Evaluation of contaminated drinking water and preterm birth, small for gestational age, and birth weight at Marine Corps Base Camp Lejeune, North Carolina: a cross-sectional study

BACKGROUND Births during 1968-1985 at Camp Lejeune were exposed to drinking water contaminated with trichloroethylene (TCE), tetrachloroethylene (PCE), and benzene. METHODS We conducted a cross-sectional study to evaluate associations between residential prenatal exposure to contaminated drinking water at Camp Lejeune during 1968-1985 and preterm birth, small for gestational age (SGA), term low birth weight (TLBW), and mean birth weight (MBW) deficit. Birth certificates identified mothers residing at Camp Lejeune at delivery. We analyzed exposure data for the entire pregnancy and individual trimesters. For each period examined, births were categorized as unexposed if mothers did not reside at Camp Lejeune or if their residence on base received uncontaminated drinking water. Ground water contaminant fate/transport and distribution system models provided monthly estimated contaminant levels at residences. For PCE and TCE, the exposed group was divided into four levels: < median value, ≥ median value, ≥75th percentile, and ≥90th percentile. For benzene, the exposed group was categorized as <1 part per billion (ppb) versus ≥1 ppb because of sparse data. Magnitude of effect estimates and exposure response relationships were used to assess associations. Confidence intervals (CIs) indicated precision of estimates. RESULTS For the highest TCE exposure category during the entire pregnancy, odds ratios (ORs) were 1.5 (95% CI: 1.2, 1.9) and 1.3 (95% CI: 0.8, 2.2) for SGA and TLBW, respectively, and reduced MBW β = -78.3 g (95% CI: -115.0, -41.7). The OR =1.3 (95% CI: 1.0, 1.6) for preterm birth and the highest PCE exposure category during the entire pregnancy. Monotonic exposure-response relationships were observed for benzene exposure during the entire pregnancy and TLBW (highest category OR =1.5, 85% CI: 0.9, 2.3). Although a monotonic association between benzene and adjusted MBW difference was also observed (highest category β = -36.2 g, 95% CI: -72.3, -0.1), the association disappeared when TCE was also added to the model. We found no evidence suggesting any other associations between outcomes and exposures. CONCLUSION Findings suggested associations between in utero exposures to TCE and SGA, TLBW and reduced MBW; benzene and TLBW; and PCE and preterm birth.