O Kravdal1. 1. Cancer Registry, Institute for Epidemiological Cancer Research, Montebello, Oslo, Norway.
Abstract
BACKGROUND: Several studies have documented significant relationships between a woman's reproductive career and her risk of developing certain types of cancer. This can be due to biological processes triggered by pregnancy or various other factors that are influenced by or are influencing family size. In this paper it is discussed how data on men can be used to gain insight into the nature of the cancer-parity relationship. METHODS: Hazard regression models for cancer incidence are estimated on the basis of register data for complete Norwegian birth cohorts. Survey data are used to assess differences in tobacco and alcohol consumption according to parity. RESULTS: It has been shown in previous studies that the effect of parity on the incidence of thyroid cancer, colorectal cancer and Hodgkin's disease is confined to women. By contrast, the present analysis demonstrates that the incidence of oral and pharyngeal cancer (379 cases) is just as strongly related to parity when the focus is on men as it is in female-oriented models. Except for the low incidence among childless men, the relationship between parity and the incidence of cutaneous malignant melanoma (4470 cases) is also similar for men and women. This suggests an association that operates through lifestyle factors, and runs counter to the hypothesis that sex hormones play an important role in the aetiology of malignant melanoma. With respect to malignancies in the oral cavity and pharynx, more frequent use of alcohol and tobacco among low-parity men and women can be intervening factor. CONCLUSION: This research suggests that future investigations into the importance of reproductive factors for women's cancer incidence should be supplemented with corresponding model estimates for men whenever possible or relevant.
BACKGROUND: Several studies have documented significant relationships between a woman's reproductive career and her risk of developing certain types of cancer. This can be due to biological processes triggered by pregnancy or various other factors that are influenced by or are influencing family size. In this paper it is discussed how data on men can be used to gain insight into the nature of the cancer-parity relationship. METHODS: Hazard regression models for cancer incidence are estimated on the basis of register data for complete Norwegian birth cohorts. Survey data are used to assess differences in tobacco and alcohol consumption according to parity. RESULTS: It has been shown in previous studies that the effect of parity on the incidence of thyroid cancer, colorectal cancer and Hodgkin's disease is confined to women. By contrast, the present analysis demonstrates that the incidence of oral and pharyngeal cancer (379 cases) is just as strongly related to parity when the focus is on men as it is in female-oriented models. Except for the low incidence among childless men, the relationship between parity and the incidence of cutaneous malignant melanoma (4470 cases) is also similar for men and women. This suggests an association that operates through lifestyle factors, and runs counter to the hypothesis that sex hormones play an important role in the aetiology of malignant melanoma. With respect to malignancies in the oral cavity and pharynx, more frequent use of alcohol and tobacco among low-parity men and women can be intervening factor. CONCLUSION: This research suggests that future investigations into the importance of reproductive factors for women's cancer incidence should be supplemented with corresponding model estimates for men whenever possible or relevant.