Maternal preconception diet and the sex ratio

Maternal preconception diet and the sex ratio www.feb.uva.nl/ke/UvA-Econometrics

Amsterdam School of Economics
Department of Quantitative Economics
Roetersstraat 11
1018 WB AMSTERDAM
The Netherlands
Maternal preconception diet and the sex ratio
Key words sex ratio, nutrition, famine
Abstract
Temporal variations in the sex ratio or the ratio of boys over girls at birth have been widely studied and variously attributed to social changes, conditions of war, and environmental changes. Recently, Mathews, Johnson and Neil (2008) studied the direct evidence of individual pregnancies and established an association between the sex at birth and the mother’s preconception diet. We examined the hypothesis using new evidence from the wartime famine in Holland in 1944/45 and failed to show an association between maternal diet in pregnancy and the sex ratio. This makes a causal link highly improbable. 1 University of Amsterdam, Department of Quantitative Economics, Roetersstraat 11, 1018 WB Amsterdam, the Netherlands 2 Department of Epidemiology, Mailman School of Public Health, Columbia University, 722 West 168 Street, New York, NY 10032, USA 3 Lorentz Fellow, Netherlands Institute for Advanced Study in the Humanities and Social Sciences, 2242 PR Wassenaar, the Netherlands * Author for correspondence [email protected] Introduction
In most human populations there is a small excess of boys over girls at birth, or a sex ratio slightly over 1 (Russel 1936). In several industrialized countries this sex ratio has been falling over a long period (Parazzini, La Vecchia, Levi et al 1998, James 2000) and the reasons for this are not well understood. Trends in Denmark since 1850 and in the Netherlands since 1950 have been interpreted in terms of increasing reproductive hazards, such as environmental estrogens, but long-term developments in Germany have been attributed to other factors like nutrition or social changes (Moller 1996, van der Pal, Verloove-Vanhorick and Roeleveld 1997, Bromen and Jockel 1997). Then there are a number of studies that suggest an increase in male births in conditions of war (Graffelman and Hoekstra, 2000; and for the Netherlands, van den Broek 1997). A recent report by Matthews, Johnson and Neil (2008) that links the sex ratio to the maternal preconception diet is therefore of particular interest, the more so as the authors make use of direct evidence from individual pregnancies. Matthews, Johnson and Neil have interviewed women attending an antenatal clinic in the south of England in the 1990s and collected 721 retrospective reports of the usual diet prior to conception, using a food frequency questionnaire. The preconception dietary intake of women delivering a boy were higher than of women delivering a girl (with averages of 2413 and 2283 kcal/day respectively), and 45% of women in the lowest third of food intake had boys against 56% in the highest third. These differences are statistically significant. There was no association with the diet in early pregnancy. The authors interpret this association as a causal link, and suggest inter alia that the secular decline in the sex ratio in industrialized countries may be due to slimming diets of young women. We here examine the hypothesis that a mother’s diet around conception and in pregnancy has a direct effect on the sex ratio with new data for the Dutch famine during the winter of Data and methods
We use new data for the Dutch hunger winter of 1944/45, viz. the monthly birth statistics for the city of Amsterdam from January 1938 to December 1948, as published by the Municipal Bureau of Statistics of that city (Gemeente Amsterdam 1938-1948). The data represent a large population living under uniform (but varying) conditions; with between 1,000 and 1,200 monthly births in normal years, and about 600 in the worst months of the famine. The study period includes the war years, and in particular the famine of the 1944/45 winter, during the last months of the German occupation, when food rations in Amsterdam dropped dramatically to a level of only 500 kcal/day (Trienekens 2000). In order to trace the effects of war and famine, we examine the monthly births in Amsterdam for five distinct periods of conception, reflected by births nine months later. These are the prewar period , the war period without famine, the war period with famine, and the early postwar and later postwar periods. The period of war without famine starts with the German invasion of the Netherlands in May 1940, the famine period in Amsterdam starts in November 1944, the early postwar period starts with liberation in May 1945, and the later postwar period starts six months thereafter. The precise delimitation of these periods is given in Table 1. The monthly data on fertility and the sex ratio are shown in Figures 1 and 2, and the mean Figure 1 shows a dramatic decline in fertility during the famine and an equally dramatic increase after the end of the war and of the German occupation. While it is known that in the course of the war conceptions responded to major