A Survey of Attitudes and Actions on Dual Use Research in the Life Sciences: A Collaborative Effort of the National Research Council and the American Association for the Advancement of Science (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Summary Over the past 50 years, rapidly expanding knowledge in the biological sciences has brought great benefits to society. But the same technologies that fuel scientific advances also pose potential risks—that the knowledge, tools, and techniques gained through legitimate biotechnology research could be misused to create biological weapons or for bioterrorism. This is often called the dual use dilemma of the life sciences. The fear is that some research—dual use research of concern—could be used by those with malicious intent to do great harm. Yet even research with the greatest potential for misuse may offer significant benefits as well. Determining how to constrain the danger without harming essential scientific research is critical for national security as well as prosperity and well-being. Growing concerns about the potential risks of dual use research of concern have led to calls for raising awareness within the scientific com- munity about dual use issues. Several reports from the National Research Council, including Biotechnology Research in an Age of Terrorism (NRC 2004a, herein called the Fink report), Seeking Security: Pathogens, Open Access, and Genome Databases (NRC 2004b), and Globalization, Biosecurity, and the Future of the Life Sciences (NRC 2006a) share a common message: The scientific community should take preemptive steps to protect the integrity of science and to minimize the risk of misuse of dual use research of concern. These reports also contain recommendations for enhanced education and outreach programs to raise awareness of the potential unin- tended harm from dual use research. They recommended that scientific societies and professional associations undertake programs to educate  

 DUAL USE RESEARCH IN THE LIFE SCIENCES scientists about dual use issues and their responsibilities to help mitigate the risks of misuse. In addition to proposed efforts by professional and scientific societies, the National Science Advisory Board for Biosecurity (NSABB), created in 2004 in response to the Fink report (NRC 2004a), has an explicit man- date to “provide recommendations on the development of mandatory training programs for education and training in biosecurity issues for all scientists and laboratory workers at federally funded institutions.” A few universities, nongovernmental organizations, and professional societies have undertaken or are planning education efforts even before there is any government mandate to do so. For example, though certainly not exhaustive, in the United States the Federation of American Scientists, the Southeast Regional Center of Excellence for Emerging Infections and Biodefense, and the Center for Arms Control and Nonproliferation all have online materials or programs available. THE AAAS-NRC SURVEY PROJECT In September 2005, NRC and the American Association for the Advancement of Science (AAAS) Center for Science, Technology, and Security Policy (CSTSP) hosted a meeting, “Education and Raising Aware- ness: Challenges for Responsible Stewardship of Dual Use Research in the Life Sciences,” that brought together over 50 participants to share information and explore ways to engage and educate the research com- munity most effectively. The discussions underscored how little data exist about levels of awareness and attitudes about biosecurity issues in the life sciences community and highlighted the critical need to move beyond anecdotes to empirical evidence. Building on the results of their 2005 planning meeting, CSTSP and NRC developed a plan to survey a sample of AAAS members in the life sciences about their knowledge of dual use issues and attitudes about their responsibilities to help mitigate the risks of misuse of their research. In addition to providing essential baseline data, it was hoped that the results of the survey would generate more attention to the continuing challenges of dual use issues and foster additional debate among life sci- entists about their personal and professional responsibilities. The project used consultations with experts and practicing scientists as well as four focus groups in 2007 to design and refine a Web-based survey question- naire that could be e-mailed to AAAS members in the life sciences. The focus of the survey was on practicing scientists in the biological, health, and agricultural sciences working in the United States. AAAS is the largest general scientific society in the world and has more than 64,000 life scientists among its members. Since the membership is largely American 

