A case-control study on breast cancer risk factors in a southern European population.

A case-control study has been carried out among women attending a screening service in Palermo (Sicily) from 1974 through 1983 to ascertain the distribution of the most frequently investigated risk factors for breast cancer in a southern European population. Information has been obtained from the archives of the screening service. The analysis was separately conducted for pre- and post-menopausal cases and non-cases. Risk factors for pre-menopausal women are: nulliparity (nulliparous versus parous: OR 2.17, 95% CI 1.41-3.32); age at first birth (25-29 versus less than 20: OR 2.16, 95% CI 1.17-4.00); interval between menarche and first birth (greater than 20 years versus less than 6: OR 5.34, 95% CI 2.08-13.66); number of births (greater than 4 versus 1-2: OR 1.98, 95% CI 1.10-3.50). Risk factors for post-menopausal women are: nulliparity (nulliparous versus parous: OR 2.18, 95% CI 1.59-2.99); age at first birth (greater than 29 versus less than 20: OR 1.84, 95% CI 1.13-2.99); interval between menarche and first birth (16-20 years versus less than 6: OR 2.15, 95% CI 1.20-3.85). Age at menarche, age at menopause, breast feeding and family history were not found to be risk factors for breast cancer in the investigated population. The existence of influencing differences between northern and southern populations has been postulated.

Survival of patients with cancer of the stomach, colon and rectum, lung and breast in Ragusa, Sicily.

Survival of 1747 patients with cancer of the stomach, colon/rectum, lung and female breast was investigated on incident cases registered by the Ragusa Cancer Registry, covering the Province of Ragusa (275,000 inhabitants) between 1981 and 1986. Cases known from the death certificate only were excluded. Date of death was obtained by a linkage with death certificates issued for Ragusa residents. Living status was assessed directly at the Registrar's Office. Analysis was performed also by sex, age group and presence of histologic verification. Relative survival at 5 years of Ragusa cancer cases was of the same order of magnitude as that observed in the U.S.A. (1980) and in Switzerland (1970-1980). However, the slight differences observed were mostly in favor of the Swiss and American patients. Sex did not significantly influence survival, but younger patients survived longer than older ones. Lack of histologic verification was associated with poor survival.

[Analysis of the quality of death certificates for neoplasms in the province of Ragusa]. / Analisi della qualità dei certificati di morte per neoplasia in provincia di Ragusa.

The reliability of the death certificates issued in Ragusa for 379 of the 909 cancer patients registered in 1986 by the Ragusa Cancer Registry, who had died within 31 october 1988, was investigated. Data were available for 365 cases deceased. A consistent proportion (32.8%) of disagreements between diagnosis of the Registry and cause of death reported on death certificate was observed, concerning errors of the second and third digit of ICD-9 (21.9%) and for lack of mention of cancer in the death certificate. The main causes of the loss of information were the issuing of the certificate by a M.D. of the permanent medical ward and not by the family doctor, and the lack of the document released by the hospital upon discharge of the patient, usually containing indication on diagnosis and treatment.

[Multiple virus vaccines and the interference phenomena]. / Vaccinazioni antivirali multiple e fenomeni di interferenza.

It is well known that interference between two viruses simultaneously infecting the same host may bring about the exclusion of one. When administering attenuated viruses for immunization purposes this fact should be theoretically taken into account. The present tendency of associating attenuated virus into multiple vaccines (trivalent polio, mumps-rubella-measles) is supported by their proved effectiveness. The reasons of the non-operativity of the interference mechanism in such conditions are discussed.

Comparison of cancer incidence among similar geographic areas: a better source of etiologic hypotheses?

Difference in incidence rates of cancer of specific sites among different populations has been a powerful tool for the generation of etiologic hypotheses. The use of strictly comparable data from cancer registries, on populations from Italy, France, Switzerland and Spain, similar for geographic location, race and ancestry, shows unexpected differences of rates of single cancer sites, being significantly higher or lower than those of the same subgroup of registries. The study of risk factors associated with such differences might be more profitable than those carried out on the basis of comparisons between distantly located populations.

A case-control study on lip cancer risk factors in Ragusa (Sicily).

Incidence rates of lip cancer in males in Ragusa (Sicily) are amongst the highest in Europe [age-standardized rate (world) for 1980-82: 7.5 per 100,000]. A case-control study was conducted on 53 male cases and 106 controls matched for sex, age (+/- 2 1/2 years), residence and hospital from which cases had been drawn. Individual interviews were carried out for the evaluation of ethnic, environmental, pathologic and occupational risk factors. Lip cancer was significantly associated with: fair, brown, or red hair (relative odds = 2.3), blue eyes (r.o. = 5.3), fair skin (r.o. = 8.0), sensitivity to sunburns (r.o. = 4.1), working outdoors (r.o. = 4.9), coexistence of non-specific lesions of exposed body parts (r.o. = 12,2), low socioeconomic status (r.o. = 15.8), farming (r.o. = 2.6) and working in greenhouses (r.o. = 12.0). Recall of recurrent Herpes labialis was not significantly associated; also association with tobacco smoking was not significant after adjustment for socioeconomic status. Risk increased exponentially with the number of ethnic characteristics of northern European populations.

