Molybdenum in the diet: an estimate of average daily intake in the United States.

Previous studies have estimated the average intake of molybdenum (Mo) from the diet at approximately 300 to 400 micrograms/day. Foods collected in a grocery basket sampling program were analyzed for Mo content. The Mo concentration of these foods was combined with published United States Department of Agriculture estimates of food consumption to estimate the average daily dietary intake of Mo in the United States. This estimate is less than those previously reported and varies between 120 and 240 micrograms Mo/day, depending on age, sex, and income.

Inorganic arsenic: a need and an opportunity to improve risk assessment.

This paper presents views on the current status of (inorganic) arsenic risk assessment in the United States and recommends research needed to set standards for drinking water. The opinions are those of the Arsenic Task Force of the Society for Environmental Geochemistry and Health, which has met periodically since 1991 to study issues related to arsenic risk assessment and has held workshops and international conferences on arsenic. The topic of this paper is made timely by current scientific interest in exposure to and adverse health effects of arsenic in the United States and passage of the Safe Drinking Water Act Amendment of 1996, which has provisions for a research program on arsenic and a schedule mandating the EPA to revise the maximum contaminant level of arsenic in drinking water by the year 2001. Our central premise and recommendations are straightforward: the risk of adverse health effects associated with arsenic in drinking water is unknown for low arsenic concentrations found in the United States, such as at the current interim maximum contaminant level of 50 microg/l and below. Arsenic-related research should be directed at answering that question. New epidemiological studies are needed to provide data for reliable dose-response assessments of arsenic and for skin cancer, bladder cancer, or other endpoints to be used by the EPA for regulation. Further toxicological research, along with the observational data from epidemiology, is needed to determine if the dose-response relationship at low levels is more consistent with the current assumption of low-dose linearity or the existence of a practical threshold. Other recommendations include adding foodborne arsenic to the calculation of total arsenic intake, calculation of total arsenic intake, and encouraging cooperative research within the United States and between the United States and affected countries.

Biochemical abnormalities in workers exposed to molybdenum dust.

Exposure to molybdenum in dust was measured in a molybdenite roasting plant. This exposure was accompanied by large elevations of serum ceruloplasmin and smaller increases in mean serum uric acid levels in the workers. Absorption of molybdenum from the dust was demonstrated by increases in plasma and urinary molybdenum levels. It remains necessary to demonstrate whether such exposure results in long-term health effects.

Scaling toxicity data across species.

The response of various species to doses of chemicals can often give the impression that some (such as cattle in the case of molybdenum) are much more susceptible than others to these chemicals. These impressions usually rely on an underlying assumption that equivalent doses are based on mg of the chemical per kg body weight of the animal. That is, that doses scale as the first power of body weight. This assumption is more often wrong than right. When viewed in a more general way, where the scaling is proportional to a power of the body weight and the exponent determined empirically, it is often found that equivalent doses scale with an exponent in the range of 0.6 to 0.8. As a result, larger animals are indeed more susceptible to toxicity on a mg kg(-1) body weight basis, but this is not because of unique differences in the species, but only because of different body sizes. This method of scaling is called allometry or allometric scaling. An early version of this approach was based on body surface area where the exponent is 2/3. More recently, pharmacokinetics has revealed that the reason for the different response of larger animals is related to the slower metabolic and clearance rates for larger animals which give rise to larger biological half-lives for chemicals in the body and to higher tissue concentrations per given dose.

Does the animal-to-human uncertainty factor incorporate interspecies differences in surface area?

Risk assessment practices for noncarcinogens typically employ an uncertainty factor (UF) for animal-to-human extrapolation when defining acceptable levels for humans based on animal studies. EPA has interpreted the use of this factor as addressing interspecies differences due to dose normalization via surface area (exposure-dose relationships) and to innate differences in species susceptibility (dose-response relationships). Thus EPA has concluded that dose normalization via surface area is not necessary when using animal studies to define acceptable levels for noncarcinogens for humans. In this report we challenge this position on both theoretical and practical grounds. It is recommended that the UF for animal-to-human extrapolation for noncarcinogens in the risk assessment process and the technique for dose normalization be considered distinctly.

Assessment of blood lead levels in children living in a historic mining and smelting community.

Lead poisoning in childhood is an important public health problem, and thus, it is important to determine how children are exposed to lead. In 1987, the authors conducted an exposure assessment and blood lead screening for children aged 6-71 months living in Leadville, Colorado. High levels of lead had been found in the soil as a result of both past mining and smelting activities and natural mineralization. Blood was collected from each child for lead analysis, and behavioral characteristics were identified through an interview with a parent or guardian. Three sources of exposure to lead were associated with blood lead levels: lead in a core sample taken from the backyard of the family's home, lead brought home on the clothes of a miner, and lead from soldering in the home. Two pathways of exposure were associated with blood lead levels: the child swallowing things other than food, and taking food or a bottle outside to play. Multivariate regression using these variables found effect modification by age. For children aged 6-36 months, only sources of exposure were independent predictors of blood lead levels, while in children aged 37-71 months, a pathway of exposure in addition to sources of exposure independently predicted blood lead levels.

Is ingested inorganic arsenic a "threshold" carcinogen?

Ingested inorganic arsenic (As) is known to be a human carcinogen. An intriguing question is whether there is a threshold for the carcinogenic effects of As, i.e., is there a level below which it does not induce the development of cancer(s)? This Roundtable will discuss the United States Environmental Protection Agency's As risk assessment using the Taiwan data from different viewpoints. It will also consider the hypothesis that there is a threshold for As and data for or against this hypothesis. For example, some scientists believe that epidemiological data cannot answer this question, while others feel that different study designs and larger sampling will provide adequate data. Reasons for each position are given. This Roundtable discussion demonstrates the controversy surrounding the use of the Taiwan data for risk assessment.