© 1990 Oxford University Press
research-article |
Cancer Incidence and Risk Factors among Montreal Residents of Italian Origin

*Dipartimento di Scienze Biomediche e Oncologia Umana, Università di Torino Via Santena 7, Torino 10126, Italy
**Epidemiology and Preventive Medicine Research Centre, Institut Armand-Frappier 531 blvd. des Prairies, Laval-des-Rapides, Quebec, H7V 1B7 Canada
Epidemiology and Biostatistics, McGill University Montreal, Quebec, Canada
Terracini B (Dipartimento di Scienze Biomediche e Oncologia Umane, Università di Torino, Via Santena 7, Torino 10126, Italy), Siemiatycki J and Richardson L. Cancer incidence and risk factors among Montreal residents of Italian origin. International Journal of Epidemiology 1990; 19: 491497.
Cancer risks associated with Italian ethnicity were investigated using data from a large case-control study on the aetiology of several cancer sites in the male population of Montreal. Two approaches were taken. First, incidence rates were computed for Italians and for others in the Montreal area using our ascertained cases as numerators and census-derived denominators. Secondly, for respondents to the case-control study, an analysis was carried out with Italian ethnicity as the risk factor and a number of covariates as potential confounders. Out of 4553 incident cases in men aged 3569, 301 were of Italian origin. As compared with other Montreal males, those of Italian origin had higher incidence rates for cancers of the stomach (p = 0.016, based on 31 cases) and of the colon and rectum (p = 0.102, based on 75 cases) and lower rates for cancer of the lung (p = 0.006) and prostate (p = 0.102, based on 24 cases). For other sites the differences between Montrealers of Italian and non-Italian origins were small. Montreal Italians manifested risks similar to those of the country of origin for cancer of the prostate and similar to the host country for cancers of the colon, rectum and liver. For other sites it was difficult to characterize the pattern because of wide variations among Italian registries. Over 80% of the study subjects in Montreal were interviewed and odds ratios (OR) for Italian ethnicity were estimated for each cancer site using all other sites as controls, adjusting for five potential confounders. The excess stomach cancer risk among Italians was attributable to two variables: socioeconomic status and an indicator of urban or rural upbringing. By contrast, the highly significant deficiency of lung cancer among Montreal Italians could not be attributed to any of the covariables analysed, including tobacco consumption.
Revised 1 February 1990