of over two in males in the North/South ratio) than for other chiefly epithelial carcinomas or nonepithelial cancers. In general, variations for nontobacco-related cancers tended to level off over more recent calendar periods. However, there was little tendency towards decreasing differences in gastric cancer mortality (which was markedly elevated in the North and Center), at least in males. During the 1970's death certification rates from cancer of the (cervix) uteri decreased in northern and central more than in southern Italy. This pattern of trends may have been influenced by a different impact of cervical screening in various areas of the country.

Cancer Mortality in Italy, 1994, and an Overview of Trends from 1955 to 1994

Background

Data and statistics are presented on cancer death certification in Italy, updating previous publications covering the period 1955-1993.

Methods

Data for 1994 and the quinquennium 1990-94 subdivided into 30 cancer sites are presented in 8 tables, including age- and sex-specific absolute and percentage frequencies of cancer deaths, and crude, age-specific and age-standardized rates, at all ages and truncated for the 35-64 year age group. Trends in age-standardized rates for major cancer sites are plotted from 1955 to 1994.

Results

The age-standardized (world standard) death certification rates from all neoplasms steadily declined from the peak of 199.2/100,000 males in 1988 to 186.3 in 1994, and in females from 102.5 in 1989 to 98.6 in 1994. Ever larger was the decline in truncated rates, for males from the peak of 275.1/100,000 in 1983 to 223.2 (-19%) in 1994, and for females from 151.6/100,000 in 1987 to 136.4 (-10%). A major component of the favourable cancer mortality trends in males was lung cancer (accounting for 31,000 deaths in both sexes combined in 1994), whose overall age-standardized rates declined from 60.3 in 1987-89 to 54.6/100,000 males in 1994 (-9%), and from the peak of 96.7 in 1983 at ages 35 to 64 to 72.7 in 1994 (-25%). In contrast, female lung cancer rates have remained stable from 1992 onwards, but have increased from 7.2 to 7.7 at all ages and from 10.6 to 11.0 at age 35-64 between 1985-89 and 1990-94. These different trends in the two sexes reflect the patterns and trends in smoking among Italian males and females.

Conclusions

Cancer mortality trends in Italy over the period 1990-94 were relatively favourable, mainly reflecting the decline in lung cancer rates in males, together with the persistent declines in gastric cancer in both sexes and in cervix uteri for women. Continuous advancements were registered for neoplasms amenable to treatment, essentially testicular cancer, Hodgkin's disease and childhood leukaemias. The major unfavourable trends were observed for non Hodgkin's lymphomas, and require therefore further monitoring, besides a clearer understanding of their determinants. Italy maintains an intermediate level of cancer mortality on a European scale, suggesting that further progress is possible, mostly for tobacco-related neoplasms in males.

Comparative Completeness and Correspondence of Cancer Mortality Data as Collected by Istat and Cancer Registries

The authors have compared individual cancer mortality records from the file of the Central Statistics Office (ISTAT) with those of the Cancer Registries of Varese and Torino for the period July-December 1976. Most diagnoses ascertained by the Cancer Registries were histologically confirmed. The results showed a high degree of reciprocal completeness (over 90 %) for both systems (Cancer Registries and mortality statistics) and a satisfactory level of concordance (75 % at Varese, 70 % at Torino). The discrepancies, concentrated in certain nosologic codes, were analyzed. The conclusions suggest that the correspondence between mortality statistics and clinical diagnoses is hight, at least in the geographical areas examined.

