Document Type



This study tests the hypothesis that mothers of dizygotic (DZ) twins experience periodic ovarian dysfunctions leading to a slow rate of depletion of oocytes and culminating in delayed age at menopause. The data used are population-based. The time interval (TI) between the twin and the last birth is the independent variable, the dependent variable being the risk of a twin birth. The methodology uses a case/control and a pair-matched design. Maternal cohort and age, and the sex and birth order of offspring were controlled. Results show an interaction between maternal age and TI, and the type of twin birth; specifically, the risk of a DZ twin birth born to a mother aged 30 years or more increases with TI. Other results show the unexpected finding that the risk of a like-sex (presumably monozygotic) twin birth bom to a mother aged less than 30 years is inversely related to TI. It is concluded that late menopause in older mothers of DZ twins is likely, and cannot be ascribed to differences in maternal cohort or birth order. The potential mechanism of slow oocytes depletion with age is discussed. Because the finding of a short TI associated with like-sex twinning at younger maternal age was merged by induction, further testing must be considered. However, premature ovarian decline is a likely etiologic mechanism of the finding.