ANOTHER USDA coveup
Researcher alleges CDC cover-upUSDA microbiologist was infected with E. coli O157:H7, and a report says she was responsibleBy John Dudley Miller
A former government researcher has charged the US Centers for Disease Control and Prevention (CDC) with falsely concluding that she was responsible for the oversight of a dangerous experiment that almost killed her. Ru-ching Hsia, a microbiologist who worked in a US Department of Agriculture (USDA) lab in Beltsville, Maryland, from June to December 2003, became sick five days after another researcher's technician conducted an experiment in her lab.
Hsia told The Scientist that a CDC report–which she said incorrectly concluded that she was supervising Santiago Rossi, the technician, during the experiment–would lead others to believe that she is a sloppy scientist and will ruin her reputation forever. "I feel that I have a moral responsibility to expose this cover-up," she wrote in an E-mail yesterday (June 14). "I believe in science and I could not respect myself if I allow this lie and deceit to continue when there is risk of more infections happening." The report, which has not been officially released, was obtained by Hsia through a Freedom of Information Act (FOIA) request.
The basic facts of the case are not in dispute. In an experiment he performed on December 5, 2003, to determine whether liquid sanitizers could eliminate virulent E. coli O157:H7 from produce, Rossi, working on a bench he shared with Hsia and not under a containment hood, as is required by CDC regulations, began by dipping apple slices in the bacteria. He then put the contaminated slices into three home salad spinners, spun out the excess liquid by hand, put them in the sanitizer liquid, spun them out again, and cultured the apples to see if any E. coli survived. Afterward, he sprayed alcohol on the wet spinners and rinsed them in the sink. Hsia later touched the sink faucet while washing her hands, which is when she believes she became infected.
Stanley Maloy, president-elect of the American Society for Microbiology, said that Rossi should have used a harmless mutant of O157, because as few as 10 cells of the bacterium–not millions, like most other bacteria require–can cause infection. Rutgers University professor Richard Ebright said last month that Rossi should have used O157 only in a final experiment to prove that real 0157 and the harmless mutant give the same results. Maloy and Ebright agreed that spraying alcohol on wet spinners is insufficient to decontaminate the equipment used.
Hsia developed hemolytic uremic syndrome, lapsed into a 30-day coma, was put on a ventilator, and nearly died. She said she now has short-term memory problems and recurrent severe headaches, and she can no longer conduct research with food-borne pathogens because she cannot risk a relapse from reexposure. She said yesterday that her "dream to be an independent researcher in microbiology has been nearly demolished." Hsia receives 70% of her former salary, tax-exempt, from workers' compensation through the Department of Labor, which is trying to find her a new job. The USDA dismissed her in January, saying that her illness made her ineligible for any laboratory position.
In January 2004, the CDC sent a team of five people to investigate the accident, including a representative of the Maryland Department of Health and Mental Hygiene, and an FBI agent to investigate the suggestion made by coworkers that another lab employee intentionally infected Hsia. The employee was exonerated.
The report of the CDC investigation–which Hsia and her husband, Patrik Bavoil, a professor of microbiology at the University of Maryland Dental School, have posted on their own Web site with a rebuttal–strongly implies Hsia was responsible for the mishap. "Researcher A is a 44-year-old microbiologist who was working with E. coli O157:H7 on December 4th and 5th, 2003," it reads. The report does not indicate that Hsia is Researcher A, but she was 44 at the time, and other descriptions in the report and on her lab's Web site leave little doubt that she is that person.
The report never mentions any other researcher involved in the experiment, only the technician, leaving the impression this was Hsia's research. Notes made by Leslie Edwards, the Maryland investigator, and made available after a FOIA request by Hsia and Bavoil show that when the CDC team interviewed Hsia after she came out of her coma, Hsia denied the experiment was hers, denied being Rossi's supervisor, and gave detailed explanations to prove the faulty experiment wasn't her research. None of her comments were included in the report, even though Robert Tauxe, chief of the CDC branch that investigated the accident, said the team had them when the report was written.
Hsia maintains the experiment belonged to another researcher, Yaguang Luo, who started a month-long vacation the day the technician conducted the experiment. In an interview earlier this week, Tauxe said that Hsia is right. Asked whether Luo designed the experiment and trained the technician to do the work, Tauxe told The Scientist, "that would be my understanding."
