Drinking water disinfection by-products and duration of gestation

Appendix materials to be available online

1

Appendix 1:Details regarding estimation of personal exposure

Information on individual water use collected during the baseline and follow-up interviews was integrated with residential DBP concentrations to estimate personal DBP exposure. The estimation of personal DBP exposure was implemented in four main steps: (1) adjustment of residential DBP concentrations for water treatment (boiling and filtering), (2) estimation of the amount of DBPs ingested daily, (3) estimation of the amount of THMs inhaled/absorbed through the skin daily while showering/bathing, and (4) integration of THM exposure through ingestion and showering and bathing into an estimate of total THM exposure.

Step 1- Adjust THM and HAA residential concentrations for heating and filtration

We considered two water treatment activities that can alter DBP concentrations in tap water before ingestion: thermal treatment (e.g., heating water in the microwave or a kettle) and point of use filtration (e.g., filtering water using a home filtration system or filtration pitcher). Study collaborators previously conducted a series of experiments to derive correction factors for adjusting residential DBP levels for boiling and filtration1. Thermal correction factors for individual THM and HAA species in chlorinated and chloraminated water were also available from a study by Krasner and Wright2. These correction factors were applied to residential DBP concentrations to estimate the amount of DBP concentrations under eight different combinations of heating (cold/hot) and filtration (no filtration/faucet filtration/pitcher filtration/filtration of unknown type).

Step 2 - Estimate amount of THM and HAA exposure from to ingestion

Women were asked about their typical tap water consumption in detail during baseline and follow-up interviews. The following information was collected at baseline:

  • The number of glasses of tap water and tap water-based drinks typically consumed per day and the usual size of glasses.
  • Whether the frequency of consumption had changed over the past 4 months, and if so, how many glasses were previously consumed per day and when consumption changed.
  • The frequency and type of water filtration system a woman used to treat water before consumption.

Women were asked these questions about the number and size of tap water drinks separately for cold and hot tap water consumption. These questions were also asked separately for home and work if a woman reported working at a location outside of the study site. At follow-up interview, women were again asked about the typical number and size of glasses of cold and hot tap water they consumed per day at work and home.

To estimate the average daily amount of DBPs ingested (ounces per day), adjusted DBP concentrations estimated in step 1 were combined with interview information to estimate the amount of DBP ingested under each of the eight different water treatment scenarios. The following assumptions were made:

  • The volume of cold/hot water contained in small, medium and large glasses/cups was equal to the mid-range of ounces given to describe size categories during interview.
  • Self-reports of filtering “all or nearly all”, “most”, “some”, “very little” and “none” of water corresponded to filtering 100%, 75%, 40%, 20% and 0%, respectively, of all cold and hot water that was ingested.

Treatment-specific values were then summed to derive daily-ingested amounts in units of “micrograms/day”. If a woman reported a change in tap-water ingestion over the period of pregnancy, a time-weighted average was calculated for all periods during which a change was reported to have occurred. For changes reported between baseline and follow-up interview, that change was assumed to have occurred halfway through the period between interviews. If no change was reported, this value was simply the reported average daily amount.

Bottled water was assumed to have no DBP content and did not contribute to the estimated amount of DBPs ingested.

Step 3 - Estimate amount of THM exposure from showering and bathing

Residential THM concentrations and self-reported information on the average duration and frequency of showering and bathing were integrated using uptake factors to estimate an absorbed dose (microgram/day) through inhalation and dermal absorption. Information on the frequency and length of time typically spent showering and bathing was collected during baseline and follow-up interviews and was converted to the average minutes per day engaged in each activity. Uptake factors were obtained from previous toxicokinetic studies that measured blood THMs in relation to known duration of bathing and showering in water of known THM concentration3,4,5. For chloroform, uptake factors were assumed to be 0.001536 µg and 0.001321 µg into the blood per minute per microgram while showering and bathing, respectively. For all other THMs, uptake factors were assumed to be 0.001352 µg and 0.001296 µg of other THMs in blood per minute per microgram from showering and bathing, respectively.

Step 4- Estimate total amount of THM exposure from ingestion, showering and bathing

Estimated THM exposure through ingestion in microgram/day, derived in step 2, was multiplied by 0.00490 to estimate an absorbed dose (microgram/day) from ingestion.1,5 The estimated absorbed dose through ingestion and the estimated absorbed dose from showering and bathing, derived in step 3, were then summed to estimate total THM exposure through ingestion, showering and bathing.

  1. Savitz DA, Singer PC, Hartmann KE, et al. Drinking Water Disinfection By-products and Pregnancy Outcome. Final report for AwwaRF Project #2579. Denver, CO: American Water Works Association Research Foundation, 2005.
  2. Krasner SW, Wright JM. The effect of boiling water on disinfection by-product exposure. Water Res. 2005;39:855-864.

3.Backer LC, Ashley DL, Bonin MA, Cardinali FL, Kieszak SM, Wooten JV. Household exposures to drinking water disinfection by-products: whole blood trihalomethane levels. J Expo Anal Environ Epidemiol. 2000;10:321-326.

4.Lynberg M, Nuckols JR, Langlois P, et al. Assessing exposure to disinfection by-products in women of reproductive age living in Corpus Christi, Texas, and Cobb county, Georgia: descriptive results and methods. Environ Health Perspect. 2001;109:597-604.

5.Whitaker HJ, Nieuwenhuijsen MJ, Best NG. The relationship between water concentrations and individual uptake of chloroform: a simulation study. Environ Health Perspect. 2003;111:688-694.

1

eFigure - Predicted risk of preterm birth by second trimester average residential concentrations of total trihalomethanes (TTHMs) and five haloacetic acids (HAA5), 2000-2004.

Model adjusted for maternal age, race/ethnicity, education level, annual household income, employment status during pregnancy, marital status, pre-pregnancy BMI (kg/m2), daily caffeine intake (mg/day) and parity. Solid line is the predicted probability of preterm birth and dashed lines are lower and upper pointwise 95% confidence interval band.

1