Cancer Mortality in Italy: An Overview of Age-Specific and Age-Standardised Trends from 1955 TO 1984

Number of certified deaths, age-specific and age-standardised rates and percentages of all cancer deaths from 30 cancers or groups of cancers (plus total cancer mortality) for each five-year calendar period between 1955 and 1984 in Italy are presented in tabular form. From these data, three graphs are derived, including trends in age-standardised rates, age-specific rates centered on birth cohorts and maps plotted in different shades of grey to represent the surfaces defined by the matrix of various age-specific rates. These analyses quantified the rises in overall cancer mortality in males (from 137 to 192/100,000 world standard), chiefly due to increases in lung and other tobacco-related neoplasms. Overall cancer mortality was stable in females (around 100/100,000). Appreciable cohort effects were evident for tobacco related neoplasms, but also for other major cancer sites, such as Intestines or breast, whose rates, after earlier rises, are now stable in earlier middle age. Since the early 1970's, cancer mortality rates have been declining in all age groups below 40 in males and below 55 in females. These declines reflect Improvements in therapy for leukemias, lymphomas and germ cell tumors, and general improvements in food availability and storage, hygiene and early diagnosis, which have led to the declines in stomach and cervical cancer. Although moderate in absolute terms and smaller than in other western countries where tobacco-related neoplasms have also been falling in more recent cohorts, these declines are encouraging for the indication they provide on the most likely patterns over the next decades in the same and subsequent generations.

Descriptive Epidemiology of Hodgkin's Disease in Italy

Death certification data on Hodgkin's disease in Italy over the period 1955-84 were studied in terms of age-standardized and age-specific national trends, and of geographical variation in mortality. There were substantial declines in death rates from the early 1970's onward, which can be largely attributed to therapeutic improvements. These led to avoidance of about 350 deaths, with a total 950 reported, which is probably the major absolute therapeutic advance identified for any cancer site. The declines started earlier in childhood and young adult age, and were restricted to population below age 60. The age distribution of the disease was different in the two sexes, since the age curve for males showed steady rises up to age 75, whereas that for females was clearly bimodal, with a peak around age 30, and another at oldest age. This divergent pattern is consistent with different exposure to (infectious) agent(s) in children of the two sexes, but also to occupational exposures potentially related to the risk of the disease. Examination of rates in various geographical areas showed generally higher rates in the North, and a few provinces with exceedingly high mortality in the central part of Northern Italy, particularly in a chiefly rural province (Mantua). This excess mortality (and, more in general, the observation that rates for Northern Italy are higher than in any other area of the EEC) could not be explained by obvious diagnostic or classification problems, were evident in both sexes, appeared to be consistent over the last decade and are reflected in available Italian cancer registration data.

Cancer Mortality in Migrant Populations within Italy

Italian death certification rates from all causes of death, all diseases of the circulatory system, all neoplasms, and cancers of the upper digestive and respiratory tract, stomach, intestines, lung and breast in middle age (45-64 years) were analyzed according to selected geographic areas of birth and residence at death. For total cancer mortality and most neoplasms considered, the rates in middle age were closer to those of place of birth than to those of area of residence, although this pattern was more evident for some sites (e.g., mouth or pharynx, esophagus, larynx, stomach or bladder) than for others (e.g., intestines or breast). In most cases, migration had an adverse effect on cancer rates, and the lowest mortality was reported among stable populations (i.e., those with the same area of birth and death). These findings are discussed in relation to the major migration fluxes within Italy during the current century. Moreover, these analyses give information on the quality of Italian cancer death certification, since the observation that area of birth is often a more important determinant of cancer rates than area of residence provides indirect evidence that cancer death certification in various Italian geographic areas is satisfactorily reliable and consistent.

[Patients' satisfaction for hospital care among the Italian Regions between 1997 and 2009]

Patients have a privileged perspective on health care quality, thus it is important to obtain from them judgments about offered services. Currently, there are no well developed quality monitoring systems related to the patient's perspective, though Istat data sources would b helpful on this issues. We performed a descriptive analysis relating to the satisfaction degree of Italian population for hospital admissions between 1997and 2009. Data were taken from the Istat Multipurpose analysis on some aspect of daylife. Our results show a positive appraisal for medical and nursing assistance; however the rating for the quality of food is lower Our analysis would give a contribution to the appraisal of an outstanding dimension of patients' appraisal of quality of care among the Regions, by considering the implementation of devolution in healthcare since 2001.

