In April 2014, the Department of Homeland Security (DHS) canceled its procurement of the new “BioWatch Gen-3” (next generation) biosurveillance program originally approved for development by the DHS in 2009. Following the cancelation, the U.S. House of Representatives Subcommittee on Emergency Preparedness, Response, and Communications held a hearing on 10 June 2014 to identify “lessons learned” from the failed procurement. Subcommittee Chairman Susan Brooks (R-IN) stated that it was “not the first failed acquisition in the Department’s history.” A June 2014 study by the U.S. Government Accountability Office (GAO), which testified at the hearing, found significant problems with the procurement process for the Gen-3 system and identified several challenges facing DHS as it works to maintain its current system while planning for future enhancements.
An Established Early Warning System
Charged with early detection of a biological attack in the United States, the BioWatch program has been operating using its current (Gen-2) technology platform for the past 10 years. Through the daily collection and analysis of air samples in a cooperative effort involving federal, state, and local agencies, BioWatch is designed to identify several biological pathogens – for example, anthrax – and serves as an early warning system to a potential biological attack in more than 30 cities, including Washington, D.C., and New York.
As a labor-intensive activity involving the manual collection of samples, the transportation of samples to nearby laboratories, and the requisite analysis itself, the current BioWatch system typically returns results within 12-36 hours. Due to the potential for rapidly spreading pathogens and subsequent person-to-person contact, DHS has been working to identify and implement new solutions that can significantly reduce the detection time and thereby mitigate large-scale exposure with various countermeasures, including emergency medical treatment, targeted vaccinations, and shelter-in-place orders.
During the past 10 years of operation, the current Gen-2 BioWatch solution has issued several “BioWatch Actionable Results” following the detection of a specific pathogen in a collected sample. For example, in October 2003, the tularemia organism (Francisella tularensis) was identified from a BioWatch detector in Houston, Texas. Although later determined to be naturally occurring, and not a biological attack (“false positive”), the finding nonetheless demonstrated the effectiveness of the BioWatch monitoring technology itself.
Reports on the Next Generation of Detection
In October 2009, DHS approved acquisition of a new BioWatch Gen-3 technology platform. The goal of Gen-3 was to automate the collection and analysis phase, thereby shortening the detection time to six hours and eliminating some manual labor costs currently incurred using Gen-2 technology. At that time, DHS’s Office of Health Affairs was overseeing BioWatch and suggested implementing new and automated “lab-in-a-box” solutions to achieve the performance objectives for Gen-3. However, concerns regarding an increasing estimated life-cycle budget for Gen-3, which increased from an initial DHS estimate of $2.1 billion in 2009 to $5.8 billion in 2011, coupled with questions regarding the acquisition process itself, led the House Homeland Security Committee to request that GAO conduct an assessment of the BioWatch Gen-3 acquisition.
In September 2012, the GAO issued its first report regarding Gen-3, finding that DHS had not fully followed its own acquisition process and, notably, that it had not performed an adequate analysis of alternatives (AOA) as required by the DHS Life-Cycle Framework for program development. In fact, at the completion of DHS’s “Phase 1” Gen-3 development effort, designed to assess the capability of market-available technology against its requirements for Gen-3, DHS conducted only limited field testing using a single vendor’s automated detection technology.
GAO also found that DHS’s Office of Health Affairs and its Science and Technology Directorate had each separately contracted with the Sandia National Laboratory to conduct performance assessments of new Gen-3 autonomous detection technology. However, each office requested that Sandia use different metrics to assess technical performance – Office of Health Affairs supporting a metric of the “fraction of the population covered,” while Science and Technology Directorate preferred “probability of detection.” Although these metrics may be complementary in terms of the overall analysis on the effectiveness of the new technology, GAO noted that this discrepancy could affect the ability of a chosen solution to meet the operational needs of the BioWatch program.
