3
REGULATORY AND ADVISORY PERSPECTIVES
The workshop’s first panel session featured four presentations from representatives of agencies that play a role in regulating animal welfare. Susan Silk, Director of the Division of Policy and Education in the NIH Office of Laboratory Animal Welfare, and Carol Clarke, Research Program Manager at the U.S. Department of Agriculture’s (USDA) Animal and Plant Health Inspection Service, provided the NIH and USDA’s perspectives on performance standards. Gilly Griffin, Director of Standards at the Canadian Council on Animal Care, discussed Canada’s view of performance standards, and Judy MacArthur Clark, Head of the United Kingdom’s Animals in Science Regulation Unit, spoke about the U.K. perspective and how it fits within the European regulatory system. An open discussion moderated by Planning Committee Co-Chair David Kurtz followed the four presentations.
The NIH Office of Laboratory Animal Welfare (OLAW), explained Susan Silk, oversees the welfare of research animals according to the standards in the Public Health Service (PHS) Policy on Humane Care and Use of Laboratory Animals (PHS, 2015). The PHS policy, first published in 1985 and revised in 2015, incorporates by reference the 8th edition of the Guide. It is a tool that enables federal agencies to give legal effect to materials published elsewhere rather than creating a new set of technical standards simply to serve regulatory purposes. Since the Guide provides the best-practice standards for biomedical animal care and use programs,
_________________________
1 This section is based on the presentation by Susan Silk, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.
PHS policy requires PHS-funded institutions, including those receiving NIH funding, to base their programs on the Guide.
OLAW’s interpretation of the Guide holds it to be a starting point for how to operate a quality program. The PHS oversight system relies on self-monitoring, self-regulation, and self-reporting, Silk explained, because the NIH is a scientific, not a regulatory agency. This grassroots approach, established by Congress in 1985, supports the development of a culture of compliance and caring. The U.S. Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training were promulgated in 1985 (IRAC 1985), and an accompanying report provides Congressional support to OLAW’s reliance on IACUCs to oversee PHS-funded animal activities, Silk explained.
PHS’s initial policies and guidelines derived directly from the 6th edition of the Guide, and the evolution of these policies and the Guide has resulted from 26 years of cooperation and careful work supported by documentation and information sharing within the animal research community, said Silk. In 1997, for example, a symposium on performance standards in animal welfare, convened after the publication of the 7th edition of the Guide, led to the recommendation that every performance standard required associated measures of assessment; development of such measures is an activity ongoing today.
In 2007, the ILAR Council determined the Guide needed updating, and a letter sent to NIH noted the necessity of preserving and perhaps even increasing the performance-based nature of this document. As a result, NIH and other organizations provided funds for ILAR to update the Guide. The statement of task to the expert panel charged with revising the Guide provided the following instruction: “Where scientifically warranted, the guidance and recommendations of the 1996 Guide will be changed to reflect new scientific evidence, while maintaining the performance standards of the 1996 Guide.” NIH, which played no role in developing the Guide, took about a year to adopt the Guide after its publication and requires all PHS-Assured institutions to implement the new Guide.
The Guide, Silk noted, is not an operations document, but rather a collection of experts’ wisdom assembled “under the benevolent leadership of the National Academies and ILAR, which serves the interest of nonpartisan scientific integrity.” She clarified a misunderstanding as to whether federal agencies strongly support the use of performance standards. “Performance standards are the most important component of the infrastructure of PHS oversight of animal programs, and OLAW stands behind this statement. We expect IACUCs to meet their responsibility to ensure humane animal care and use while advancing quality scientific
research through the use of performance standards in the IACUC’s oversight of institutional animal programs,” said Silk. These statements, she added, are posted on the OLAW website (http://grants.nih.gov/grants/olaw/positionstatement_guide.htm#performance).
OLAW expects PHS-Assured institutions to implement performance standards by enlisting diverse expertise to develop outcome-based performance standards that enhance the quality of animal care and use programs, and to apply professional judgment and experience to develop the policies and procedures needed to maintain a quality program that provides humane care to animals. The 8th edition of the Guide, said Silk, supports performance standards that were already in use by most programs, and as a result, OLAW expected few institutions would need to make major changes to quality programs as they adopted the new Guide. Institutions not meeting those standards were given a year to develop a reasonable plan and schedule for implementing the Guide. OLAW has been holding webinars to help the community understand its expectations regarding well-established performance standards and to address a variety of concerns among members of the community. OLAW has posted the webinars on its website (http://grants.nih.gov/grants/olaw/educational_resources.htm) and is in the process of indexing the comments and questions prompted by these webinars. The website also contains a Frequently Asked Questions (FAQ) section addressing some of the key issues the community has raised, such as on how to assess the adequacy of housing and how to develop performance standards for housing.