[Sensorineural hearing loss in childhood: evaluation of economic impact in view of vaccine prevention of cases due to congenital cytomegalovirus infection]

Sensorineural hearing loss (SNHL) is a serious public health problem which affects 1-3% per hundred live born babies in developed countries. The congenital cytomegalovirus (CMV) infection is its most important non-genetic cause. The evaluation of the effectiveness of future programs of anti-CMV vaccination requires an assessment of the present costs of SNHL. Direct costs for the Italian public system were calculated per prosthesis child until his full age and turned out to add up to 260,000 euro. Private costs are difficult to be assessed and anyhow are highly dependent from the socio-economic level. This preliminary assessment suggests that the vaccination would be cost-saving if SNHL cases due to congenital CMV were more than 21 per year, corresponding to a congenital infection prevalence higher than 0.21%o.

[Health care provider payment systems]

Health care provider payment systems regulate the relationship between patients, providers, and third payers in order to maximise benefits and minimise costs of the whole health care system. Health care providers could be paid by a price or a fee for service, by capitation systems, or by reimbursement of production costs. It would be interesting to develop innovative payment systems aimed to the payment of the entire health care pattern of patients. This would be particularly desirable for certain health conditions where it is impossible to divide the health care delivery pattern into single health services e.g. psychiatric care, long term and rehabilitation care.

[The evaluation of efficacy and efficiency in the health care sector: separate or integrated moments?]

Studies on efficacy (clinical trials) and efficiency (Cost Benefit) in health care are frequently disjoint and carried on by researchers with different background in distinct moments. The origin of this division can be found in the profound conviction existing in the healthcare researchers that efficiency and efficacy are distinct and distant concepts, the former pertaining to the economist, the latter to the clinician. Many are the factors at the basis of this separation which consequently lead to the divergence between the two sector of analysis; among those, probably the most relevant factor is the distinction, in the healthcare sector, between the consumer (the patient) and the purchaser (private and/or public insurance). In reality, the organizational evolution of the health care systems, the consciousness of the interdependency between health and economic benefits, and the progressive shortage of economic resources for the growing healthcare needs, require a major integration of analyses concerning efficacy and efficiency. On this line of thought is moving the operational research where models such as the Data Envelope Analysis and the Semi-Markov Decisional Models have been developed.

Pollen mixtures used as health food may be a harmful source of allergens

We describe herein one case of systemic anaphylaxis due to the ingestion of an undefined mixture of pollens, sold as a dietary supplement. The patient, who suffered from rhinoconjunctivitis due to grass pollen (with sensitization to several trees), had a severe episode of anaphylaxis immediately after eating this health food. The episode required emergency care. We attempted to study the pollen mixture responsible, but no pollen granules could be identified. We prepared a solid phase with the pollen mixture, and we observed a RAST positivity with the patient's serum and pools of sera containing specific IgE to trees. Furthermore, a RAST-inhibition assay of the patient's serum showed highly positive results with grasses, birch, alder and Compositae. Therefore, we concluded that the pollen mixture contained determinants capable of cross-reacting with the patient's IgE. This case report is evidence of the possible risks due to the use of undefined herbal products by allergic patients.