The CDC report charges that "[Hsia] was officially supervising [Rossi] while working on another experiment in the same laboratory." However, time sheets provided by the USDA in response to a FOIA request show that Luo was Rossi's official supervisor, signing almost all of his biweekly time sheets. While Luo was on vacation, Luo's and Hsia's boss, Kenneth Gross, signed them.
Hsia and Bavoil complained to the CDC that the report was seriously flawed. But a November 30, 2004 letter from Judith Aguilar, acting director of the division of bacterial mycotic diseases in CDC's National Center for Infectious Diseases, said "the report is an accurate summary of the information obtained and the observations" with the exception of a few small clarifications and corrections she offered.
The couple has considered filing a lawsuit against the USDA in the matter, but they said lawyers told them they had very little chance of winning. After she complained to the USDA that she was not Rossi's official supervisor, Edward Knipling, the administrator of USDA's Agricultural Research Service (ARS), wrote back April 11 acknowledging she was not. In Luo's absence, he wrote, Gross "is the official second-level supervisor of [Rossi]." Nevertheless, he wrote, both Rossi and Luo told the CDC that Hsia "had informal, temporary oversight of [Rossi]" in Luo's absence. "This is common practice in all of our laboratories," he wrote. Earlier this week, Tauxe agreed, saying that Hsia was Rossi's informal supervisor.
But three USDA veterans said the concept of informal supervision does not exist. One of them, Sandra Hays Miller, director of public relations at the ARS's Beltsville facility, told The Scientist last month that she had never heard of the phrase "and I've been with the agency for 17 years."
The CDC's Tauxe said Hsia became Rossi's informal supervisor automatically when Luo left because she was senior to Rossi and in the same room with him. Hsia, Tauxe said, "as a senior scientist in that room would have been the person to get advice and support from." Last month, Miller called the concept of automatic informal supervision "ridiculous."
Tauxe noted that Hsia had inoculated a culture of O157 days before the experiment, and that shortly after her 0157 infection, she came down with an infection from a strain of salmonella she used in her own research after spilling some of it. The report also says several coworkers complained about her lab techniques, especially the fact that she didn't always wear protective gloves.
Hsia acknowledged inoculating the culture and spilling a tiny amount of salmonella-infected liquid a few days before the experiment. She said she recorded both events in her lab notebook. She said the small amount of salmonella spilled would otherwise not have made her ill.
Hsia said she doesn't always wear gloves because she isn't always handling contaminated surfaces. Several former supervisors and colleagues, including her PhD advisor at Stanford, John Boothroyd, and David Ojcius, who worked with her at the Pasteur Institute and is now a professor at the University of California, Merced, told The Scientist she is one of the safest and most conscientious researchers they have ever known. James Kaper, who studies 0157 at the University of Maryland, noted that she has 20 years of microbiology experience, far more than her lab mates. However, Maloy said he requires gloves to be worn at all times in his lab, because surfaces researchers think are clean could be contaminated.
Contacted this week and asked who trained Rossi, Luo said, "it's a group," mentioning three other researchers in the lab. "Normally it's done by the microbiologists," she said. Hsia said that it is a group on paper, but in practice each scientist has his or her own individual projects. "She is a food technologist," Hsia said of Luo, "and she has all the sanitizer equipment and food processing in her lab."
Asked if she worked with 0157, Luo said, "I'm not a trained microbiologist," and said microbiologists in the lab worked with it. When pressed, she admitted doing "collaborative research" with O157, but said, "I didn't do anything myself, because I wasn't trained for it."
However, in April 2004, four months after Hsia left the lab, Luo herself became infected with the same strain of O157 that infected Hsia. She recovered quickly. In the past three years, Luo has published two journal papers about O157 research and has made several national conference presentations. Although those reports include collaborators, only one of her group is ever listed as a coauthor. A 2003 paper, for example, substituting carrot shreds for apple slices in the salad spinners, did not include a microbiologist author.
Of the CDC report's accuracy, Luo said: "I don't know why the CDC wrote the way they did."
Bavoil has tried to drum up support for his wife by publishing letters in the January ASM News and the April Applied Biosafety Journal. He and Hsia have also requested that the USDA's Office of the Inspector General investigate statements by laboratory personnel used in preparing the CDC report. The OIG informed Hsia on June 3 that it would reinvestigate, "since the matter involves a possible criminal offense" of misconduct.
Correction (posted June 17):
