Registration Dossier

Administrative data

Endpoint:
epidemiological data
Type of information:
migrated information: read-across from supporting substance (structural analogue or surrogate)
Adequacy of study:
weight of evidence
Study period:
1998-2002
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Relatively low number of cases. Exposures were estimated and quality of these estimates were difficult to quantify. Lack of data on infertility treatment which may be a risk factor for hypospadias.

Data source

Reference
Reference Type:
publication
Title:
Maternal exposure to water disinfection by-products during gestation and risk of hypospadias
Author:
Luben et al
Year:
2007
Bibliographic source:
Occup Environ Med 2008;65:420-427.

Materials and methods

Study type:
case control study (retrospective)
Endpoint addressed:
developmental toxicity / teratogenicity
Test guideline
Qualifier:
no guideline followed
Deviations:
not applicable
Principles of method if other than guideline:
The use of chlorine for water disinfection results in the formation of disinfection by-products (including MCA), which may be associated with birth defects, including urinary tract defects. Birth records were used to conduct a population-based case-control study investigating the relationship between hypospadias and two classes of OBPs, trihalomethanes and haloacetic acids, including MCA. ORs were calculated for hypospadias and estimated exposure to MCA between 6 and 16 weeks' gestation, also taking into account personal water consumption.
GLP compliance:
no

Test material

Reference
Name:
Unnamed
Type:
Constituent
Details on test material:
Haloacetic acids in drinking water.

Method

Type of population:
general
Ethical approval:
other: waived under 45 CFR 46.116d.
Details on study design:
Study was approved by the Human Research Committee, Office of Regulatory Compliance, at Colorado State University, USA.

Birth certificate and birth defect registry data for 647 cases of hypospadias occurring in Arkansas between 1 January 1998 and 31 December 2002 identified by the Arkansas Reproductive Health Monitoring System (ARHMS) were used to conduct a population-based case-control study investigating the relationship between hypospadias and two classes of OBPs, trihalomethanes (THM) and haloacetic acids, including MCA. Monitoring data, interpolation of mean quarterly monitoring data and geographical information systems were used to link daily concentrations from 263 water utilities to 320 cases and 614 controls. ORs were calculated for hypospadias and exposure to MCA between 6 and 16 weeks' gestation.

Arkansas participants in the NBDPS were evaluated in subset analyses to explore the potential effects of exposures to MCA based only on an estimate of water consumption (total 40 cases and 242 controls).
Exposure assessment:
estimated
Details on exposure:
TYPE OF EXPOSURE: via drinking water.

TYPE OF EXPOSURE MEASUREMENT: other:
Estimated based on monitoring data of water utilities and geographical information systems (details specified in attached document).

EXPOSURE LEVELS: drinking water: mean: 2.3 micrograms/L; median: 2.5 micrograms/L; SD: 2.6 micrograms/L; maximum concentration: 33.1 micrograms/L.
Exposure estimates were obtained by averaging the daily MCA concentrations for each case and control mother during an exposure window, defined as the period between weeks 6 and 16 of gestation, as calculated using the date of the last menstrual period reported by the mother on the birth certificate.
Exposure via drinking water was estimated using various exposure metrics.

EXPOSURE PERIOD: birth records between 1998-2002 were used. Weeks 6 to 16 of gestation were used as the exposure window in this study.

POSTEXPOSURE PERIOD: not applicable.

DESCRIPTION / DELINEATION OF EXPOSURE GROUPS / CATEGORIES:
647 cases of hypospadias occurring in Arkansas between 1 January 1998 and 31 December 2002 identified by the Arkansas Reproductive Health Monitoring System (ARHMS). Subset analyses were conducted using data for ARHMS subjects who also participated in the National Birth Defects
Prevention Study (NBDPS), an ongoing population-based case control study.
Controls (n = 1264) were randomly selected from birth certificate records for all male live births in Arkansas during the study period.
Birth certificate records of infants without mention of a congenital malformation were used to randomly select two controls for each case, frequency matched on race.
Exposures were categorised into quartiles based on the distribution of concentrations among controls.

Data for potential confounders were abstracted from birth certificate records for cases and controls and included maternal and paternal age, race and education, marital status, maternal use of alcohol and tobacco, parity, prenatal care, 1 and 5 min Apgar scores, birth weight, method of delivery, and any risks, procedures or complications associated with delivery.
Statistical methods:
Logistic regression was used to obtain crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the relationship between hypospadias and exposure to MCA. Exposures were categorised into quartiles based on the distribution of concentrations among controls.

Potential confounders were accounted for (see details on exposure). Potential confounders that were statistically significant at the
p<0.10 level or that changed the odds ratio of the covariate by 10% or more were retained in the models.

A backwards selection method was used to create multivariate models. For analyses based on the subset of cases and controls from
the NBDPS, exposure was analysed by tertiles based on the controls due to the smaller sample size available for this subset of participants.

Results and discussion

Results:
There was no evidence of an association between risk for hypospadias and concentration of MCA when the analysis was based on concentrations from routine water utility monitoring.

When water consumption was included in the exposure assessment there was a pattern of increased risk for exposures in the intermediate tertiles (see also attached full study report), which suggests that exposure misclassification may have occurred between tertiles when only ingestion was taken into account, distorting a dose-response trend. The results of the analyses were based on the small number of cases and controls with exposure data during the 10-week exposure window which may not be representative of the results that might have been obtained had the total sample of cases and controls been included. These results may also have occurred due to chance.

Confounding factors:
- Little information on infertility treatment (which may be a risk factor for hypospadias).
- Potential for confounding exposures before/during window of exposure assessed in this study.
Strengths and weaknesses:
- Small sample size; low power for statistical analyses of the subset participants due to the small number of hypospadias cases in the NBDPS that could be linked to disinfection by-product concentrations (including MCA) in their exposure window; at 80% power and alpha= 0.05 adequate power to detect a risk estimate of 3.0 in analyses utilising the NBDPS population of 40 cases and 243 controls was obtained.
- Exposures were estimated.
- Limited ability to link subjects to utilities and DBP data for exposure assessment; a high level of success in geocoding participants' residences (97%) was obtained but ability to link the geocoded addresses to a water utility was limited (70%).
- Limited information on water consumption behaviours to allow estimation of more specific exposure measures among the ARHMS participants, and limited information to assess the quality of the exposure measures for all participants.
- Potential selection bias from case under-ascertainment and from systematic differences in characteristics between participants for the two
study groups; no statistically significant differences in maternal age and race between ARHMS and NBDPS participants were obtained. NBDPS participants were more highly educated than ARHMS participants but there was no evidence that educational level was related to exposure to DBPs based on water consumption. Thus this difference should not introduce bias.

Any other information on results incl. tables

See attached full study report.

Applicant's summary and conclusion

Conclusions:
Birth records were used to conduct a population-based case-control study investigating the relationship between hypospadias and two classes of OBPs, trihalomethanes and haloacetic acids, including MCA. ORs were calculated for hypospadias and estimated exposure to MCA between 6 and 16 weeks' gestation, also taking into account personal water consumption.

The results provide little evidence for a positive relationship between MCA exposure during gestation and an increased risk of hypospadias, but emphasise the necessity of including individual level data when assessing exposure to MCA. A clear relationship between exposure to MCA and risk of hypospadias could not be obtained in this study.

The biological mechanisms that might be responsible for an association between exposure to DBPs and hypospadias are not well understood.