Anaphylactic reaction to the ingestion of raw onion. A case report

A case of severe systemic reactions (intense itching, urticaria, confusion, blurred vision, transient loss of consciousness, sweating, tachycardia) after ingestion of raw or lightly-cooked onion is described. The patient, a 44-year-old woman, had no troubles with well-cooked onions. Differently from the cases of sensitivity to onion described in literature, this patient was monosensitized, being skin tests negative to pollens, inhalants and other foods. The patient had 3.7 kU/L of onion-specific segum IgE, as determined by REAST. The density of onion-specific IgE (calculated as percent ratio to total IgE) was 30.8%. The reactivity of patient's serum IgE towards thermolabile and thermostable components has been tested with unheated and heated (30' at 100 degrees C) onion extracts bound to polystyrene beads and tested in the RAST system. Unheated extract resulted positive in class 2, heated extract negative, demonstrating that this patients, differently from similar clinical cases described in literature, had IgE antibodies recognizing just thermolabile onion fraction. This is the first case described in literature of a monosensitization to the thermolabile component of onion, negative also to related foods (Liliacee) and characterized by severe systemic reactions. The importance of specific-IgE density (%) rather their absolute amount (kU/L) as parameter predictive for the clinical severity of allergic reactions is discussed.

First 10,000 chorionic villus samplings performed on singleton pregnancies by a single operator

Chorionic villus sampling (CVS) was performed in 10,000 consecutive singleton pregnancies by a single principal operator, working in two institutions. The procedure was performed between 8 and 32 gestational weeks: transabdominal (TA) sampling was carried out in 8479 cases and transcervical (TC) in 1521. Patients were referred for chromosomal risk in 89.1 per cent of cases, Mendelian disorders in 10.5 per cent, and DNA investigations for paternity or infectious agents in 0.4 per cent of cases. The sampling success rate for both TA and TC techniques by the second insertion was 99.8 and 99.2 per cent, respectively. TA sampling succeeded in a higher number of cases at the first insertion (98 per cent vs. 86.8 per cent) and was associated with smaller samples (< 10 mg) in fewer cases (3.2 per cent vs. 4.9 per cent). Cytogenetic analysis was highly successful (99.4 per cent) and accurate; however, in one case a de novo structural rearrangement of chromosome I was not recognized. Mosaicism or rare trisomies were reported in 1.30 per cent of cases. Five diagnostic errors in DNA investigation (0.51 per cent) ended with the birth of affected fetuses. Fetal loss through 28 weeks' gestation in the pregnancies intended to continue was 2.58 per cent; the rate increased with maternal age (1.22 per cent at less than 30 years to 3.8 per cent at 40 years or more), while gestational age affected the abortion rate only at 8 weeks (odds ratio=2.22, P<0.05). Rates of premature delivery, low birth weight, and perinatal mortality did not differ from the Italian standards. By comparison with the Italian Birth Defects Registry data, no differences were found for the major malformations, including transverse limb reduction defects (TLRDs) (4.34 vs. 3.28 x 10,000). Total malformations and TLRDs did not show any pattern relation to either maternal age or gestational age.

[Geographic distribution of digestive system tumor mortality in Italy, 1970-87]

The geographical distribution of mortality rates from tumours of digestive tract in Italy is analyzed in this paper. The analysis is based on official mortality data collected by the National Institute of Statistics (ISTAT). Age-adjusted mortality rates for stomach cancer presented the highest values in some provinces of the North and the Center, and the lowest values in the South of Italy. A same geographical pattern was observed for men and women, and during the whole considered period. Colorectal cancer presented the highest rates in the North-West of the country, and in some provinces of Liguria and Tuscany. The lowes rates were observed in the South, particularly in Calabria and Sicily. Mortality for this cancer was positively associated with degree of urbanization. The geographical pattern remained fairly constant in time, but the North-South differences narrowed during the years 1980s. A similar geographical distribution, characterized by the highest mortality levels in the north-eastern regions of Italy, was observed for cancers of the oral cavity and pharynx, of the oesophagus, of the pancreas, and for male tumours of the liver. Female liver cancer presented, on the contrary, the highest mortality levels in the southern regions.

Exploratory tools for disease mapping in geographical epidemiology

In the present paper a new shrinkage estimator of relative risk, useful in disease mapping, is compared with the empirical and full Bayes estimators, using death certificate data for lung cancer (males and females) and breast cancer (females) 1982-1988 in the Emilia-Romagna region (Italy). The estimates are obtained averaging the relative risks obtained from recursive partitioning of the set of observed values. The number of partitions is fixed in advance and behaves as a smoothing parameter, the lower the number the higher the degree of shrinkage toward the overall mean of the observed values.

