The aim of this study was to examine available data on breast cancer and age at first marriage from a new perspective: that is, marriage involves the closest contact and contact effects are relevant to the question of infection, a possibility long considered in this disorder. The large Seven Country Study, carried out in 1964–1968, investigated age at first marriage; its reports were examined carefully for details of possible relevance. Intriguing gaps were noted in the grounds for the conclusion by this study that late age at first birth explained an earlier reported association with late age at marriage, with risks presented by age at first marriage for nulliparous, but not for parous, married women. Only in one centre, Glamorgan Wales, and only for two age groups could risks by combined ages at first marriage and first birth be derived. When both events occurred at age 30 or older, the risk estimate was 7.0 (95% confidence interval: 5.2, 9.1) relative to when both events occurred younger than age 20, whereas the corresponding risk was 1.4 (95% confidence interval: 1.1, 1.8) when age at first birth was 30 or older but marriage was younger than age 30. The above findings are consistent with an effect of age at first marriage, and a basis in contacts or infection is considered plausible. However, other explanations may exist, and this report primarily aims to encourage examination of the subject in other datasets, particularly where intersexual contrasts in infective exposures have probably existed.
Cancer Epidemiology Unit, Nuffield Department of Medicine, University of Oxford, Oxford, UK
Correspondence to Leo J. Kinlen, MBBS, PhD, Cancer Epidemiology Unit, University of Oxford, Richard Doll Building, Roosevelt Drive,Headington, Oxford OX3 7LF, UK Tel: +44 1865 289600; fax: +44 1865 289610; e-mail: email@example.com
Received February 8, 2013
Accepted March 29, 2013