CHARLESTON, W.Va. -- A week ago, state and federal public health officials in West Virginia faced an almost impossible task. Thousands of gallons of a coal-cleaning chemical had contaminated the Elk River, the water supply for 300,000 people in a nine-county region. Little information was available about the potential dangers of the material. There was little time to act, and little information to help make smart decisions. Given the difficult hand they were dealt, national experts say that officials from federal agencies and the state of West Virginia probably did about the best they could responding to the leak of the chemical Crude MCHM from the Freedom Industries' tank farm just north of Charleston. Still, public health and toxic chemical experts also say officials here did a terrible job of communicating the risks -- and the important uncertainties -- to the public, and could have done more to ensure their estimate for a safe level of MCHM in water was sufficiently protective. "They did the best they could with very limited information," said Glenys Webster, an epidemiologist and postdoctoral fellow at Simon Fraser University. But, Webster said, the entire system the U.S. uses to study and regulate toxic chemicals left decision-makers in West Virginia with a huge vacuum of data and analysis they really needed. 'It's impossible to know with no information what level is safe," Webster said. "I don't think we have the information to make a decision that will satisfy the public." Dr. Lynn Goldman, dean of the School of Public Health and Health Services at George Washington University, agreed, recalling her days dealing with major chemical accidents at the California Department of Public Health and regulating toxic chemicals in the Clinton administration's Environmental Protection Agency. "I think it's putting too much on people," Goldman said. "What they have is not enough data to establish a drinking water standard. I can appreciate why they went down that road, but you can have a database that is too fragile for doing that." Throughout the week, state officials and West Virginia American Water have been giving residents the go-ahead to flush out their home plumbing systems and resume using tap water. They've based those decisions on water samples that show contaminant levels dropping below an emergency screening level set by scientists from the U.S. Centers for Disease Control. West Virginia officials announced the figure -- 1 part per million -- but did little to explain it publicly, and CDC officials spent most of last week dodging media questions about how they calculated the number. The CDC's number was based largely on an April 1990 Eastman study in which rats were exposed to varying levels of 4-methylcyclohexanemethanol. The study concluded that a concentration of 100 milligrams per kilogram of the chemical was the "No Observable Adverse Effects Level, or NOAEL," for the material. CDC officials divided that 100-milligram-per-kilogram level by 10 three times, once each to account for differences between rats and humans, differences between different humans and the lack of data on the chemical. Then, they set the final screening level for water based on a standard estimate for a 1-year-old child of about 22 pounds of body weight and consumption of a little more than a quart of water per day. In an interview Friday, Webster explained that what the CDC's exercise actually does is set an estimated safe dose for daily exposure to the chemical of 0.1 milligrams per kilogram of body weight. An average adult weighing about 150 pounds and drinking a little more than two quarts of water a day would get an exposure dose of about 0.03 milligrams per kilogram of body weight, Webster said. That amounts to three times lower than the level that the CDC considers as unacceptable under its screening level. But for a 1-year-old child weighing 22 pounds and drinking about a quart of water per day, she said, the daily dose would be 0.1 milligrams per kilogram of body weight. That's right at the level the CDC set with its screening level. "The people who are most vulnerable to any chemical exposure are the fetus, infants and young children," Webster said, and in this instance, children "will be closer to that maximum level considered safe by the CDC." Webster said that it's also important to consider the levels of Crude MCHM in the water, something that Tomblin administration officials have emphasized are dropping to well below the 1-part-per-million number. In a statement issued late Friday afternoon, Gov. Earl Ray Tomblin said, "My team's number one priority is the health and safety of those impacted by this crisis. My family is in the affected area. Many of the team's families and neighbors are in the affected areas. "This crisis hits home," the governor said. "The efforts we are taking are for all of our families." Still, while praising the state and CDC for issuing the bottled water advisory for pregnant women, outside experts say the advisory came too late and should have also included young children. The experts noted that the CDC calculation contained a wide variety of uncertainties that could mean the actual "safe" level is far different from what the agency estimated. "I think the CDC tripped up when they delayed before issuing the pregnant woman warning," said Jennifer Sass, senior scientist with the Natural Resources Defense Council. "I would recommend that not only pregnant women, but also infants and children avoid the water as long as the chemical contaminants are detected. And, the public needs to know what the detection limit is, so an informed decision can be made about how long to wait after the chemicals are no longer detected." Experts pointed to a variety of weaknesses in the CDC's approach. Most importantly, they said, it was based on little data. Also, rat studies based on high doses of chemicals can often mask adverse health effects that occur at lower levels. "It's a very rough calculation," Webster said. "And each step has uncertainty associated with it. It is not an exact science."Reach Ken Ward Jr. at email@example.com or 304-348-1702.