policies related to the management of high-containment laboratories and report incidents to senior officials.10 Agencies have made progress in implementing many of our recommendations but the United States is still without a national strategy and does not have a single entity charged with overseeing the implementation of such a strategy to identify the aggregate risks associated with the expansion of high-containment laboratories and the type of oversight needed. In addition, the National Academy of Sciences, the White House, and federal committees and task forces have raised concerns about biological safety and security, including the management and extent of independent oversight over highcontainment laboratories. In this context, you asked us to evaluate issues related to the inactivation of pathogens in high-containment laboratories. In this report, we evaluated (1) the extent to which incidents involving incomplete inactivation occurred from 2003 through 2015; (2) any challenges that may affect the implementation of inactivation in high-containment laboratories; and (3) the extent to which the Select Agent Program referred violations and enforced regulations related to incidents involving incomplete inactivation. To conduct this work, we reviewed relevant laws, regulations, and guidance, including guidance issued by the Select Agent Program. We also reviewed relevant documentation and interviewed officials from the federal departments that own and operate high-containment laboratories, as well as officials from some academic and private high-containment 9 GAO-09-574 and GAO-08-108T. 10GAO-16-305. Page 5 GAO-16-642 Inactivation of Pathogens laboratories.11 We convened, with the assistance of the National Academy of Sciences (NAS), a meeting with 19 experts to discuss issues related to the inactivation of pathogens in high-containment laboratories. These experts represented academia, the federal government, and industry, and had combined expertise in pathogen and toxin inactivation and control, biological safety, risk assessment, legal requirements, standards development, incident reporting, epidemiology, and statistics. To evaluate the extent to which incidents involving incomplete inactivation occurred, we analyzed documentation on incidents reported to the CDC, APHIS, and NIH since 2003—when the Select Agent Program began requiring reporting of the theft, loss, and release of select agents— through 2015—the most recent year for which data were available. We took several steps to determine the reliability of the agencies’ incident databases, including reviewing agency documents and interviewing agency officials. We determined that the Select Agent Program incident database did not capture some incidents involving incomplete inactivation and was therefore not reliable on its own for establishing the number of incidents, as further discussed in the report. We verified through interviews and documentation each incident identified in the Select Agent Program database as well as additional incidents that we identified. We conducted site visits for 7 of 10 high-containment laboratories and interviewed officials from 8 of the 10 high-containment laboratories that the Select Agent Program originally reported to us as having incidents involving incomplete inactivation.12 We also interviewed officials from a nongeneralizable sample of 19 high-containment laboratories selected to 11The federal departments and their component agencies were the Department of Homeland Security; DOD and its departments of the Army, Navy, and Air Force; Department of Energy and its National Nuclear Security Administration and Office of Science; Department of the Interior and its Fish and Wildlife Service and U.S. Geological Survey; Department of Veterans Affairs and its Veterans Health Administration; HHS and its components of CDC, Food and Drug Administration, and NIH; USDA and its APHIS, Agricultural Research Service, and Food Safety and Inspection Service; and the Environmental Protection Agency (EPA). According to officials from the Department of Veterans Affairs, Interior’s Fish and Wildlife Service, and EPA, inactivation is not conducted in any of their high-containment laboratories so we excluded them from our additional work. In addition, the Department of Energy’s Office of Science has not operated a high-containment laboratory since 2006 so we also excluded it from our work. 12We contacted officials from all 10 high-containment laboratories at which incidents involving incomplete inactivation were originally reported to us by the Select Agent Program to arrange interviews; however, officials from one university and one private laboratory declined to be interviewed. Page 6 GAO-16-642 Inactivation of Pathogens represent a range of laboratories that work with human and animal pathogens and biological safety levels that had not reported incidents.13 We compared information learned from interviews with laboratory and agency officials and from federal documents about the definition of inactivation and incidents involving incomplete inactivation with comments from our expert meeting, guidance, and our past work. To identify challenges that potentially affected the implementation of inactivation in high-containment laboratories, we reviewed relevant