Radium has been found to be elevated in the shallow Kirkwood-Cohansey aquifer, a major drinking water supply source in parts of southern New Jersey, and the Englishtown, Old Bridge and Farrington aquifers in central New Jersey. Recent groundwater and drinking water surveys by the New Jersey Department of Environmental Protection and the U.S. Geological Survey provide a systematic basis for examining the potential public health implications of exposure to this naturally-occurring radium in water. High levels of exposure to radium isotopes have caused the bone cancer, osteosarcoma, among radium dial painters and patients receiving oral radium treatment for ankylosing spondylitis. Previous population-based studies of ingestion of the relatively lower levels of radium in drinking water have also shown associations with osteosarcoma mortality or incidence.
The New Jersey Department of Health and Senior Services conducted an exploratory study comparing rates of osteosarcoma in areas of southern and central New Jersey with varying amounts of radium in community drinking water supplies. Community water supplies were mapped using a geographic information system (GIS). Two measures of exposure were constructed. One measure classified water systems into those with radioactivity levels above either the U.S. Environmental Protection Agency (USEPA) maximum contaminant level (MCL) for combined radium-226 and radium-228 (5 picoCuries per liter (pCi/L)) or the gross alpha radioactivity MCL of 15 pCi/L. The second exposure measure was based on predicted cancer potency of the radium isotopes, and was expressed as equivalents of radium-228.
An association of osteosarcoma with radium exposure was observed among males, but not among females. For the sexes combined, the rate of osteosarcoma incidence in areas receiving drinking water above either the combined radium or gross alpha MCLs was 100% higher than in areas below either MCL. Among males, the osteosarcoma rate was more than three times higher (rate ratio = 3.3, 95 percent confidence interval 1.6, 6.0). When exposure estimates were categorized in terms of total radium cancer potency, expressed as equivalents of radium-228, the resulting incidence rate among those exposed at > 4 pCi/L was 90% higher than those whose tap water had less than 0.5 pCi/L. The elevated incidence of osteosarcoma was entirely associated with exposed males. For males in areas receiving water with > 4 pCi/L and 2.0-3.9 pCi/L, compared to those receiving <0.5 pCi/L, the rate ratios were 3.4 (95% CI 1.5, 6.7) and 3.1 (95% CI 1.3, 6.0), respectively. For males 25 and over, rate ratios were 6.2 (95% CI 2.0, 14) and 5.5 (95% CI 1.8, 13), respectively.
The results of this study, viewed in the context of previous studies of radium in drinking water, and biological knowledge of radiation effects, provide further evidence that exposure to radium at levels found in drinking water pose a measurable risk of this bone cancer. The findings are quantitatively consistent with risk estimates based on extrapolation from studies of higher level occupational and medical exposures, which have provided the basis for regulatory actions.