Council on Environmental Health
Article Review

Prepared by: Mark A. Anderson, MD, MPH
September 2008

TITLE
Drinking Water Disinfection By-Product Exposure and Fetal Growth

AUTHORS
Caroline S Hoffman, Pauline Mendola, David A Savitz, Amy H Herring, Dana Loomis, Katherine E Hartmann, Philip C Singer, Howard S Weinberg, Andrew F Olshan

REFERENCE
Epidemiology, September 2008, 19(5): 729-737

STUDY DESIGN
Prospective cohort

BACKGROUND
Exposure to drinking water disinfection by-products (DBPs) during pregnancy has been linked to adverse pregnancy outcomes. Earlier studies describe a moderately increased risk of delivering a small-for-gestational-age (SGA) infant among women exposed to high levels of total trihalomethanes (TTHMs). It is not clear what the mechanism for this relationship might be; it is also unclear if TTHMs as a whole, or some constituent species, is responsible.

Previous studies of DBP exposure have been limited by reliance on retrospective data on DBP concentration obtained from regulatory databases, which do not account for individual variation in tap water uptake. This study had two objectives: 1) to examine the association between DBP measures and fetal growth using improved exposure data that would account for variations in exposure between individuals; and, 2) to examine the effect of exposure to individual TTHMs and haloacetic acid species on fetal growth.

METHODS
Women who were either planning to become pregnant or were early in pregnancy (≤12 weeks gestation) were recruited between 2000 and 2004. The women were from 3 communities that included one site with moderate levels of predominately chlorinated DBPs, one site with moderate levels of predominately brominated DBPs, and one site with low levels of all DBPs. A total of 2766 pregnancies were enrolled; analysis was restricted to 1854 live births (≥37 weeks gestation).

Weekly water samples were collected from the chlorinated and brominated DBP sites and biweekly from the low DBP site. Concentrations of the 4 individual TTHM species, 9 haloacetic acid species and total organic halide were assessed. The 4 TTHM species were summed to calculate TTHM; 5 haloacetic acid species were also summed (HAA5). DBP exposure was characterized based on the estimated tap water concentrations (the concentration in the water system serving a woman’s residence) and the estimated personal exposure (estimated residential concentrations and detailed self-report data on tap water exposure).

Information on fetal growth was obtained from medical records, vital records and self-report (<1%). SGA was defined as a birth weight <10th percentile for gestational age at birth.

RESULTS

  • The probability of delivering an SGA infant was greater among women with an average 3rd trimester residential TTHM concentration ≥80µg/L compared to women with exposure <80µg/L (risk ratio=2.0 [95% CI=1.1-3.6]).
  • Haloacetic acids and total organic halide were not associated with SGA or term birth weight.
  • A dose-response in birth weight and TTHM, HAA5, or total organic halide was not found.
  • No consistent association was found between any DBP species and fetal growth.

DISCUSSION
The findings of this study did not suggest an adverse effect of TTHM or HAA5 exposure on fetal growth at residential concentrations below the current regulatory standards. Similarly, the findings did not suggest a negative effect of haloacetic acid or total organic halide exposure on fetal growth.

Two major strengths of this study were the concurrent measurement of DBP concentrations over the course of pregnancy and the fact that pregnant women were prospectively followed. The authors also note that their selection process resulted in overrepresentation of non-Hispanic whites and, consequently, their results may underestimate of the effect of DBP exposure in the general population.

BOTTOM LINE
This study was the most extensive study of DBP exposure and fetal growth conducted to date. The results do not indicate an adverse effect of residential TTHM or HAA5 levels with the regulatory limits on fetal growth. Constituent acid species were not associated with fetal growth restriction either.

Return to COEH Page