SUMMARY  (about 84 percent) and primarily composed of scientists with advanced degrees (e.g., Ph.D.s or M.D.s), there was ample opportunity to attempt to survey the attitudes of American researchers across the full range of life science subfields. Because the survey was conducted via e-mail the study was restricted to 24,194 members who had validated e-mail addresses out of 64,787 life scientists who belong to AAAS. A random sample of 10,000 from these 24,194 AAAS members was selected to be contacted. The survey was fielded from early August to early October 2007 by the AAAS Office of Member Services, with several follow-up e-mails to encourage a higher response rate. Among those sent the survey, 2,713 individuals viewed the survey (i.e., clicked on the link to the question- naire provided in an e-mail); 1,954 individuals completed part of the survey; and 1,570 completed the entire survey. This leads to a response rate of about 16 percent for completed surveys and 20 percent including partial responses. Almost all of the respondents had conducted or managed life sci- ences research (and three-quarters of them are currently doing so), were employed, had a postgraduate degree, and were U.S. citizens. In addition, a substantial majority of the scientists were academics and most were mid-career. Given the low response rate, the lack of information by which the characteristics of the nonrespondents could be compared to those of the respondents, and the fact that the sampling frame included only those AAAS members whose e-mail addresses were known to AAAS, the sur- vey results should not be generalized to the general population of U.S. life scientists. The methodological difficulties encountered in this project with regard to obtaining a representative sample and a high enough response rate to make generalized conclusions provide valuable lessons for future surveys on this as well as other topics of interest to the scientific com- munity. Although it is necessary, because of these issues, to confine the report to the respondents and not to generalize beyond them, the com- mittee believes that the survey results (including respondents’ anecdotal comments) provide interesting indications of how the U.S. life sciences community may view dual use research that merit further investigation. SURVEY RESULTS The results of the survey provide some of the first empirical data about the perceptions of a sample of U.S. life scientists across a variety of disci- plines about the potential risks of misuse of legitimate scientific research for malicious purposes. The survey data provide evidence about how the respondents perceive the sources of risk related to dual use research, the actions that some of these scientists are taking to reduce the risk of misuse 

 DUAL USE RESEARCH IN THE LIFE SCIENCES of science, and the prospects for acceptance of various policy proposals aimed at reducing the risks of misuse of legitimate life science research, although, as indicated earlier, the results of the survey must be viewed with caution because of the low response rate and possible response bias. Scientists who may be involved in biodefense research or who use select agents, for example, may be more aware of the dual use dilemma and thus more likely to have responded to the survey. In addition, a few of the questions could have been interpreted in multiple ways so that, for example, all “Yes” or “No” responses may not be comparable. Despite these potential problems, the committee believes the data obtained in this study offer valuable insights and new information. Overall, the survey findings suggest that there may be considerable support for models of oversight that rely on the responsible conduct of research and self-governance by the scientific community. The responses also suggest, however, that there is a critical need to clarify the scope of research activities of high concern and to determine the appropriate actions that members of the life sciences community can take to reduce the risk of misuse of science for biological weapons development and bioterrorism. Perceptions of Risk The findings suggest that, on average, the scientists who responded to the survey perceive a potential, but not overwhelming, risk of bioter- rorism and that the risk is greater outside the United States. On average, the respondents believed that there is a 51 percent chance that there will be an act of bioterrorism somewhere in the world in the next 5 years and a 35 percent chance that there will be an act of bioterrorism in the United States in the next 5 years. Three-quarters of the respondents believe that a preference for other means of attack is the primary reason why there have been only a few acts of bioterrorism to date; overwhelmingly, 87 percent of respondents said that they believe that terrorists are not deterred by the threat of being caught and punished. Fewer scientists considered a lack of knowledge (46 percent) or access to equipment (51 percent) or agents (36 percent) to be significant barriers. It may be that one’s perceived risk of such an attack is related to one’s support for taking measures to reduce the risks that life sciences research might be misused. With regard to the chance that the knowledge, tools, or techniques from dual use research will facilitate bioterrorism, the respondents per- ceive a 28 percent chance, on average, of such a bioterror attack within the next 5 years. Half of the respondents thought that if someone wanted to create a harmful biological agent, the Internet would be the most likely place to provide sufficient information for life scientists with college- 

SUMMARY  level training. Other sources of information—articles in scientific journals (40 percent), personal communications (38 percent), and presentations at professional meetings (18 percent)—were considered relatively less likely sources, although on average 45 percent of respondents answered “Don’t Know” to these questions. Actions Taken by Life Scientists in Response to Dual Use Concerns Although the responses to the survey indicate that bioterrorism prob- ably is not perceived to present a serious immediate risk to U.S. or global security, the survey results also indicate that there is already concern about dual use issues among some of the life scientists who responded. Fifteen percent of the respondents (260 individuals out of 1,744) indicated that they are so concerned about dual use research that they have taken actions, even in the absence of guidelines or mandatory regulations from the U.S. government. Some respondents reported that they had broken collaborations, not conducted some research projects, or not commu- nicated research results. The results indicate that more scientists have modified their research activities than some members of the committee expected on the basis of previous reports of manuscripts that have been modified or not published because of dual use concerns. Interestingly, many of the actions that the respondents reported tak- ing to mitigate concerns occurred before the publication stage; much of the behavior change occurred during the research design, collaboration, and early communication stages. Of particular interest and concern to the committee, a few respondents commented on their concerns about for- eigners as potential security risks, which may be reflected in the reported avoidance of some collaborations. The survey results suggest that: (1) some life scientists in the United States may be willing to consider self-governance aimed at the respon- sible scientific conduct for dual use research, and (2) some life scientists in the United States are already acting, even in the absence of govern- ment regulations and guidance, to protect against the perceived risk of misuse of dual use research. Oversight Mechanisms With a proposed oversight framework for dual use research of concern proposed by NSABB in June 2007 now under consideration within the U.S. government, the survey was an opportunity to assess scientists’ atti- tudes toward specific policy options. Many of the respondents indicated that they believe that personal responsibility, including measures such as codes of conduct, could foster a positive culture within the scientific com- 

