Arsenic in community drinking water systems and small for gestational age birth, pregnancy-related hypertension, and stillbirth in Utah, 1989--2006

The purpose of this dissertation research was to investigate whether there is an association between exposure to low to moderate levels of arsenic in drinking water in community water systems (CWSs) and small for gestational age birth (SGA), pregnancy-related hypertension, and/or stillbirth. The study included over 633,000 live births and stillbirths to Utah residents during 1989 to 2006 where the maternal addresses recorded on birth and fetal death certificates were within the boundaries of a CWS. Over 97% of the maternal addresses in each county were geocoded and then spatially linked to georeferenced data layers of 476 CWS service areas statewide and to elevation data. Water quality data collected for regulatory purposes were used to estimate annual average arsenic levels for each CWS; these values were assigned to the births and stillbirths based on the first trimester of the year of pregnancy and the CWS providing water to the maternal residence. Arsenic levels were less than 2.5 micrograms per liter (mug/L) for the majority of residences (73.8%); arsenic levels were greater than 10 ig/L at only 3.7% of the residences. There was a small but statistically significant association between arsenic concentration and SGA. Using <2.5 mug/L as the reference, the adjusted odds ratio (aOR) for SGA was 1.04, (95% confidence interval (CI) 1.00, 1.07) when arsenic levels were 5.1 to 9.9 mug/L (p-value 0.03), and aOR 1.07 (CI 1.03, 1.12) when levels were 10 mug/L or greater (p-value 0.002). At arsenic levels from 2.5 to 5 mug/L, there was a small, but not statistically significant (p-value 0.40), increase in SGA (aOR 1.01, CI 0.98, 1.04). Arsenic was not found to be associated with pregnancy-related hypertension, nor was there an association between low to moderate levels of arsenic in drinking water and stillbirth. An additional finding was that, compared with births at elevations less than 3,000 feet(ft), the frequency of SGA increased with every 1,000 ft increase in elevation to an aOR of 1.91 (CI 1.65, 2.22) for women residing above 6,000 ft.