Risk factors of nonmelanoma skin cancer in Ragusa, Sicily: a case-control study.

A case-control study on 133 consecutive incident cases of nonmelanoma skin cancer registered by Cancer Registry of Ragusa (Sicily) was carried out in order to evaluate the correspondence between risk factors observed in other geographic areas with those of a Latin country with a strong solar exposure. A multivariate analysis showed that family history of skin cancer, cancer-related cutaneous diseases, fair skin color, residence more than 400 meters above sea level, and prolonged solar exposure without protection, significantly and independently increase the risk of skin cancer. Ease of suntanning was an important protective factor. Solar exposure was a stronger risk factor for squamous cell than for basal cell cancer. The results of the study enable us to identify a high risk group of people to whom simple methods of protection against solar exposure may be suggested to reduce the risk of skin cancer.

Risk factors of female cancers in Ragusa population (Sicily). 2. Breast cancer.

A case-control study on breast, cervix and corpus uteri cancer cases registered in Ragusa between January 1, 1983 and June 30, 1985 has been conducted. Information on risk factors has been obtained by means of a structured questionnaire. Risk factors for breast cancer were: few pregnancies (1-2 vs greater than 4 OR 2.14, 95% CL 1.13-4.04), few children (for postmenopausal only, chi trend 4.84), previous breast disease (OR 1.97, 95% CL 1.20-3.23), family history (OR 3.57, 95% CL 1.92-6.63), alcohol (OR 1.68, 95% CL 1.12-2.53), high socioeconomical status (1 vs 4 OR 2.93, 95% CL 1.22-70.03). A protective role was evident for: early age at first birth (for premenopausal only, less than 20 vs greater than 20 OR 0.11, 95% CL 0.01-0.90), previous ovary disease (OR 0.26, 95% CL 0.08-0.88). Age at menarche, age at menopause, years of fertile life and breast feeding were not found to be related to breast cancer. A different distribution of risk factors among cases with and without family history has been suggested. A synoptic table shows the distribution of the most important risk factors of the three investigated female cancers in the Ragusa population, reported in the present and in an accompanying paper.

Cancers of the female genital tract in Ragusa, Sicily.

A descriptive study of cancers of the female genital tract (cervix, endometrium, ovary, labia, vulva and vagina) in the province of Ragusa (Sicily) was carried out using incidence and mortality data of the Ragusa Cancer Registry, covering the years 1981-1987. Corpus uteri was the most frequent site of cancer, followed by cervix utery and ovary; the highest mortality rates were exhibited by ovary cancer, whose survival at five years was 18.1 percent. A comparison of incidence and mortality rates with those observed by Italian and European registries shows that while in Ragusa rates for cancers of all sites were lower, female genital tract cancers, particularly of the uterus, were more frequent in Ragusa than in other Italian and European areas. In contrast with what has been reported by cancer registries of most developed countries, incidence of cervix cancer in Ragusa has apparently not decreased between 1981 and 1987, while mortality has increased.

[Studies of influenza in Palermo, 1970-75]. / Indagini sull'influenza a Palermo, 1970-75.

Epidemiological surveillance of influenza was carried out in Palermo betwen 1970 and 1975. Antigenic structure of viral strains A/H3N2 isolated from influenza patients during winter outbreaks was studied in comparison with reference strains circulating in the same periods in other parts of the world. Antigenic drift of hemagglutinin, of neuraminidase, or of both, was observed in Palermo isolates, which were progressively different from Hong Kong/68 strain; 1973 and 1975 isolates were similar to A/England/72 and A/Port Chalmers/73. HI antibodies against Palermo isolates were titered in serum samples obtained from adults living in Palermo. Seronegativity for each strain progressively decreased from 1971 to 1975.

Epidemiological aspects of hepatitis B in Palermo: changes in HBV spread.

Prevalence of HBV infection in Palermo was investigated in 1989, in comparison with similar data observed in 1982. Between September 1988 and April 1989, 1001 serum samples taken from healthy individuals or from patients with pathology unrelated to liver diseases, in the age groups 0 to 59 years, were collected. Prevalence of HBsAg and anti-HBc was, respectively, 2.7% and 24.3% in males, 2.3% and 19% in females. Prevalence was inversely related to socioeconomic level as defined by occupation, educational level and area of residence. Prevalence of HBV markers in 1989 was compared with that of a sample of sera taken in 1982 and stored at -20 degrees C. Prevalence was lower in 1989 for age groups under 20, while no difference was seen in older age groups.

HBV infection risk in hospital workers.