The Animal Welfare Regulations (AWR) mix performance and engineering standards to show there is more than one way to achieve animal welfare goals, explained Carol Clarke. The regulations, for example, require the IACUC to review animal care and use programs and inspect animal facilities every six months, a prescriptive engineering approach with no exceptions and no special circumstances. However, the regulations allow the IACUC to determine the best means of conducting these evaluations. The defined goal is to get inspections done with a flexible, performance approach.
_________________________
2 The section is based on the presentation by Carol Clarke, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.
Housing is another example of where engineering standards prescribe a minimum space requirement for each species. The regulations, however, provide the opportunity to develop innovative cages that may not meet the space requirements but do provide sufficient space and opportunity to allow for provide species-specific behavior. This latter provision creates a flexible, performance-based approach.
Clarke pointed out two important phrases appearing throughout the regulations - “in accordance with established veterinary and medical practices,” and “currently accepted professional standards” – that are important because standards and practices evolve. For example, in the 1980s animals did not routinely receive postoperative pain medications after surgery, but this is now a required veterinary practice. At one time, toys were the central feature of efforts to promote the psychological wellbeing of nonhuman primates, but today’s practices require employing innovative social housing and behavioral monitoring. This kind of evolution is important, said Clarke, because it has a direct bearing on what constitutes compliance.
She then dispelled some of the myths associated with AWR. Some believe these regulations are based on specific engineering standards and have not yet incorporated performance standards. To show why this is not the case, Clarke cited the following example: regulations regarding operative procedures on non-rodents call for surgery to be performed only in facilities intended for that purpose and maintained under aseptic conditions. One facility, however, was unable to build a dedicated surgery suite, so it asked USDA if it could use mobile laminar flow hoods for such surgeries. Upon review, USDA ruled that laminar hoods met the requirements of being clean, aseptic, and dedicated to surgery, and approved this request.
Another example concerned naked mole rats, for which there is no species-specific standard. These animals are eusocial and live in underground colonies in a caste system similar to that of ants. USDA ruled that a network of plastic pipes designed to mimic the underground conditions in which these animals live in nature fulfilled the requirement that housing must enable the animals to express their normal social behavior.
The role of the USDA inspector, Clarke then explained, is to ensure a facility is in compliance with the regulations. The inspector understands compliance can be achieved in many ways because of the built-in flexibility of the regulations and variation among facilities. The inspectors can draw upon expertise within the Animal and Plant Health Inspection Service (APHIS), including diplomates from the American College of
Laboratory Animal Medicine (ACLAM) and the American College of Animal Welfare (ACAW), as well as other subject-matter experts. They can also use other references and standards, such as the Guide and taxon-specific publications. Clarke encouraged everyone to join the online APHIS Stakeholder Registry to receive updates and notice of future developments.
The Canadian Council on Animal Care (CCAC), established in 1968, oversees the ethics of animal experimentation in Canada, explained Gilly Griffin. It is a non-legislated, peer-reviewed system acting on behalf of the people of Canada. Griffin said there is true public involvement in every aspect of its programs, including members of the public who serve on the Council and on CCAC animal care committees, the equivalent of U.S. IACUCs. The CCAC program, said Griffin, is unique because it both sets standards and maintains them through assessments and certifications. The strong link between assessment and certification programs, she said, creates a learning system that allows the CCAC to adapt to new research realities, incorporate them into the standards, and get information back into the research community.
Canada’s approach to its guideline documents is to create modules or chapters that can be updated more regularly than large volumes. Modules are developed by expert subcommittees in a process overseen by CCAC’s standards committee (Figure 3-1). After reviewing all of the relevant scientific evidence, an expert subcommittee develops a preliminary draft of a new guideline. Several review steps follow, producing multiple drafts posted on CCAC’s website for public comment, followed by a final review. At each point in the process, the standards committee acts as a gatekeeper to ensure each draft is ready for review and aligns with other CCAC guidance documents. The final step is approval by CCAC’s board of directors, at which point the guidance document is published. The process takes a long time, Griffin said, but it allows for significant buy-in from the community.
