Am J Epidemiol 2003; 157:665-666.
Copyright © 2003 by Johns
Hopkins Bloomberg School of Public Health
LETTERS TO THE EDITOR |
RE: "BIRTH SIZE, GROWTH, AND BLOOD PRESSURE BETWEEN THE AGES OF 7 AND 26 YEARS: FAILURE TO SUPPORT THE FETAL ORIGINS HYPOTHESIS"
Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, Boston, MA 02215
In a recent issue of the Journal, Williams and Poulton concluded that their analysis of birth weight, adult weight, and blood pressure failed to support the "fetal origins hypothesis" (1). In fact, they misinterpreted their own statistical models: their results showed clear support for an inverse association of size at birth with later blood pressure.
Williams and Poulton (1) were attempting to respond to the challenge thrown down by Lucas et al. (2), who argued that it is necessary to include an interaction term between birth weight and adult weight to distinguish fetal from postnatal growth. Williams and Poulton constructed such a model from their Dunedin, New Zealand, data. In their "interaction model," a standard deviation increase in birth weight was associated with a –0.96 mmHg change in systolic blood pressure (95 percent confidence interval: –1.38, –0.54). The interaction between birth weight and adult weight accounted for a –0.32 mmHg change (95 percent confidence interval: –0.69, 0.05). Of note, the addition of this interaction term made no change to the birth-weight coefficient from the previous "combined" model that included birth weight and adult weight (95 percent confidence interval: –1.38, –0.53).
Williams and Poulton (1) mistakenly chose to interpret the interaction term, rather than the birth-weight term, as a measure of the association of fetal growth with blood pressure. In fact, it is the birth-weight term that represents the main effects of fetal growth. If we discount the interaction (because it fell short of conventional statistical significance), a standard deviation increase in birth weight was associated with a –0.96 mmHg change in blood pressure, regardless of adult size. If we choose to accept the interaction term (because it was close to statistical significance), then the increase in birth weight was associated with a –0.64, –0.96, and –1.28 mmHg change in blood pressure at –1, 0, and 1 standard deviations of adult weight, respectively.
In other words, Williams and Poulton’s data (1) show that birth weight is inversely associated with adult blood pressure whether or not there is an interaction between birth weight and later weight. Rather than refuting the fetal origins hypothesis, the results of their analysis support the notion that fetal growth is inversely associated with adult blood pressure.
REFERENCES
- Williams S, Poulton R. Birth size, growth, and blood pressure between the ages of 7 and 26 years: failure to support the fetal origins hypothesis. Am J Epidemiol 2002;155:849–52.
[Abstract/Free Full Text] - Lucas A, Fewtrell MS, Cole TJ. Fetal origins of adult disease—the hypothesis revisited. BMJ 1999;319:245–9.
[Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||