[The Italian mortality map at the municipal level]

The Italian Atlas of mortality at municipality level is the result of a research project coordinated by the Emilia-Romagna region and supported by the Health Ministry. To the realisation of this project have collaborated the Institute of Medical Statistics of Milano and the CILEA under the supervision of Prof. Cesare Cislaghi. Compared to previous equivalent products, this Atlas contains a number of methodological and content innovations. This was a consequence of the development of new statistical methods and the need of achieving different aims. First of all, the Atlas was aimed at describing the mortality at small area level, which is obtained via Kernel estimates; secondly, the objective was to identify suspect clusters of deaths which may suggest the existence of high risk areas. The Mortality Atlas is formed by 31 tables, one for each of the analysed cause of death; each table has a circular shape of 100 hundred kilometers radius and contains a variable number of municipalities; each municipality may be present in more than one circle. The Atlas is available on magnetic support and for each cause of death are provided several statistical analyses and indicators included in different files. One of these files can be directly used to build high quality maps using the graphical package MAPINFO.

[A new approach for the automatic detection of health event clusters]

The Italian Morality Atlas at municipality level, sponsored by the Health Ministry and realized by the Emilia-Romagna region in collaboration with the Institute of Medical Statistics of Milano, is aimed at: 1) supporting and stimulating the local production of mortality maps and the use of geographical data; 2) providing a working and research tool to local health planners; 3) setting a routine procedure for the detection of spatial clusters. This procedure is organized in several steps: the definition of the spatial domain and of the spatial distance metrics; the choice of the effect indicator and the standard population; the estimate of the density risk surface via kernel methods; the identification of "local maxima"; the analysis of local maxima for cluster detection; the description of the identified clusters. The paper investigates the meaning and the characteristics of spatial clusters, describes the procedure for their identification and discusses the opportunity to use an automated approach for map reading.

[The analysis of the spatial aggregation of health events using risk surface density estimation methods]

Methods for the analysis of the spatial aggregation of health events has received growing attention under the pressure of public opinion concern and as tools for the identification of potential risk sources, for monitoring relevant geographical areas and, finally, for public health decisions. The development of statistical methods for the detection and localization of spatial clusters has mainly concerned individual data. This paper is aimed at describing one of the methods proposed for the identification of clusters in the case of information at individual level and at presenting its extension to grouped data. This method, the surface density estimation method using the Kernel approach, offers remarkable advantages in terms of simplicity of implementation and flexibility, this latter being an extremely important characteristic in the case of exploratory analyses. For exemplification purposes, the density estimation method has been applied to individual data concerning the spatial distribution of cerebral tumors in Campi Bisenzio (FI) and to the distribution of gastric cancer mortality in the municipalities around Arezzo and Pesaro.

First-trimester Down's syndrome screening using nuchal translucency: a prospective study in patients undergoing chorionic villus sampling

The value of the measurement of nuchal translucency thickness for predicting fetal Down's syndrome and other aneuploidies was prospectively evaluated at 8-15 weeks of gestation in 1819 consecutive pregnancies scheduled for karyotyping by chorionic villus sampling. In 43 cases, a chromosomal unbalanced aberration was found. Two teams of ultrasonologists who examined patients attending either National Health Service (Series 1) or private practice clinics (Series 2) were involved in the study. The same type of ultrasound machine and standardized approach were used in both study groups. In those cases in which the maximum subcutaneous thickness of the translucency was 3 mm or greater, the incidence of chromosomal aberration was 18.6% compared to 1.7% in the cases in which this was below 3 mm. The sensitivity, specificity and relative risk for all aneuploidies were 30%, 96% and 10.83, respectively, and no difference was found between trisomy 21 and other types of aneuploidy. The sensitivity and specificity and relative risk were significantly higher at 9-10 weeks than between 11 and 15 weeks. The results were concordant in the two series; however, the overall values for sensitivity (20% vs. 39%), specificity (94% vs. 98%) and relative risk (4.13 vs. 24.20) were clearly higher in the group of private patients. The results obtained confirm the potential application of the measurement of nuchal translucency thickness for fetal aneuploidy screening before the end of the first trimester and suggest that a multiplicity of individual, structural and organizational factors may interact and play a crucial role in determining the actual efficiency of ultrasound screening programs.