 DUAL USE RESEARCH IN THE LIFE SCIENCES munity to evaluate the potential consequences of their research for public safety and national security. They also indicated that they believe that individual researchers, professional scientific societies, institutions, and scientific journals should be responsible for evaluating dual use potential of research and/or fostering the culture of scientific responsibility. A majority of those who responded to the survey favored self-gov- ernance mechanisms for dealing with dual use research of concern, such as those proposed by the Fink report (NRC 2004a), rather than addi- tional mandatory government regulations. In addition to the low level of support for greater federal oversight (26 percent), the individual com- ments indicated a belief that increased government oversight of dual use research would be counterproductive by inhibiting the research needed to combat emerging infectious diseases and bioterrorism as well as being potentially harmful to the scientific enterprise more generally. The survey suggests that most of the respondents (82 percent) favor their professional societies’ prescribing a code of responsible conduct to help prevent misuse of life sciences research. However, many respondents (66 percent) did not know whether the societies to which they belonged already had codes that address dual use issues, and some of the societies most frequently cited do not in fact have a code. There was substantially less support (38 percent agree or strongly agree) for a Hippocratic-style oath. The results also indicate potential support for journals having bios- ecurity policies. Yet, most of the respondents did not know if any of the journals in which they have published or to which they have submitted manuscripts have those policies. Moreover, more than half of those who responded to the survey strongly disagreed or disagreed with restrictions on personal communication, altering or removing methods or findings from scientific publications, or limiting publication itself. The survey points to a likely preference for self-governance measures to provide oversight of dual use research. There was substantially less support for mandatory measures that might be imposed by regulation, although the results varied for different policy measures. The results indicate that there may be greater support for restrictions on access to biological agents (just under 50 percent of the respondents said they agree or strongly agree) and certifications of researchers (just over 40 percent of the respondents said they agree or strongly agree) than for any control of scientific knowledge generated from the research or through informa- tion exchange (only 20 to 30 percent of respondents supported these measures). Table S-1 provides a list of the level of support for the various measures addressed in the survey. 

SUMMARY  TABLE S-1  Summary of Results Regarding Support for Measures of Personal and Institutional Responsibility Strongly Agree or Agree Measures of Personal or Institutional Responsibility (or Respond Yes*) (%) Principal investigators should be responsible for the 87 initial evaluation of the dual use potential of their life sciences research. Principal investigators should be responsible for training 86 lab staff, students, and visiting scientists about dual use research. Should professional science societies have codes for the 82* responsible conduct of dual use life sciences research? University and college students should receive 68 educational lectures and materials on dual use life sciences research. Scientists should provide formal assurance to their 67 institution that they are assessing their work for dual use potential. Funding agencies should require grantees to attest on 60 grant applications that they have considered dual use implications of their proposed research. Should scientific journals have policies regarding 57* publication of dual use research? Institutions should provide mandatory training for 55 scientists regarding dual use life sciences research. Greater restrictions should be placed on access to 47 specific biological agents or toxins. Researchers conducting dual use research should be 42 certified. All grant proposals for life sciences research with dual 41 use potential should be reviewed by a researcher’s institution prior to submission for funding. Scientists conducting or managing research should take 38 an oath. Research findings should be classified based on their 28 dual use potential. Dual use research needs greater federal oversight. 26 Certain experimental methods or findings should 22 be altered or removed prior to publication or presentation. Certain biological equipment that is commonly used in 21 life science research should be licensed. There should be restrictions on disclosure of details 21 about the research or its findings through personal communication. There should be restrictions on publication of findings 21 based on their dual use potential. SOURCE: NRC/AAAS Survey of Life Scientists; data analysis by staff. 

