AbstractBACKGROUNDScientists disagree on whether prenatal malnutrition has long-term influences on women's reproductive function, and empirical evidence of such long -term effects remains limited and inconsistent.METHODSUsing the retrospective pregnancy history of 12,567 Chinese women collected in a nationally representative sample survey in 2001, this study conducted difference-in-differences analyses to investigate the relationship between prenatal exposure to the 1959-1961 Great Leap Forward Famine in China and the subsequent risk of involuntary foetal loss, including miscarriage and stillbirth, and how this relationship changes between the rural and urban populations.RESULTSPrenatal exposure to the Great Leap Forward Famine had no long-term effect on women's risk of miscarriage. Such an exposure increased the risk of stillbirth among urban women but not among rural women.CONCLUSIONSThe results support the foetal origins hypothesis. The significant urban-rural difference in the effect of prenatal famine exposure on stillbirth suggests the presence of a long- term negative foetal origins effect and a strong selection effect caused by famine-induced population attrition.(ProQuest: ... denotes formulae omitted.)1. IntroductionScientists disagree on whether prenatal malnutrition has long-term influences on subsequent reproductive outcomes. On the one hand, the 'foetal origins' hypothesis and its straightforward extensions predict that prenatal malnutrition has a long-term negative influence on the female reproductive system by affecting the in utero development of the female foetus' organs that are responsible for the production and regulation of reproductive hormones (Drake and Walker 2004; Gardner, Ozanne, and Sinclair 2009; Lumey and Stein 1997; Lummaa 2003). On the other hand, based on the evolutionary life-history regulation theory, if prenatal malnutrition is associated with impaired health later in life, as predicted by the foetal origins hypothesis, an enhanced reproductive function is likely to present as a life-history compensation (Bateson et al. 2004; Painter et al. 2008). Both arguments are based on sound scientific theories; therefore the final adjudication depends on which argument is better supported by empirical evidence. Given the rampant poverty and malnutrition and high prevalence of involuntary foetal loss in many developing countries, testing the relationship between prenatal malnutrition and subsequent involuntary foetal loss risk is not only of important scientific value but also of significant policy relevance.Due to the legal and ethical restrictions on the use of human subjects in scientific research, empirical evidence on the relationship between prenatal malnutrition and subsequent reproductive function has mainly been derived from observational studies. A number of such studies showed that women's prenatal conditions, as measured by their birth weight, are negatively associated with their subsequent risk of foetal loss and preterm birth, and positively associated with babies' birth weight and chance of survival (Emanuel et al. 1992; Hackman et al. 1983; Klebanoff, Meirik, and Berendes 1989; Little 1987; Sanderson, Emanuel, and Holt 1995; Skjaerven, Wilcox, and Magnus 1997). However, due to the weaknesses of observational design, the causal nature of these findings remains unclear. More specifically, the observed association between women's birth weight and their subsequent reproductive outcomes may be the result of the causal mechanism articulated by the foetal origins hypothesis or the result of inadequate control of the common determinants of both women's birth weight and their reproductive outcomes that cannot be directly observed (Clarke 2005; Murnane and Willett 2010).Famine provides an 'experiment-like' opportunity to address this important question. Similar to a randomised experiment, the assignment of the famine 'treatment' condition (i. …