Evaluation of allergenic potency by REAST inhibition. A new tool for the standardization of allergenic extracts

The potency of allergenic extracts can be determined in vitro by RAST inhibition, and this has become the preferred method for the standardization of allergens. A disadvantage of this technique is the impossibility of obtaining data about allergens bound to the solid phase, i.e., the counterpart of the inhibiting extract. The REAST (reverse enzyme allergosorbent test) is based on the capture of IgE by a specific antibody bound to microtiter wells, the reaction of captured IgE with biotinylated allergen and the development of a colour reaction by subsequent addition of streptavidin-peroxidase and chromogenic substrate. The addition of an allergen extract in a dose-response fashion competes with the biotinylated allergen and inhibits the test. In the present study REAST inhibition has been evaluated with Dermatophagoides pteronyssinus, Parietaria judaica and mixed grass pollen extracts. The correlation of REAST inhibition with RAST inhibition and both intra-assay and inter-assay reproducibility have been evaluated. REAST inhibition is a potentially valuable new tool for the standardization of allergenic extracts.

[Health problems of foreign immigrants in Italy: the answer from the institutions]

The possibility and the result of the meeting between immigrant's health problems and the health services system are conditioned by legal bonds and by the ability of the system to meet different cultures and values. In the present work we outline the Italian legislation about immigration to search those critical points that may help to understand some features of immigration in Italy. We describe the foreigner's health problems, according the present documentary evidence, and we outline adequate answers.

Effects of treatment with verapamil SR and captopril on the lipid profile of hypertensive patients

The potential beneficial effects of antihypertensive drugs on cardiovascular morbidity and mortality may be compromised by their adverse effects on serum lipid levels. In our study we compared verapamil and captopril and evaluated their effects on blood pressure and on serum lipid and lipoprotein levels, with particular attention to lipoprotein(a) [Lp(a)]. 20 hypertensive patients were treated with sustained release verapamil 240mg once daily or captopril 25mg twice daily for 3 months in a double-blind randomised study. Diastolic blood pressure was reduced from 100 +/- 3mm Hg to 87 +/- 6mm Hg (p < 0.01) and from 100 +/- 5mm Hg to 92 +/- 7mm Hg (p < 0.05) in the verapamil and captopril groups, respectively. Small but significant changes in serum lipid levels were noted: total cholesterol was reduced from 6 to 5.8 mmol/L (verapamil) and from 6.1 to 5.9 mmol/L (captopril); low density lipoprotein (LDL) cholesterol was reduced from 4 to 3.8 mmol/L (verapamil) and from 4.2 to 3.9 mmol/L (captopril); apolipoprotein C-III was reduced from 0.3 +/- 0.07 to 0.2 +/- 0.06 mmol/L (9.7 +/- 2.5 to 9.2 +/- 2.3 mg/dl) [verapamil] and from 0.2 +/- 0.1 to 0.2 +/- 0.09 mmol/L (9.1 +/- 3.7 to 8.3 +/- 3.4 mg/dl) [captopril]; apolipoprotein A-II increased only with verapamil (p < 0.02). Lp(a) levels showed only minor changes in individual patients. In conclusion, in our study verapamil and captopril were effective antihypertensive agents and did not adversely effect the lipid profile.

