|
|
George Alter, University of Michigan
Julia Jennings, University at Albany
Gilles Pison, Institut national d'études démographiques (INED)
This paper explores historical patterns of twin births to advance our understanding of the genetic and biological factors underlying fertility. Although women cannot choose whether they will have a multiple birth, twinning rates vary in important ways across time and place (Pison, Monden, and Smits 2015). Twinning is known to vary by age of mother, and genetic differences appear to play a role in regional patterns. Nutrition may also play a role in the regulation of twinning, especially during breastfeeding (Bortolus et al. 1999; Smith and Grove 2002). Dizygotic twinning has been linked to key hormones controlling ovulation, and there are strong indications that women prone to twin births conceive more quickly and have shorter birth intervals than other women (Ferrari et al. 2007). Women who are prone to twinning may also give birth to more sons, and the sex composition of twins has been linked to infant mortality (Ouedraogo et al. 2025). In this paper we test a model predicting how twin-prone women will respond to the death of a breastfeeding infant (Alter and Pison 2008). Our model predicts that twin-prone women will have shorter than average birth intervals following an infant death, but early births will be singletons rather than twins. It also explains other patterns, such as the peak in twinning after World War I (Pison and D'Addato 2006). We will look for evidence that the genetic characteristics producing twinning are inherited in the female line. Research on twins requires a large database, because less than two percent of births are twins. We take advantage of the LINKS database of births, marriages, and deaths in the Netherlands (Mandemakers et al. 2023).
No extended abstract or paper available
Presented in Session 5. Fertility in Changing Contexts