_________________________
3 This section is based on the presentation by Gilly Griffin, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.
FIGURE 3-1 Canada’s animal welfare guidelines development process
SOURCE: Griffin slide 5
This last point is important, she said, because the guidelines are written for several audiences, including:
- Investigators, who write protocols and want to know how their animals are going to be managed;
- Animal care committee members, who review those protocols and ensure they comply with CCAC guidelines and policies;
- Veterinary and animal care staff, who maintain the health and well-being of those animals; and
- Assessment panels, who use the guidelines as their standards during assessment visits and make recommendations based on the guidelines themselves.
As with the Guide, Canada’s guidelines use “must” for mandatory requirements and “should” to indicate an obligation for which any exception must be justified and approved by an animal care committee based on strong scientific justification for changing that requirement. The guidelines, said Griffin, evolve through an iterative process reflecting how good practices develop into best practices, and they serve as a framework that enables institutions to develop best practices. As an aid for Canada’s institutions, CCAC maintains the 3Rs microsite at which it curates best practices documents. CCAC expects these best practices to be published and tested over time and serve as the basis for the next revision of the relevant guidelines.
CCAC’s assessment panels, comprising scientists, veterinarians, and community representatives, are responsible for assessing performance standards. These panels do not function as an inspectorate, explained Griffin, but rather as a vehicle to encourage learning and for sharing best practices. After each assessment, the panel prepares a report with recommendations based on CCAC’s guidelines and policies, and because CCAC uses a performance approach, these assessments rely heavily on the experience and range of expertise on the panels.
One result of CCAC’s historic reliance on a performance approach, Griffin noted, is that the guidelines tend to be a mix of performance standards and more specific requirements. For example, the CCAC guideline on the care and use of farm animals in research, teaching, and testing includes a performance standard stating that flooring should provide a dry, comfortable lying surface, it should allow animals to go through their normal movements and postural changes without slipping, and it should not result in injuries. This performance standard does not specify the flooring material, allowing institutions to choose based on their local circumstances and the types of barns they have. However, the guidelines do include a specific requirement for dairy cattle housed in free stalls mandating at least one stall for each cow within the group. This requirement was based on scientific evidence showing that the more cattle are able to lie down, the lower the chance they will develop hock injuries and lameness while at the same time improving milk production.
As another example, Griffin discussed CCAC’s 2010 guidelines on euthanasia of animals used in science. These guidelines include ten general guiding principles, but also permit animal care committees to accept new methods that conform to these principles. The guidelines include a summary chart of acceptable methods of euthanasia for experimental animals that meet those ten principles, as well as six methods that are conditionally acceptable.
The most recent CCAC guideline, published in March 2015, covers training of personnel working with animals in science and it emphasizes performance standards. The guideline makes institutions responsible for documenting that all personnel involved with animals have the appropriate knowledge, skills, and competency to perform their required tasks, but the document does not prescribe how to deliver programs or record training. As an adjunct to the guideline, CCAC provides a recommended syllabus so that institutions can be certain of what knowledge and skills they need to teach.
CCAC’s experience, said Griffin, has been that guidelines often require additional supporting materials, such as implementation tools to help
investigators and animal care committees determine what a performance standard means for the particular animals kept at their institutions. With the euthanasia guidelines, for example, CCAC provides a publication documenting the effects of the different euthanasia methods on research results to help animal care committees look critically at the evidence on the impact of euthanasia methods on common research goals. For the latest guidelines on training, CCAC might provide some information on how to assess and record competency, including a template institutions might use verbatim or as a guide for developing their own way of assessing and recording competency.
THE UNITED KINGDOM AND EUROPEAN PERSPECTIVE4
European Union (EU) regulations for the protection of animals used for scientific purposes are stated in EU Directive 2010/63, Annex III (European Parliament, 2010), which lists mandatory standards for application throughout EU member states, explained Judy MacArthur Clark. Referenced in that directive is an EU Commission recommendation based on a document known as Appendix A (Council of Europe, 2006), which is advisory in nature and not mandatory. These two documents – Annex III and Appendix A – need to be referenced with regard to how MacArthur Clark and her team apply EU legislation to U.K. legislation. The result is a set of mandatory standards in compliance with Annex III and what she called delivered advice based on Appendix A.
