Already late for a vote, Rep. Bart Stupak (D-Mich.), chairman of the House of Representatives Energy and Commerce Subcommittee on Oversight and Investigations, offered a quick bit of advice to a panel of public health officials before slipping out of the hearing room: "Less construction, more research."
It was a message delivered many times by Stupak during a hearing on the "Silent Proliferation of Bio-Laboratories in the United States" that underscored a growing concern about biodefense research . The number of labs handling some of the most potent biological agents is increasing, accident reports and previously unreported cases are surfacing and federal requirements for reporting them remain ambiguous, the chairman and others contend.
"Shouldn't Congress want fewer people in fewer labs dealing with these agents?" Stupak asked during the hearing.
"What has changed that mandates the proliferation of labs over the last five years -- other than the money available" from the National Institutes of Health?" he continued.
Stupak wasn't the only one raising questions. Citing the preliminary findings  of a U.S. Government Accountability Office report due out in its entirety this spring, Keith Rhodes, chief technologist in the agency's Center for Technology and Engineering, said he's concerned that so much remains unknown about the number of high security labs and what goes on inside them.
Revelations this spring that Texas A&M University failed to report to the Centers for Disease Control and Prevention incidents of human exposure to dangerous biological agents until a year after they occurred have raised public consciousness of the issue of lab safety. Some have criticized the CDC for failing to uncover the problems at Texas A&M during a visit last year.
"Oversight is fragmented and for the most part relies on self-reporting," Rhodes said. "This incident raises serious concerns about oversight and whether the safety of the public is compromised."
The GAO report found that the number of Level IV labs, which require the highest-level security, has increased from 5 before the 2001 terrorist attacks to 15, including at least one in the planning stage. The number of Level III labs, a category to which Texas A&M's belong, is much higher -- with estimates ranging from a few hundred to well over 1,000. No single agency defines as its mission tracking the overall number of labs and where they are located, according to the GAO's survey of federal agencies.
More and more labs are being located at research institutions, prompting Rep. Joe Barton (R-Tex.), the ranking Republican on the Energy and Commerce Committee, to ask Rhodes: "Should labs be at universities?"
"I don't see why not," he responded. "It's absolutely not an issue of where the lab is located."
Many agree that since the disclosure of the problems at Texas A&M, reports of accidents have increased, in part spurred by the work of the Sunshine Project , an organization that tracks potential dangerous research and is dedicated to bioweapons control. At the committee's request, the CDC turned over information about 100 incidents that resulted in possible injury over the past four years at the nation's biolabs.
Richard Besser, director of a terrorism preparedness and emergency response office at the CDC, said accidents remain rare. "I think our labs are very safe," he said, adding that the recent cases have prompted the agency to change some practices during site visits, including looking at more documents and expanding the scope of interviews to include more lab workers during inspections.
During the hearing, Stupak pressed the health officials to explain the need for the increase in labs.
Hugh Auchincloss, deputy director of the National Institute of Allergy and Infectious Disease, which is within the NIH, defended the expansion of biodefense research, saying that a panel of scientists in academe, government and the private sector has supported the growth. When a highest-level security lab opens, he said, the public is made fully aware.
Both Besser and Auchincloss pointed to existing federal guidelines on handling so-called "select agents" when asked by Stupak how the agencies were promoting good lab practice. But neither gave the chairman direct answers when he asked what is an appropriate number of biodefense labs and which agency should take the lead in oversight. The officials said an inter-agency panel would be best suited to take up the latter issue.
"It seems like every agency is making its own assessment and doing its own thing," Stupak said.
Rep. Michael C. Burgess (R-Texas) said the committee should avoid recommending a decrease in research on potentially dangerous biological weapons, because it would discourage work that benefits national security. Rhodes and Stupak said the number of labs shouldn't increase until federal agencies conduct an overall assessment of the research needs and potential dangers.
Edward Hammond, director of the Sunshine Project, called for a reduction in the number of current labs and suspension of some biodefense research until the assessment takes place.
Eddie J. Davis, interim president of Texas A&M, received a friendlier line of questioning from the congressional panel. He apologized for the incidents, which prompted the CDC to shut down its biodefense research, and said the university is taking steps to get the program back.
"We have made a mistake in failing to report this in a timely manner," Davis said. "We're satisfied with an internal investigation showing it was human error. We're committed to getting it right."
Davis said he'd like to see clarifications to 2005 federal regulations that cover the handling of select agents and toxins. It remains unclear to many, he said, what defines an occupational exposure, for instance.
Hammond and others have criticized federal agencies for creating an environment that doesn't encourage reporting. Several panelists said -- and the GAO preliminary report also notes -- that coming up with a non-punitive solution is optimal.
Added Besser: "We don't want to have a system in place that leads to less transparency in reporting."