Prevalence of HBV infection markers and its association with some risk factors has been studied on hospital staff of the University Polyclinic of Palermo. The results show that male sex, job category (technicians, nurses, cleaners) and age are significantly associated with a higher prevalence of HBV infection markers; length of service and working in departments with a presumably higher exposure to blood did not result as a risk factor of higher prevalence of HBV infection when submitted to multiple regression logistic analysis. It is suggested that results of this study may be affected by the elevated spread of HBV infection in this area, and extra risk associated to hospital exposure is too small to be demonstrated.

Risk factors of female cancers in Ragusa population (Sicily)--1. Endometrium and cervix uteri cancers.

A case-control study on breast, cervix and endometrium cancer cases registered in Ragusa between January 1, 1983 and June 30, 1985 has been conducted. Information on risk factors has been obtained by means of a structured questionnaire. Risk factors for endometrium cancer were: few children (1-2 vs greater than 4 OR 15.18, 95%CL 1.96-117.64), oestrogenic treatment (OR 2.20, 95%CL 1.05-4.90), obesity (Quetelet index greater than 30 vs less than 22 OR 10.42, 95%CL 1.30-83.86), family history (OR 2.87, 95% CL 1.05-7.83). Risk factors for cervix uteri cancer were: multiple abortions (greater than 2 vs 0 OR 9.87, 95%CL 1.46-66.66), no contraception (OR 8.33, 95%CL 2.38-25.00), younger age of mother at birth (OR 6.89, 95%CL 1.71-27.70). Age at menarche, age at menopause and years of fertile life were not found to be related to either endometrium or cervix uteri cancer. The existence of influencing differences (ancestry, environment, lifestyle) has been postulated.

Familial aggregation of tumors and detection of hereditary non-polyposis colorectal cancer in 3-year experience of 2 population-based colorectal-cancer registries.

The clinical data of 2 population-based registries, located in areas with different incidence rates of colorectal cancer, were used in order to assess the role of familial factors in the pathogenesis of these tumors. The occurrence of tumors in family members was investigated in 389 subjects with colorectal cancer registered in Modena (Northern Italy, an area characterized by a high incidence of colorectal malignancies) between 1984 and 1986; similar information was obtained in 213 patients with tumors of the large bowel registered in Ragusa (Sicily, Southern Italy, an area of similar magnitude and with low incidence rates for these tumors) in the 3-year period 1988 to 1990. In both series, colorectal cancer occurred significantly more often among relatives of patients. Controls were patients of the same sex and age (+/- 5 years) hospitalized during the study periods, but not for gastrointestinal or neoplastic diseases. There were 89 cancer cases (3.1%) among 2,851 relatives of patients in Modena, vs. 17 cases among 1,744 relatives (1.0%) in Ragusa (p < 0.01). Apart from colorectal cancer, there was no excess of other types of tumors in patients' families (in both series). During the 3 years of registration, 17 cases of hereditary non-polyposis colorectal cancer (HNPCC, or Lynch syndrome) were diagnosed in Modena; in contrast, this syndrome was more rare in Ragusa (one case only during 3 years of observation). Similarly, many more families with clinical suspicion of HNPCC were recorded in Northern regions (44 vs. 10). Although incidence rates of colorectal cancer are appreciably higher in Northern than in Southern Italian regions, the excess of this cancer type among close relatives is similar. However, full-blown HNPCC or suspected Lynch syndrome were significantly more frequent in Northern Italy.

[Observations on the diffusion of hepatitis A and B virus during an epidemic in a Sicilian town with a high rate of infection]. / Osservazioni sulla diffusione di virus epatitici A e B durante una epidemia in un comune Siciliano con alta prevalenza di infezioni.

Epidemiological investigations were carried out during a viral hepatitis outbreak occurring in a Sicilian town of 30,000 inhabitants with poor sanitary standards, with the aim to study the mode of spread of HAV and HBV in conditions of high incidence of infections. HBsAg, anti HBs, anti HAV (RIA), HBeAg, anti HBe and anti HBc were investigated in serum samples from patients, their family contacts and from healthy individuals of different age groups. Morbidity was inferred from case notifications; search of unreported cases among school children, through the study of absenteeism, did not reveal further cases. In all 148 cases, occurred from August 1976 through July 1977 with a peak in January, 44% were under 5 and 93% under 10 years of age. All but 8 of 59 cases in which laboratory data were available were due to HAV. Anti HAV antibodies were highly prevalent in serum samples obtained in February through April 1977: 62% were positive in the 1 to 3 years age group, and more than 90% among school children. Prevalence of HBsAg was age and sex dependent, ranging from 4% to 15%; anti HBs was present in 8% of the children 1-10 years and in 30% or more in age groups 30-41 and over. It is suggested that direct contact between very young children was the main mode of spread of HAV, and inapparent cases the main source of infection, although ambient diffusion through water contamination could not ruled out. HBV was probably propagated mostly by intrafamilial spread with little overt pathology.