Regarding U.K. legislation, some parts of the mandatory standards are applicable to all species, and in general, these are performance standards. Other parts of the standards are species-specific, and these tend to be engineering standards. These mandatory standards are published in the Code of Practice for the Housing and Care of Animals Bred, Supplied, or Used for Scientific Purposes. The advice component of the U.K.’s code of practice is based largely on Appendix A, but it includes information on how to achieve the performance standards as well as some engineering standards.
In the United Kingdom, engineering standards are defined, measurable parameters with a range of appropriate values. Engineering standards can apply to cage sizes, temperature ranges, photoperiod, trough length, and perch length, for example. Performance standards are outcomes-based parameters, and MacArthur Clark gave two examples:
_________________________
4 The section is based on the presentation by Judy MacArthur Clark, and the statements are not endorsed or verified by the National Academies of Sciences, Engineering, and Medicine.
the exposed surface of walls and floors should be made with a material resistant to heavy wear and tear; and noise levels, including ultrasound, shall not adversely affect animal welfare. Engineering standards, she continued, serve as a hard bottom line and tend to be the welfare safety net for individual species. As such, they set clear expectations for both institutions and equipment manufacturers, and they satisfy public expectations. Historically, engineering standards have been the model for compliance monitoring, but in general they are no longer what regulators focus on with regard to compliance. Performance standards reflect the fact that every institution and every experiment is different, and they provide the flexibility needed to accommodate those differences. MacArthur Clark reiterated what the other speakers had noted: it is not essential to mandate how good animal welfare should be achieved given the many ways this can be accomplished.
The challenge with performance standards, she said, is to agree on what is acceptable in terms of animal welfare and who defines the performance outcomes. While growth and reproduction can serve as performance outcomes, for example, there are situations where growth and reproduction do not necessarily measure good welfare. “We need to be open-minded and think carefully about how we are going to measure good welfare,” said MacArthur Clark.
U.K. and European regulators, she said, have gone to great lengths to involve a broad range of stakeholders in thinking about how to best measure animal welfare, partly to provide individual EU member states with the opportunity to share ideas and raise their own practices to best practice standards. International expert panels, run by the European Commission, help develop some understanding of performance outcomes, and MacArthur Clark’s unit in the United Kingdom develops advisory notes and codes of practice. Often, there are diametrically opposed views on the need for these notes and codes of practice, but getting people with diverse views to talk about their differences and to better understand the role of these advisory documents results in advisories that have broad support.
The cage-side view of the animal care staff, veterinarians, and inspectors is equally important, said MacArthur Clark. Home Office inspectors visit facilities on a monthly or even more frequent basis, depending on the size of the facility, and they serve both as inspectors and as part of the advisory system to share performance outcomes. Strategy and oversight are provided by the Animal Welfare and Ethical Review Boards, the U.K. equivalent of the IACUC. Guidance, she added, helps develop a shared understanding of what is or is not acceptable, and
it provides the flexibility to allow practice standards to evolve. Regulations, on the other hand, are inflexible and require an Act of Parliament. She noted that empowering and assisting institutions with benchmarking drives institutions in their own cycles of self-improvement.
MacArthur Clark explained that her office recognizes that in a normal distribution curve there will be some situations that are at the extremes of the curve and there can be good justification for those situations. In that respect, she said, guidance for performance standards describes what is normally considered appropriate or suitable for a particular species, but also allows for exceptions provided animal welfare standards are maintained. Guidance includes suggestions for enrichment for each species, suggested temperature and humidity ranges, typical species housing needs, typical social needs of a species including those for different sexes and ages, and species-specific dietary advice. All of these items include links to background papers and evidence.
The U.K. Code of Practice contains both engineering and performance standards that are mandatory today and standards that will be mandatory as of January 2017. It also contains non-mandatory advisory information and a bibliography listing a selection of literature supporting that advice. The Code of Practice was published at the end of 2014, and since its publication MacArthur Clark and her team have learned a number of lessons:
- Non-mandatory advice is useful for guidance on how to approach mandatory outcomes and for setting future expectations.
- Users and regulators appear comfortable with the idea that this advice is flexible.
- There is an increased emphasis on institutions justifying their decisions and strategies.