Cancer mortality in young adults: Italy 1955-1985

Although cancer mortality in young adults accounts for only a small proportion of all cancer deaths, it is important since it provides useful indications of the most likely future trends, and relevant information on the role of exposure to specific, or newer, carcinogens. We, therefore, analysed trends in cancer mortality between 1955 and 1985 among Italian men and women aged 20-44 years. In those three decades, overall cancer mortality declined steadily, by 27% in young women (from 33.8 to 24.7/100,000, world standard) but only by 3% (from 27.3 to 26.4/100,000) among men. The decline for men, however, was 16% from the peak rate of 31.5 reached in 1970-1974. The major underlying component causing the different trends in the two sexes was lung and other tobacco-related neoplasms, which had been considerably on the increase in young men up to the early 1970s, and levelled-off thereafter, while showing no appreciable change in women. The falls were about 50% for stomach cancer in both sexes, and over 80% for cervical cancer. A clear impact of improved treatment was reflected in the substantial declines in Hodgkin's disease, of testicular cancer in the last decade and, possibly, in the favourable trends in cancers of the breast, bone, brain and leukemias over the most recent calendar periods. Only two sites showed appreciable and persisting upward trends: oral cavity in men and skin melanoma in both sexes. They therefore constitute priorities for intervention in the near future.

The application of trend surface models to the analysis of time factors in Swiss cancer mortality

To study different temporal components on cancer mortality (age, period and cohort) methods of graphic representation were applied to Swiss mortality data from 1950 to 1984. Maps using continuous slopes ("contour maps") and based on eight tones of grey according to the absolute distribution of rates were used to represent the surfaces defined by the matrix of various age-specific rates. Further, progressively more complex regression surface equations were defined, on the basis of two independent variables (age/cohort) and a dependent one (each age-specific mortality rate). General patterns of trends in cancer mortality were thus identified, permitting definition of important cohort (e.g., upwards for lung and other tobacco-related neoplasms, or downwards for stomach) or period (e.g., downwards for intestines or thyroid cancers) effects, besides the major underlying age component. For most cancer sites, even the lower order (1st to 3rd) models utilised provided excellent fitting, allowing immediate identification of the residuals (e.g., high or low mortality points) as well as estimates of first-order interactions between the three factors, although the parameters of the main effects remained still undetermined. Thus, the method should be essentially used as summary guide to illustrate and understand the general patterns of age, period and cohort effects in (cancer) mortality, although they cannot conceptually solve the inherent problem of identifiability of the three components.

Descriptive epidemiology of intestinal cancer in Italy

Trends in the death rates from cancers of the intestine (including colon and rectum) in Italy from 1956 to 1981 were analyzed with a standard cross-sectional approach and a log-linear age/period/cohort model. In both sexes there were steady increases in mortality rates between the middle 1950s and the middle 1970s, chiefly explainable in terms of cohort effects. This was followed by a leveling off and stabilization, starting from the younger age groups and more evident in women. The analyses of the geographic distribution of intestinal cancer mortality in the 95 Italian provinces during 1975 to 1977 showed higher rates in the north of the country and in large urban concentrations, and a bimodal distribution of mortality rates, whose minimum corresponded to a distinct north/south separation. Trend surface models fitted to intestinal cancer standardized mortality ratios showed a high determination coefficient even for the simplest models. Residuals, corresponding to outliers, were scattered in a few northern and central areas. In addition, mortality rates for cancers of the intestine in middle age people were considered according to geographic area of birth and residence at death. In both sexes rates of migrant populations were influenced both by area of birth and residence, and, in particular, there was a wide-spread unfavorable effect of residence on migrants from the South to the North (the major migrant flux in Italy). The temporal and geographic variations in intestinal cancer rates observed in this study indicate that Italy may well be a particularly interesting situation for etiologic investigations of colorectal cancers.