events and the mood of the population (van den Brink 1948), the present famine also caused widespread amenorrhoea among women (Burger, Drummond and Sanstead 1948, Stein, Susser, Saenger et al 1975). But while the sex ratio is quite variable from month to month (as Figure 2 shows), its average level stays remarkably constant (as Table 1 shows) and is not affected by the condition of war, nor by the severe malnutrition of the famine months. It is easy to test the hypothesis that the sex ratio or the proportion of male births is the same for all five subperiods that we have distinguished. In each subperiod sample, given in the third column of Table 1, the number of male births has a binomial distribution, and this permits a likelihood ratio test of the hypothesis that the proportion of male births is the same for all five. This gives a test statistic LR = .64 for a chi-square distribution with 4 degrees of freedom. The hypothesis is Discussion
Our results confirm the findings from two previous studies of the effects of the Dutch hunger winter, viz. the analysis of monthly birth data for six cities (Stein, Susser Saenger et al 1975, Appendix Table 1) and the study of 3,300 selected births in three birth clinics (Stein, Zybert and Lumey 2004). These results are also in line with recent findings from Africa (Stein, Barnett and Sellen 2004). In all these studies, the sex ratio is not affected by famine, in contradiction to Matthews’ hypothesis. Neither is there any evidence of a ‘war effect’, as postulated by van den Broek 1997 for annual births in the entire Netherlands. We can therefore dismiss the hypothesis that potential effects of malnutrition towards a lower sex ratio could be offset by a contrary effect of war conditions. These results do not invalidate the finding of Matthews, Johnson and Neill (2008) that in their data there may be an association between maternal nutrition and the sex of the child. They do however contradict any causal interpretation of that association. References
Bromen, K. and K.-H. Jockel 1997. Change in male proportion among newborn infants. Lancet 349, 804-5.
Burger, G.C.E., J.C. Drummond and H.R. Sandstead 1948. Malnutrition and starvation in western Netherlands. September 1944 to July 1945, parts I and II. ‘s-Gravenhage, Gemeente Amsterdam 1938-1948. Monthly Reports of the Bureau of Statistics. [Maandbericht van het Bureau van Statistiek der Gemeente Amsterdam] Municipality Graffelman, Jan, and Rolf T. Hoekstra 2000. A statistical analysis of the effect of warfare on the human secondary sex ratio. Human Biology 72, 433-445.
James, W.H. 2000. Secular movements in sex ratios of adults and of births in populations during the past half-century. Hum. Reprod. 15, 1178-1183.
Mathews, F., P.J. Johnson, and A. Neil 2008. You are what your mother eats: evidence for maternal preconception diet influencing foetal sex in humans. Proc. R. Soc. B 275,
Moller, H. 1996. Change in male:female ratio among newborn infants in Denmark. Lancet 348, 828-9.,
Parazzini, F., C. La Vecchia, F. Levi and S. Franceschi 1998. Trends in male-female ratio among newborn infants in 29 countries from five continents. Human Reproduction 13,
Russell, W.T. 1936. Statistical study of the sex ratio at birth. J.Hyg. (Cambridge) 36, 381-401.
Stein, Z., M. Susser, G. Saenger, and F. Marolla 1975. Famine and Human Development: The Dutch Hunger Winter of 1944-45. Oxford University Press, London and Toronto.Stein, A.D., P.G. Barnett and D.W. Sellen 2004. Maternal undernutrition and the sex ratio at birth in Ethiopia: evidence from a national sample. Proc. R. Soc. Lond. B. (Suppl) 271,
Stein, A.D., P.A. Zybert and L.H. Lumey 2004. Acute undernutrition is not associated with excess of females at birth in humans: the Dutch Hunger Winter. Proc. R. Soc. Lond. B. (Suppl) 271, 138-141.
Trienekens, G. 2000. The food supply in the Netherlands during the Second World War, in Food, science, policy and regulation in the twentieth century. International and comparative perspectives, D.F. Smith and J. Phillips, Editors., Routledge: London. p. Van den Brink, T. 1949. Eerste resultaten van een analyse van de loop der geboortecijfers in Nederland. Amsterdam, Instituut voor Sociaal Onderzoek van het Nederlandse Volk. Van den Broek, J.M. 1997. Change in male proportion among newborn infants. Lancet 349,
Van der Pal, .M., S.P. Verloove-Vanhorick and N. Roeleveld 1997. Change in male:female ratio among newborn infants in Netherlands. Lancet 349, 62.

Table 1. Amsterdam 1938-1948, monthly births by selected conception periods
Period Date

Mean sex ratio and mean % male are means of monthly values for the indicated period; the standard
deviations of these means have been derived from the variation of the monthly values.
Figure 1. Amsterdam 1938-1948: daily number of births by month for five periods of conception (see Table 1). Figure 2. Amsterdam 1938-1948: sex ratio by month for five periods of conception (see Table 1).

Source: http://aseri.uva.nl/binaries/content/assets/subsites/amsterdam-school-of-economics-research-institute/uva-econometrics/dp-2009/0911.pdf