Abstract:

Although a considerable body of occupational and laboratory toxicology data have demonstrated the carcinogenic action of vinyl chloride monomer (VCM), there appears to be little epidemiological literature which examines significant exposure to children. A previous EPA-sponsored pilot study identified a cohort of 1,363 children who attended an elementary school adjacent to a VCM processing plant in Saugus, California. A health status questionnaire was developed and administered to 450 subjects and 270 spouses. Three unusual causes of death were identified. The percentage of pregnancies ending in miscarriage and the proportion of children reported to have major illness were significantly elevated in comparison to internal controls (spouses of exposed males) and the published literature. This current study identified a non-exposed control group (N = 979), set up a computer data base management system to facilitate subject tracing, performed an analysis of the mortality experience of the two groups up to 1980, and developed a protocol to validate pregnancy outcome data. Vital status was determined for 76.2 percent of the exposed group and 78.0 percent of the control group. Although the mortality rates up to this point are small, there is some evidence of an increased proportion of non-trauma deaths among the exposed group, and an increased rate among males. There is also evidence of a decreased rate of trauma deaths among the exposed cohort. Reproductive outcome responses were validated for 16 of the 246 married females (pregnant at least once) who had been interviewed. The validation protocol results suggest that the questionnaire is a good instrument capable of recording valid reproductive information.

1.1.1 The Problem:

The carcinogenic activity of vinyl chloride monomer (VCM) has been clearly established, and steps have been taken by both the California Air Resources board (ARB) and the U.S. Environmental Protection Agency (EPA) to limit occupational and general public exposure (ARB, 1977; USEPA, 1975a, 1976). Before the health hazard of VCM was recognized, however, no restrictions other than common zoning practices limited siting of plants producing or handling the monomer.

One such plant is the polyvinyl chloride (PVC) polymerication and phonograph record-producing facility operated by the Keysor-Century Corporation in Saugus, Calfornia. Since 1958, the plant has been operating at a nominal 24 hours per day, seven days per week, producing PVC from VCM and vinyl acetate. It was not until 1974 that the plant was directed to initiate significant action to control worker exposure; since then, they have satisfied federal Occupational Safety and Health Administration (OSHA) requirements for reporting worker health records, and have installed some process controls, including a fume incinerator.

The Saugus Elementary School, which was built in 1938, is located approximately 1,000 feet (300m) from the plant. The proximity of the plant and school is believed to be unique in the United States. The latency period for, and the effects in, children could differ from those resulting from adult exposure. It has been postulated (Hefner, 1975; Watanabe, 1976) that risk to young children may be greater because of the reduced activity of several enzyme systems which are thought to be important in the metabolic processes that facilitate body elimination of VCM. For these reasons, and because VCM concentrations above the current occupational standard of 1 ppm had been measured at the school site, the EPA sponsored a two-year pilot study (1979-1981) of the health status of the students who had attended the school during the first six years of plant operation. Science Applications Inc. (SAI) was the contractor, and the findings of the pilot study were submitted to the EPA (Ziskind et al., 1981).

1.1.2.1 Objectives and Methods:

The objectives of the EPA-sponsored study were to (1) evaluate the incidence of possible carcinogenic and non-carcinogenic health effects among a cohort of former students at the Saugus Elementary School; (2) establish tracing and registry mechanisms to maintain contact, update health status, and permit reexamination of the study cohort; and (3) estimate ambient VCM concentrations at the school site during the exposure period.

The maximum time between initial exposure and the pilot study was 21 years. Since the latency commonly associated with VCM exposure is 15 to 29 years (USEPA, 1975b), it was hoped that a significant portion of any resulting health impact would be observable in the oldest student population. The pilot study cohort was selected to be the 1,363 children who attended Saugus Elementary School for at least one year during the academic years between 1958/1959 and 1964/1965, inclusive. The minimum time between first exposure and the pilot study was therefore 16 years.

Initial information on the study cohort was obtained from pupil records provided by the Saugus Union School District (SUSD). Names, addresses, birthdates, periods of attendance, and other useful data were placed in computer files, which served as a registry for the study. Since only one measurement of ambient VCM levels at the school site had been made before introduction of emission controls at the plant, estimation of student exposure during the study period was problematical. Emission factors from the literature were combined with limited plant-specific information to estimate what uncontrolled VCM emissions would have been, and then dispersion modeling was used to estimate average exposures during school hours on those days when the school was in session. The uncertainty in these exposure estimates was so great, however, that it was decided to use months of attendance as a surrogate measure for exposure. Production levels at the plant are believed to have been fairly steady throughout the time of interest; over the long run, therefore, our assumption appears to be reasonable.

In the pilot study, 586 members (43 percent) of the exposed cohort were located. Since it was not known in advance what type, if any, of pathology may be induced in children under the circumstances of the study, it was decided to administer a comprehensive questionnaire covering a wide range of symptoms, reproductive outcomes and disease states. Data on smoking, drug use, and occupational history were also obtained. Specially trained interviewers administered the questionnaire to 450 subjects and 269 spouses over 10 months.

1.1.2.2 Pilot Study Findings:

Our pilot study findings were:

• Risks of fetal deaths, induced abortion and major illnesses in children of exposed women were higher than those corresponding to unexposed spouses of male members of the cohort; and

• Three deaths among the study cohort were from causes unusual for the age group: malignant lymphoma, malignant melanoma and erythema multiforme (etiology unknown).

It should be borne in mind that these findings are very tentative; both the small study population and the lack of controls prevent us from being more conclusive. In addition, confounding factors which may have influenced the results were not explored in depth. The findings were, however, sufficiently suggestive to warrant further investigation.

1.1.2 Objectives of the Present Study:

The objectives fo the ARB-sponsored study were to build upon previous research by:

(1) Defining, locating and setting up a data base for a suitable control group;

(2) Designing and implementing a computerized registry to facilitate tracing of subjects and controls during this study and in any future studies;

(3) Conducting a mortality study of the exposed cohort and the controls;

(4) Developing and testing a protocol for validating the reproductive outcome results of the pilot study.

[...]