The US Government Accountability Office has made public its report criticizing the study methods used by the US Department of Homeland Security in concluding that a site in Kansas was the best place to build a National Bio- and Agro-Defense Facility.
Six sites were studied as potential locations to locate the $523 million facility, which would replace the Plum Island (NY) Animal Disease Center.
In its report, the GAO faulted the homeland security agency for what it termed a "quick and limited" economic study that was "unrelated" to the accidental release scenarios reviewed.
As a result of what it identified as flaws in DHS's methodology, "The conclusion that [foot-and-mouth disease] work can be done as safely on the mainland as on Plum Island is not supported," GAO concluded in its report.
The GAO arrived at its conclusion after questioning the models DHS used to gauge the economic effects of an outbreak of foot-and-mouth disease on the six sites; assessing the path of the plume from an accidental release of FMD or other pathogens; and evaluating the size and characteristics of the populations that would be exposed to that release.
DHS answered back in a two-page rebuttal letter printed as an appendix to GAO's 64-page report, Biological Research: Observations on DHS's Analyses Concerning Whether FMD Research Can Be Done as Safely on the Mainland as on Plum Island. The homeland security agency defended its methodologies and conclusions, and said it "will conduct a site-specific biosecurity risk mitigation assessment for the Manhattan, Kan., site to determine the required facility design and engineering controls needed to adequately protect NBAF during operations."
DHS also cited the operation of animal disease laboratories on mainland sites worldwide — including the US, where five labs carry the highest biosafety laboratory designation of BSL-4:
• Two Atlanta labs, each run by the US Centers for Disease Control and Prevention and Georgia State University.
• The US Army Medical Research Institute of Infectious Diseases at Fort Detrick, Md.
• The University of Texas Medical Branch in Galveston and the Southwest Foundation for Biomedical Research in San Antonio.
"The fact that live FMD virus research is already being performed on the mainland in other countries should not be dismissed by the GAO, as it clearly demonstrates that such work can be conducted safely on the mainland (with appropriate biosafety and biosecurity protocols in place to minimize the risk of release)," Bradley Buswell, DHS' acting undersecretary for science and technology, wrote in the agency's rebuttal, which was addressed to Nancy Kingsbury, the GAO's managing director of applied research and methods.
"While the study of contagious diseases anywhere is not without risk, modern biocontainment technology has made the likelihood of an accidental release of a pathogen extremely low," Buswell added. "Modern biocontainment technology has eliminated the need for locating animal-disease research on an island as was done decades ago."
The GAO quietly released its report on July 30, the day Buswell and Kingsbury were to have testified about it at a hearing of the oversight and investigations subcommittee of the US House of Representatives' Energy and Commerce Committee. That hearing was canceled after the subcommittee chairman, Rep. Bart Stupak (D-Menominee, Mich.), joined members of the full committee in marking up a bill among several being considered for overhauling the nation's healthcare system.
Three days earlier, the Washington Post summarized and quoted from a draft of the report that had been leaked to the newspaper.
[ pagebreak ]
Plume, Economics at Issue
At issue in the GAO report was the methodology behind DHS' conclusion that its NBAF biolab could carry out research into FMD as safely on the US mainland as at Plum Island.
In January, DHS chose to build NBAF in a 30-acre site on the Manhattan campus of Kansas State University, immediately adjacent to the Biosecurity Research Institute. The $54 million BRI has a BSL-3 ag-research lab and a BSL-3 Enhanced lab, mirroring the kind of labs NBAF plans to operate.
DHS chose Kansas over competing proposals from Georgia, Mississippi, North Carolina, and Texas, after concluding that the Sunflower State best met all four of the agency's criteria for its NBAF decision: Proximity to research capabilities and workforce; acquisition/construction/operations, or ACO, requirements, including costs; and community acceptance.
Of the rejected states, Texas leaders have objected most forcefully to DHS' decision. A consortium of leaders from the state government, the life-sci industry, and academia earlier this year sued DHS seeking to nullify the Kansas approval, only to have their lawsuit thrown out by a judge who found the project had not progressed sufficiently to be "ripe" for judicial review.