- Institutions prefer performance standards and would like fewer engineering standards.
- Equipment manufacturers and animal welfare groups would like more engineering standards.
Regulators, said MacArthur Clark, are trying to balance these lessons in a way that works best for the animals while recognizing there are other interested perspectives at play. She believes, she said in closing, that engineering standards are providing a welfare safety net for the animals and reassurance to the public, but performance standards provide the opportunity to put animal welfare first. Both types of standards need to be balanced into an effective system, with guidance as an essential
component of the end product to ensure shared interpretation and expectations of what is acceptable performance.
To start the discussion, David Kurtz asked the panelists for ideas on how to publish both positive and negative experiences from trying to establish performance standards. Clarke responded that she heard from a group of researchers that there exists an unofficial venue where investigators discuss and share information about what does not work. These researchers told her they were reluctant to make this an official publication, and Clarke wondered if the roundtable could serve as an official venue for resurrecting this idea and getting the community to think about publishing negative results. Patricia Turner said the important characteristic of any data, positive or negative, is that it must be generated by rigorous studies. “We should have a mechanism for sharing information, but we also need to recognize a certain standard for that information. If a study is performed well, even if the results are counter to the hypothesis perhaps that was the initial basis for that work, it should still be publishable,” said Turner. Griffin agreed with Turner’s statement.
Silk said that OLAW is helping its community implement new standards with significant changes to previously approved animal protocols. OLAW is also encouraging institutions to publish significant changes to their practices on the IACUC Administrators Association website. MacArthur Clark agreed with Kurtz’s idea of sharing negative results and said that the field needs a mechanism by which investigators can freely exchange their experiences and get feedback from others in the field, perhaps in the form of a wiki-type environment.
Paul Locke, from Johns Hopkins Bloomberg School of Public Health, asked Turner to clarify her use of the term practice standard. Practice standard, she replied, refers to what is acceptable in terms of how procedures are done within a facility, and it consists of the appropriate skills, treatment, and attitudes toward animals that occur within an institution. Performance standards, she said, may contribute to the practice of laboratory animal science in the long run and therefore to practice standards that are acceptable.
Locke then asked the panelists for ideas on how to fully engage the appropriate communities and the public to get new performance standards discussed, vetted, and validated.
In Canada, said Griffin, this is done via the assessment panels that share best practices with the institutions they visit. “Our assessments are
a formative experience and an opportunity for institutions to ask questions and get advice and understanding from the panel members, who then go back to their own institutions,” said Griffin. Panel members have told her that these assessments have been great learning experiences for them as well as for the institutions they assessed. Griffin also reiterated that CCAC’s 3Rs microsite is a place for curating and disseminating best practices, though she acknowledged that operating this site is resource-intensive.
MacArthur Clark said there is no one approach for sharing information that will work for everyone. One barrier her agency faces is that the environment in which it works has been rather secretive and confidential, which does not create an atmosphere that inspires investigators and institutions to share best practices. She is encouraged, though, by the drive over the past four to five years in the United Kingdom to create more transparency, but currently, unlike in Canada, inspectors in the United Kingdom operate under confidentiality and do not share best practices among institutions. One approach her office is considering is to reassign some of its resources to conduct what she called horizontal inspections, which would be thematic rather than institutional and would allow her office to use its annual report and public meetings to disseminate anonymized best practices.
Silk said she is in favor of looking to the people who have contact with the animals every day for ideas to improve what they do to the benefit of the animals. Clarke added that USDA promotes and supports meeting for sharing best practices in an open venue.
Robert Wurtz from the NIH, questioned why regulatory authority over animal experiments in the U.S. was divided among multiple agencies. Clarke explained that every U.S. regulatory agency supports a specific act of Congress, and her agency supports the Animal Welfare Act. Silk noted that USDA is a regulatory agency but NIH is a scientific agency and compliance with its guidelines is entirely optional. “Our guidelines are not laws. We conduct oversight to help you stay in compliance with guidelines that enable you to legally receive appropriated funds,” said Silk. NIH has had a Memorandum of Understanding with USDA and the Food and Drug Administration since 1985, she added, enabling the agencies to work closely with one another to harmonize expectations. For example, when NIH publishes its FAQs, it reviews with USDA those questions that concern shared issues. In addition, the two agencies have harmonized their published guidelines.
