DC Bureau - Investigative Journalism, Environmental News, and National Security Stories That Matter

Last December the Defense Health Board, a federal advisory committee for the Pentagon, reviewed the medical assessment performed by the Army’s Center for Health Promotion and Preventive Medicine (CHPPM) at Qarmat Ali and concluded that it was “timely, comprehensive and appropriate.” But according to medical experts on sodium dichromate, CHPPM used inadequate testing methods and reached tenuous conclusions about the level of chemical exposure at the water treatment facility.

Too Little, Too Late

In its report, CHPPM states that if significant exposure of sodium dichromate occurred at Qarmat Ali, it would discover elevated levels of chromium in blood samples taken from soldiers who worked at the facility. In its analysis of these samples, CHPPM found that they were “lower than published results for groups with known occupational exposure.”

But according to Dr. Herman Gibb, a leading authority on chromium health effects, by the time CHPPM conducted its sampling the results would not have adequately gauged initial exposure levels. In its report, the Army contends that testing occurred 30 days after exposure ended. But a review of the official timeline shows that there was a 72-day gap between August 12, when KBR reportedly started containment and remediation efforts at Qarmat Ali, and October 24, when CHPPM completed its site testing—meaning that blood samples were likely not taken until 40 to 60 days after exposure.

Even after 30 days, due to the half-life of chromium in whole blood, the amount of the chemical left in soldiers’ bodies would have been significantly reduced, Gibb explained in an interview. “The blood tests were probably a waste of time.” Aaron Barchowsky, the University of Pittsburgh toxicologist, arrived at a similar conclusion, writing in a letter to Sen. Dorgan that “Measuring an individual’s blood for [chromium] several weeks or months after exposure would provide meaningless information unless there is a severe tissue injury.”

Dr. Coleen Baird, CHPPM’s medical advisor, disagrees. “If you had substantial exposure you should be able to detect more [chromium] than you would expect,” she said. In a follow-up e-mail, Dr. Baird wrote that “there is no medically recognized more appropriate test for this kind of exposure.”

But government agencies like the National Institute for Occupational Health and Safety (NIOSH) and the Agency for Toxic Substances and Diseases Registry (ATSDR) insist that there is no definitive biomarker for detecting sodium dichromate exposure. While increases in blood chromium levels can serve as an indication of exposure, they “cannot be used to predict the kind of health effects that might develop,” reads the ATSDR website. Even OSHA, the Occupational Safety and Health Administration, maintains that “certain biological tests, including blood tests…are of an uncertain value.”

As evidence, in one recent study of shipyard welders in Korea, eleven workers developed “chrome holes” burned through their noses, a common symptom of significant sodium dichromate poisoning. But blood and urine samples from the workers found no traces of chromium.

Sample Size

CHPPM only examined 137 of the 600 potentially exposed soldiers at Qarmat Ali—the vast majority of them from the Indiana National Guard. The medical team determined that since the Indiana troops had spent the longest amount of time at the treatment plant, if their tests did not indicate chromium exposure, there was “limited benefit” to evaluating the West Virginia and Oregon units.

But according to ATSDR, the development of health effects from chemical exposure “depends on the dose, the duration, how you are exposed, personal traits and habits, and whether other chemicals are present.” Dr. Gibb agreed, telling DCBureau that “the presence of symptoms is not so much a function of time spent at the site as it is a function of the intensity of exposure.”

Therefore, even though Oregon Staff Sgt. Larry Roberta spent less time at Qarmat Ali than many of his Indiana colleagues, he likely suffered from a more intense exposure when a violent windstorm blew sodium dichromate onto the chicken sandwich he was eating during one visit to the treatment plant in May 2003. Roberta suffered from such severe respiratory pains afterwards that he would eventually need surgery to repair his esophagus.

“It would be difficult and inappropriate to generalize the risks [of exposure] given the seasonal differences in the environmental conditions such as temperatures that would affect chemical uptake or differences in the dust storms,” wrote Dr. Barchowsky.

Though the Guardsmen spent a relatively brief period at Qarmat Ali—an average of 20 days—half of the production workers exposed to sodium dichromate in Dr. Gibb’s study started getting sick in less than three weeks. “They [the soldiers] could have developed symptoms within one day,” he said.

Medical Misinformation

Several months after CHPPM completed its blood and site testing, the National Guard distributed fact sheets to soldiers that were filled with misleading and inaccurate information about the Army’s findings.

One fact sheet claimed that only people exposed to sodium dichromate for “over two years” are at risk of getting lung cancer, a “fact” that the Army, the VA, and KBR have repeated in numerous reports, assessments, and statements regarding Qarmat Ali. But in Gibb’s study, half of the production workers that developed lung cancer had only been exposed to sodium dichromate for less than ten months. A quarter were exposed for less than two.  

Moreover, the Army references the readings from CHPPM’s air monitoring test as evidence that sodium dichromate levels were “well below relevant government standards.” It fails to divulge, however, that the testing being cited was conducted over one month after the area around Qarmat Ali had been paved over and remediated.

The fact sheet also states that “almost all” blood tests taken from U.S. soldiers were below the limit of chromium detection. But a complete, declassified version of the CHPPM report, obtained by DCBureau, shows that the real figure was only 68 out of 137—or just over half. And despite cataloguing over half of the results as being below the detection limit, the fact sheet says that 98 percent of the results showed detected chromium levels within a range of four to five micrograms per liter, a contradiction further muddled by the lack of a clear reference value used by CHPPM to compare its findings with “normal” blood chromium levels.

“If epidemiologists can’t understand this,” asked Gibb, “how are soldiers expected to?”