Breast cancer risk is reduced by early cohabitation but not by young age at first birth
Early case-control studies reported a strong association of breast cancer risk with older age at first marriage, which at that time was a good surrogate for age at beginning cohabitation. A subsequent study in 1970 concluded, despite other possible explanations, that the apparent effect of age at first marriage was entirely due to its strong correlation with age at first birth. This has been the consensus for the past 50 years.
The effect of age of first marriage was noted again recently, leading us to re-analyse the data from a large international study conducted over 10 countries from 1979 to 1982. We estimated the independent effects of age at first marriage and age at first birth adjusted for parity and other risk factors in 2,724 breast cancers and 18,209 controls.
We conclude that age at first marriage, a surrogate for age at beginning cohabitation, is strongly associated with breast cancer risk. The effect of age at first birth, a long-accepted breast cancer risk factor, appears to be due largely – perhaps entirely – to confounding.
In this seminar, we discuss this surprising conclusion and suggest a potential underlying mechanism for breast cancer. We welcome a lively discussion of this and the role of confounding and inferences regarding causality.
Details of our analysis are now online in the International Journal of Cancer.