However, the judge allowed the Texas Bio and Agro-Defense Consortium to re-file its suit and reargue the case once "ripeness" has been established.
GAO took issue with DHS' use of the straight-line Gaussian plume model in evaluating the effects of an accidental release of pathogens from the NBAF. The model incorporated a year's worth of hourly averaged meteorological data to determine the probability of areas downwind from the release site being affected by the plume. Data from Kansas, as with Georgia, North Carolina, and Texas, came from 1991; the Plum Island data came from 1990; and Mississippi from 1992.
DHS measured the downwind dispersion of FMD virus from a hypothetical release up to the limit of the model, 10 km from the point of release for each of the accident scenarios in the six sites studied. According to the study, the average estimated air concentration of virus particles, or virions, in a cubic meter of air at five meters from the site of a spill would be the same 161,000 virions in Kansas that was recorded in Texas, Mississippi, and Plum Island.
The three locations shared identical readings at 200 meters (15,700 virions), 600 meters (29,100 virions), and 1,000 meters (13,590 virions). At 6,000 meters from the spill site, the concentration recorded for Kansas of 25 virions is less than the 40 measured for Texas, or 91 measured for both Mississippi and Plum Island. At 10,000 meters, 12 virions was recorded for Kansas, versus 14 for Texas, 16 for Mississippi, and 30 for Plum Island.
For FMD, infection is considered to result from exposure to as little as 10 virions.
The study also showed North Carolina recording the lowest concentrations of virus from an accidental spill, and Georgia the second lowest.
While straight-line Gaussian is used to evaluate the path of accidental releases of radiation releases from nuclear plants, GAO noted that the straight-line model and other Gaussian models have rarely been used to study the dispersion of biological particles because they lack a means for inputting biological decay rates. Also, according to GAO, the Gaussian models do not model dispersion less than 100 meters from a source, and lose their effective ability to predict results when the strength of the source outbreak, and weather, change over a long period of time.
The Gaussian models, GAO concluded, were better suited for studying continuous releases of a constant source strength and uniform wind field.
[ pagebreak ]
"Much better, validated, models are available and should have been used. We believe that if DHS is going to analyze something as important as the downwind dispersion of FMD virus after a release, it should use the best science and validated models available," the GAO concluded.
GAO said it based its findings on reviews by the Johns Hopkins University Applied Physics Laboratory, the Massachusetts Institute of Technology Lincoln Laboratory, and Gryphon Scientific.
GAO suggested that DHS instead use "other available models," but cited only one specific example — the RIMPUFF, a local-scale puff diffusion model developed by Risø National Laboratory for Sustainable Energy in Denmark, and used, according to the report, in several European national emergency centers. RIMPUFF is designed to help public safety agencies deal with chemical, nuclear, biological, and radiological releases to the atmosphere.
"Models like RIMPUFF are superior to Gaussian models, because they apply local wind, precipitation, and turbulence data and sophisticated scaling theory, and because puff diffusion can be calculated on many time scales. RIMPUFF also applies biological decay rates for FMD," the GAO report concluded.
In its rebuttal letter, DHS noted that the GAO "provides no analysis that would indicate that a different methodology would yield different results."
DHS' economic study, carried out by the agency's Biodefense Knowledge Center at Lawrence Livermore National Laboratory, concluded that the primary economic effect of an FMD virus release would be an export ban on US livestock products if FMD or another pathogen escaped from the biolab, regardless of where it was located.
BKC quantified the economic loss from an FMD virus release as highest in Kansas at $4.2 billion, and lowest at the Plum Island site, at $2.8 billion. BKC based its projection on estimates of foreign trade lost because of the duration of an export ban; of "direct" costs to government, and of "indirect" costs such as the disruption to industry. The indirect costs, namely industry disruption, ranged from just over $1 billion for the Kansas site to just over $31 million for Plum Island.
"We found several weaknesses in the economic analyses. For example, they (1) did not incorporate market response to an FMD outbreak or consider the effect of establishing a containment zone to moderate the costs of the export ban, and (2) were constrained by the limited outbreak scenarios used and the lack of detail," GAO contended in its report.