 DUAL USE RESEARCH IN THE LIFE SCIENCES The survey results suggest there is support for: 1. Greater oversight that is not federally mandated, 2. Self-governance mechanisms as an approach for preventing misuse of life science research and knowledge, 3. Professional and scientific societies adopting codes of conduct that include dual use research as suggested in the Fink report (NRC 2004a), 4. Establishing and implementing policies for authors and reviewers to consider the dual use potential of research manuscripts submitted to journals. The survey results suggest there is opposition to: 1. Mandatory government regulations to govern the conduct of dual use research and the communication of knowledge from that research; 2. Other mandatory oversight actions, such as oaths or licensing of scientists. Based on the survey results and its own analysis, the committee believes that a basis of support exists within the U.S. scientific commu- nity for measures that, taken together, could lead to the development of a system of self-governance for the oversight of key aspects of dual use research. Education and Outreach A major reason for conducting the survey was to inform efforts for education and awareness-raising about dual use research by providing empirical data on the attitudes of a sample of the life sciences community. In general, the respondents to this survey would likely support educa- tional and outreach activities aimed at raising awareness of the dual use dilemma. The respondents indicated that they supported educational materials and lectures on dual use research for students. They also sup- ported mandatory training by institutions for practicing life scientists regarding dual use research of concern. The survey results also highlight the need to better define the scope of dual use research of concern. Fewer than half of the respondents who indicated that they were carrying out dual use research activities felt that their research fell into one of the seven categories of research of concern specified by the NSABB. The dual use experiments of concern as listed in the Fink report (NRC 2004a) and by the NSABB are all based on microbial research, but other relevant research, such as theoretical research, scenario development, or applied research (e.g., pharmaceutical formulations or 

SUMMARY  neuroscience research) can be of dual use concern. In their individual comments, a number of respondents stressed the difficulties of defining dual use, as did participants in the focus groups used to develop the survey. Clearly a better understanding of the scope of dual use research of real concern would help any educational or outreach activities aimed at raising the awareness of life scientists so that appropriate actions can be taken. Based on the survey results and its own analysis, the committee believes that there is support for mandatory education and training about dual use issues, most likely as part of ethics and responsible conduct of research training. RECOMMENDATIONS The committee believes that the survey raises several hypotheses that merit further research about the views of life scientists about oversight policies and education and outreach efforts to address concerns about dual use issues in the life sciences. In particular, based on the survey results and its own deliberations, the committee offers the following recommendations: Oversight, Education, and Outreach 1. Explore how to continue and to expand the dialogue within the life sciences community about dual use research of concern. 2. Explore ways to provide guidance to the life sciences community about appropriate actions that can be taken to protect against the misuse of dual use research. 3. Seek to better define the scope of knowledge in the life sciences that may be at greatest risk for misuse and to provide the life sciences community with criteria for recognizing dual use research of concern. 4. Encourage journals that have biosecurity policies or plan to adopt them in the future and the professional and scientific societies that have or plan to develop codes of conduct to communicate those policies more effectively. Further Research 1. Examine the effectiveness of existing educational programs and how they can be enhanced and focused. 2. Seek to extend educational and awareness-raising efforts being conducted in the United States to the broad international scientific community. 

10 DUAL USE RESEARCH IN THE LIFE SCIENCES 3. Examine how education and outreach activities can be developed to guide the life science community’s response to concerns about dual use research so as to ensure that actions taken by the community are appro- priate and contribute to advancing scientific knowledge while protecting national security. 4. Conduct additional surveys, interviews, or focus groups of U.S. life scientists that better represent the full community, with higher response rates than the current study was able to achieve, and the ability to assess potential bias, in order to gain •  a better understanding of the awareness of a broader range of U.S. life scientists about dual use research of concern and the measure that they would support to reduce the threat that research in the life sciences could be subverted to do harm; •  a better understanding of the types of behavioral changes being made in response to dual use concerns to determine if actions by life sci- entists are contributing to national security or harming scientific research; such research is critical given the actions that the current survey suggests are being taken; •  more detailed information about the types of changes scientists are making or scientists’ thoughts about dual use issues, experiments of concern, and select agents; •  a better understanding of scientists’ experiences with education on this topic and their views about the content and delivery of educa- tional and training materials. 5. Conduct additional surveys of life scientists outside the United States that would enable comparisons of attitudes toward dual use research of concern and inform educational and outreach programs so that they can be effective on a global scale. Such knowledge could also facilitate international discussions of potential measures to address dual use concerns.