Copyright © 2004 by the Johns Hopkins Bloomberg School of Public Health
LETTERS TO THE EDITOR |
TWO AUTHORS REPLY
1 Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle, WA 98195
2 Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA 98109
We thank Zheng (1) and Clementi et al. (2) for their comments regarding our study (3). We agree with Zheng that the relation between gravidity and spontaneous abortion likely results from heterogeneous risk and selective fertility (4). If women are likely to continue becoming pregnant until they achieve the number of living children they desire (4, 5), then it follows that those of higher gravidity will include some who are inherently at higher risk of fetal loss (and, by extension, according to Zheng’s argument, at higher risk of other adverse pregnancy outcomes such as Down’s syndrome). However, although this possibility may result in an increased risk of adverse pregnancy outcomes for those of higher gravidity, it seems much less likely to have this effect on studies examining parity. In fact, both Wilcox and Gladen (4) and El-Saadani (5) have reported that those of higher parity who had had no prior miscarriages actually had a lower incidence of spontaneous abortion. These findings led Wilcox and Gladen to conclude that "the more successful pregnancies a woman has had, the more likely she is to be a woman of low risk" (4, p. 168). Thus, although heterogeneous risk and selective fertility may explain the relation between gravidity and spontaneous abortion, we do not consider this to be a likely explanation for the positive findings of our study.
Clementi et al. (2, 6) were able to assess the relation between parity and Down’s syndrome by using high-quality data from two congenital malformation registries in Italy (NEI and ISMAC). However, their analyses are subject to some of the same biases previously reported in studies of this question. Although the NEI data were presented by using narrow (2-year) categories for maternal age, the ISMAC data were analyzed by 5-year categories because of a smaller sample size. Doing so allowed for residual confounding by maternal age (7, 8). In addition, some analyses included stillborn and electively terminated cases, which may bias any association toward the null (3).
Despite these limitations, the analyses of Clementi et al. are intriguing. For data collected by using the same methodology, in two populations in the same country, the results differed on the basis of location. In the NEI data, only those women aged 35 years or older of a parity of five or more had an increased risk; in the ISMAC sample, the increased risk associated with parity occurred in all age groups and was present for women of parity two to four. What accounts for the heterogeneity in studies of parity and Down’s syndrome? It is possible that the risk associated with parity may be modified by other exposures or that the association results from uncontrolled confounding or other bias. However, another explanation is possible. A woman’s parity is the result of a complex combination of both biology and family planning decisions. One mechanism proposed for the association is a reduced likelihood of Down’s syndrome pregnancy loss for women of higher parity (3, 8, 9); women with a proven ability to carry a pregnancy to completion (as measured by parity) may be more likely to do so even in the presence of fetal chromosomal abnormalities. Thus, parity can be considered a crude measure of a woman’s fertility, and the association would depend on women in a population choosing to reproduce. If many women elect to have no (or few) children, then parity ceases to be a good measure of fertility, and the association would be blunted, which could explain a weaker association between parity and Down’s syndrome in some populations. We agree with Clementi et al. (2) that a better knowledge of the underlying mechanisms is crucial to our understanding of this question.
REFERENCES
- Zheng CJ. Re: "Parity and the risk of Down’s syndrome." (Letter). Am J Epidemiol 2004;160:609–10.
[Free Full Text] - Clementi M, Bianca S, Tenconi R. Re: "Parity and the risk of Down’s syndrome." (Letter). Am J Epidemiol 2004;160:610.
[Free Full Text] - Doria-Rose VP, Kim HS, Augustine ET, et al. Parity and the risk of Down’s syndrome. Am J Epidemiol 2003;158:503–8.
[Abstract/Free Full Text] - Wilcox AJ, Gladen BC. Spontaneous abortion: the role of heterogeneous risk and selective fertility. Early Hum Dev 1982;7:165–78.[CrossRef][Web of Science][Medline]
- El-Saadani S. High fertility does not cause spontaneous intrauterine fetal loss: the determinants of spontaneous fetal loss in Egypt. Soc Biol 2000;47:218–43.[Medline]
- Clementi M, Bianca S, Benedicenti F, et al. Down syndrome and parity. Community Genet 1999;2:18–22.[CrossRef][Medline]
- Castilla EE, Paz JE. Parity and Down’s syndrome. (Letter). Lancet 1994;344:1645–6.[Web of Science][Medline]
- Lilford RJ. Commentary: Down’s syndrome and parity. BMJ 1997;314:721.
[Free Full Text] - Pharoah PO. Increased parity and the risk of trisomy 21: study measured prevalence of Down’s syndrome at birth, not incidence. (Letter). BMJ 1997;314:1760–1.
[Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||