One example of lack of detail cited by GAO: DHS carried out a review of scientific literature that included a 2007 study of an FMD outbreak in southwest Kansas. But the homeland security agency told GAO the purpose of its literature search was to identify upper and lower bounds of potential economic losses, not to develop detailed estimates for the Kansas site or other specific sites under review.
As for market response, GAO said both supply and demand for livestock products would be likely to change once FMD was detected, which would affect the estimate of the overall costs of an outbreak.
"Since losses from export sales would be offset by domestic purchases (at lower prices) and by consumers' substituting unaffected animal products (say, chicken for pork), prices and revenues to producers of the substitutes could rise," the report stated.
And while losses may be minimized by creating containment zones limiting the animals that are subject to a ban, DHS should consider differences across the various sites being considered for the NBAF biolab, GAO added.
DHS argued differently in a letter to GAO, citing a November 2008 letter from Bernard Vallat, director general of the World Organisation for Animal Health, that differences in the national impact of an outbreak relate less to where outbreaks occur than to how a country's authorities respond.
"While we agree that the effectiveness of a country's response is paramount, we
believe that where an outbreak occurs is also significant. Building FMD scenarios that take into account geographic and animal demographic factors could reveal whether there is an advantage to sites where developing a containment zone may be facilitated by unique characteristics, such as its being an island," GAO countered.
[ pagebreak ]
GAO faulted BKC for carrying out an initial one-day study that did not study the effects of an aerosol release of pathogens, though such a release and its effect on the six locations under review were examined in a follow-up report that took one week to prepare.
The follow-up study based its findings on the Multiscale Epidemiological/Economic Simulation and Analysis Decision Support system, which measured the spread of FMD via direct contact animal movement, high-risk and low-risk indirect contact, and interstate transportation of live animals — but not by inter-herd aerosol transmission.
Despite the facility's lower costs associated with disruptions, Plum Island has long been ruled out by DHS as an option for the NBAF. The homeland security agency has contended that the facility, built in the 1950s, is too small and increasingly too outdated to carry out the volume and quantity of testing needed to protect the nation's $1 trillion agricultural industry from the potentially catastrophic results of a bioterror attack on livestock.
DHS' environmental report concluded renovating Plum Island would be the costliest option for developing NBAF, at a projected $752.5 million. However, a team of outside consultants hired by DHS last year offered an even higher estimate for Plum Island and all other NBAF options, with Plum Island projected to cost $939.3 million.
According to a July 25, 2008, site cost analysis by the NBAF Design Partnership — which included Perkins+Will, Flad & Associates, Merrick, Affiliated Engineers, and ccrd partners — costs for the biolab ranged from $648.2 million for the Mississippi site, and $652.4 million for the Texas site, to $677.3 million for North Carolina, just under $680 million for Georgia, and $724.55 million for Kansas.
Those costs don't include state incentives. Texas has said the true cost of developing NBAF in Texas would be $552 million, compared with $619 million for Kansas. But when the $210 million in Kansas incentives are included, that state again finishes with the lowest cost of the states seeking NBAF.
Texas initially offered $44 million in incentives, then sought to raise its package to $100 million after DHS' March 31, 2008, deadline for submissions by competing states. Texas has argued the deadline should have been extended because its legislature was not in session during the period set by DHS. Texas said its revised incentive package would have lowered the cost of building NBAF below the $501.7 million projected by DHS' final EIS, to $401.7 million.
In its final environmental impact statement on the project, consultants hired by DHS said the Kansas site would be able to accommodate a biolab that could be operated safely, and at the lowest cost, which was projected at $563 million. The final EIS concluded that the true cost to DHS would be roughly $200 million lower in light of a $105 million bond the state promised to issue for infrastructure improvements, another $5 million subsidy from the city of Manhattan, and roughly $100 million in additional state money toward biological research at the BRI.
DHS lowered its projected cost for the facility to $523 million earlier this year when it issued a formal solicitation for construction bids. However, that figure is still 16 percent above the project's initial cost estimate of $450 million.