New Technologies – Cost vs. Effectiveness
Following the 2012 GAO report, DHS commissioned the Institute for Defense Analysis to perform a new AOA study that included a more “exhaustive” market survey of new technologies, and which provided DHS with alternative methodologies to consider when pursuing future BioWatch technology acquisitions. In January 2014, GAO reviewed the results of the AOA and found that the Institute for Defense Analysis had properly followed DHS guidance resulting in a “more robust exploration of alternatives.” Although not intended to identify a specific solution or technology for the Gen-3 acquisition, the AOA focused on providing information that would assist DHS with considering the trade-offs between cost and effectiveness when investing in new technology. After reviewing the findings of the AOA and GAO’s updated recommendations, DHS canceled the Gen-3 acquisition, noting in the June 2014 GAO report that “the AOA did not confirm an overwhelming benefit to justify the cost” of the new system.
Despite the failure of the Gen-3 acquisition, BioWatch officials are nonetheless excited about the future of the program, and they are looking forward to exploiting new acquisition tools and research to modernize the “proven” Gen-2 system. BioWatch Program Manager Michael Walker, PhD, stated in a phone interview on 2 July 2014 that its future success “will be measured by tapping into technologies that get us faster results that enable us to respond more efficiently when something is out of the ordinary.” Adding that their metric is “better, faster, cheaper,” Walker noted that there have been some misconceptions on the current (Gen-2) BioWatch solution, which some have incorrectly identified as “failures” that are, in fact, demonstrations of the effectiveness of the program, including the tularemia detection in Houston. He added that “timeliness and cost” are the main issues today, and that BioWatch remains a costly program to maintain.
Walker also discussed BioWatch’s support for “out-of-the-box” thinking for technical solutions. At the 10 June 2014 hearing, DHS testified that they were investigating the new use of federal “prize” authority for “engaging nontraditional” partners through a “biosurveillance grand challenge.” Walker added that they are “not ruling anything off the table,” including the use of data analytics to compliment BioWatch detection.
Big Data & Social Media Analytics
Gregory Koblentz, deputy director of the Biodefense Graduate Program at George Mason University, stated in a phone interview on 27 June 2014 that “big data” and social media analytics can support biosurveillance efforts by identifying “hot spots” of symptoms occurring – for example, in tight geographic locations or identifying “unusual illnesses that are happening off-season” – and then engaging the health community to investigate. He added that the best data must be collected using a “bottoms-up” (localized) approach that can identify a small “blip” that warrants attention. Koblentz noted that engaging new partners and analyzing data – for example, from a local CVS pharmacy when there is a localized increase in Pepto Bismol sales – could serve as another type of “trigger” for detection.
In addition, Koblentz expressed a belief that localized training could greatly improve the timeliness of detection once symptoms become apparent in the local population. He noted that most police officers and many doctors are not trained on what smallpox even looks like, and that training is relatively inexpensive to administer. Koblentz also stated that the convergence of public safety and public health data can be mutually beneficial to support biosurveillance efforts “by identifying what information each community has and how we can assist each other” during a potential biological event.
At BioWatch, Michael Walker remains focused on detection “before people get sick” in order to respond in a way that can prevent the spread of illness, not simply treat it once an outbreak is underway. However, Walker recognizes that BioWatch’s point system simply cannot sample everything and, “if a plume misses a detector, then we won’t see it.” He also recognizes additional opportunities to engage nontraditional partners, including the veterinary community. Walker suggested training veterinarians on what to look for in animals, as biological attacks will affect the local pet population.
On 16 June 2014, DHS Science and Technology Directorate’s Chemical and Biological Defense Division issued a broad agency announcement solicitation to improve its ability to “prevent, detect, respond to, and remediate from a chemical or biological incident, whether intentional or unintentional.” The $50 million, 3.5-year solicitation is open to “all responsible sources” including business and academia as well as federal laboratories and research centers. By casting a wider net and soliciting novel, incremental approaches on detection and response, DHS may be well on its way to improving BioWatch and implementing a “better, faster, cheaper” solution – though the latter goal may continue to be elusive.
Rodrigo (Roddy) Moscoso
Rodrigo (Roddy) Moscoso is the executive director of the Capital Wireless Information Net (CapWIN) Program at the University of Maryland, which provides software and mission-critical data access services to first responders in and across dozens of jurisdictions, disciplines, and levels of government. Formerly with IBM Business Consulting Services, he has more than 20 years of experience supporting large-scale implementation projects for information technology, and extensive experience in several related fields such as change management, business process reengineering, human resources, and communications.