Birth cohort, time, and age effects in Italian cancer mortality

Italian death certification data from 1955 to 1979 for total cancer mortality and 30 cancer sites in the population aged 25 to 74, were analyzed using a log-linear Poisson model to isolate the effects of birth cohort, calendar period of death, and age. The most frequent cohort pattern was characterized by increases up to the generations born between 1920 and 1930, followed by stabilization or a slight decrease. This pattern was evident for total cancer mortality in men, and for several common sites, including larynx, lung, esophagus, bladder, female breast, and ovary. Only four sites (pancreas, pleura, intestines in both sexes, and kidney in men) showed cohort values still rising in more recent generations. Stable cohort and period of death curves were observed for cancers of the prostate and testis, whereas trends were steadily going down for neoplasms of the stomach, and (cervix) uteri. Finally, there were a few discontinuous trends (e.g., in the case of brain neoplasms, leukemias, and lymphomas), which probably reflect different effects of improvements in diagnosis and/or treatment. Period of death values increased for lung and other tobacco related sites (chiefly in males) and, up to the early 1970s, for a few other common sites, including intestines, and the female breast. Downward trends over the calendar period were evident for cancers of the stomach and of the (cervix) uteri. Therefore, total cancer mortality trends over the calendar period of death were moderately increasing for men, and slightly decreasing for women.

Descriptive epidemiology of gastric cancer in Italy

National trends in death certification rates from cancer of the stomach in Italy over the period 1955 to 1979 were analyzed using a standard cross-sectional approach and a log-linear Poisson model to isolate the effects of birth cohort, calendar period, and age. Overall, age-standardized death certification rates decreased from 47.04 to 30.74/100,000 males (average annual rate of change, assuming that the decrease has been constant, -1.8%) and from 34.55 to 19.27/100,000 females (average annual rate of change -2.4%). The decreases were even larger in middle age for both sexes. Both cohort and period of death curves were markedly downwards. However, cohort values did not decrease for generations born around the second world war (1935-1945), thus indirectly confirming the importance of (dietary) habits in childhood on subsequent gastric cancer risk. Further, the geographic distribution of certified mortality from gastric cancer in the 95 Italian provinces over the period 1975 to 1977 was analyzed. Death certification rates were about 10% lower for both sexes in the 14 provinces including the largest urban concentrations (over 250,000 inhabitants) than in the remaining areas. This finding might be related to earlier availability of modern food processing and storage in urban areas. It is, however, more difficult to explain the lower mortality rates (about 50% in both sexes) in the southern compared with the central and northern areas, since southern Italy is the less developed part of the country. Likewise, there appears to be at present little satisfactory explanation for the several clusters of exceedingly high mortality areas scattered in northern and central Italy, since some of these areas are several hundred kilometers apart, and there is no obvious common denominator in diet or other environmental factors that may explain their higher gastric cancer mortality rates.

Mortality from alcohol related disease in Italy

Trends in death certification rates from the five major alcohol related causes of death in Italy (cancers of the mouth or pharynx, oesophagus, larynx, liver and cirrhosis of the liver) were analysed over a period (1955-79) in which per capita alcohol consumption almost trebled. Age standardised mortality from liver cirrhosis almost doubled in males and increased over 70% in females. In males, mortality from cancers of the upper digestive or respiratory tract showed increases of between 27% and 44%, and liver cancer increased by over 100%. In the late 1970s, the four alcohol related cancer sites accounted for about 12% of all cancer deaths in males and 4.5% in females. Mortality from liver cirrhosis alone accounted for 4.8% of all deaths in males (9.2% of manpower years lost) and 2.3% in females (6.3% manpower years lost) in females. These figures were even higher in selected areas of north eastern Italy, where alcohol consumption is greater. In absolute terms, the upward trends observed correspond to about 10,000 excess deaths per year in the late 1970s compared with rates observed two decades earlier and are thus second only to the increase in tobacco related causes of death over the same calendar period.