Regulations impacting the United Kingdom are set at a European level but are then implemented by U.K. legislation, explained MacArthur
Clarke. She noted the implementation process can be tortuous, and in the end, while in theory the regulations of all the EU member states are aligned, in practice each EU member state implements the regulations differently. This creates challenges, she added, when scientists move between countries. Canada, said Griffin, has provincial rather than federal legislation, which requires her office to consider 10 different pieces of legislation when trying to implement a national system of regulation.
An online workshop participant submitted a question asking how to qualify the outcome of no observable effect when using performance standards to assess welfare in the context of trying to reach a good effect or avoid a bad effect. OLAW, replied Silk, encourages its stakeholders to ask questions and it allows its PHS-Assured institutions to suggest different approaches to achieve the objective in a “should” statement in the Guide. “If the welfare of the animals is the same or better under the mechanism that they propose, then we would be fine with them going ahead with that,” said Silk.
David Anderson, from the University of Washington, asked Silk to comment more on the challenge of being able to accurately assess impact. “Do we really require a demonstrable good effect or is a neutral effect something that is left up to the institution to judge?” he asked. Silk replied that this is a hypothetical question that everyone on the panel faces. “You are not applying performance standards in a large hypothetical universe. You are applying them to particular animals in a particular situation,” said Silk. “We would begin to unpack this question by asking specifics about the animals.” MacArthur Clark reminded the workshop that the basic operating principle is “first, do no harm,” which is why it is important to always assess the outcome of any changes to the performance standard to show there is in fact no harm being done to the animals. Griffin noted the need to put more emphasis on developing good guidance for welfare assessment, something with which institutions often struggle.
Another question submitted online asked MacArthur Clark to comment on the Stop Vivisection European Citizens’ Initiative currently in front of the European Parliament and the impact it may have on performance standards in the European Union. MacArthur Clark replied that the United Kingdom is in a pre-election period so she cannot comment about what the future may hold. She did note that it took two years of negotiation involving the European animal welfare bodies as well as the science communities to achieve the current directive, which has been implemented by all 28 member states. The Citizens’ Initiative, she explained, is asking for that directive to be abrogated.
Cathy Liss, with the Animal Welfare Institute, said the U.S. system for assuring the welfare of laboratory animals is broken, in her opinion. It is broken, she said, because the safety net is not where it should be given the disparity between the Guide, which she said has moved forward and embodies a more thorough assessment, though lacking in specificity that the welfare community would like to see, and the enforcement mechanisms residing with USDA, which does not cover the majority of the animals used in research. She claimed that USDA standards are outdated and do not reflect current knowledge that would improve animal welfare.
She noted, too, that while there is a requirement for primates, there is no requirement for improving the welfare of the other animals. She then asked if there was a chance of revising the standards and revising the Animal Welfare Act so that it embodies more of the principles within the Guide.
Clarke replied that there is a mechanism to make revisions to the Animal Welfare Act and that some revisions are going on today. The process, she explained, involves submitting a petition that USDA would then put out for comment. Once the comments are evaluated, USDA would determine if there is a need to go forward and change the standards to which Liss was referring. Though the standards were written some time ago, Clarke said USDA is not standing in the way of progress and is using provisions in the regulations relating to currently accepted veterinary practices to give it flexibility and to allow the standards to evolve. Silk said she does not believe the animal welfare system in the United States is broken. “I think our grassroots approach is providing the best welfare for the animals,” said Silk.
Liss asked Silk how OLAW, in the absence of an inspection system, ensures institutional compliance apart from the submission of an assurance. OLAW, Silk reiterated, relies on self-monitoring and self-reporting for every institution receiving PHS funding. The assurance document describes how an institution’s animal care program operates and achieves compliance with PHS policy, which incorporates by reference the standards of the Guide. Her division within OLAW also helps institutions with education and policy interpretation. “We believe that through education, we prevent non-compliances from occurring,” she said. Non-compliances are reported by the institution, by outside parties, or by individuals within the institution. OLAW’s compliance division then works with the institution to resolve the noncompliant situation to ensure people and animals are safe and then enacts policies to prevent that situation from occurring again. “I would submit, and this is my opinion, that perhaps our system is more effective than an inspection system,
which is generating a snapshot in time. We are trying to make sure that everyone is working toward a culture of humane care at all times.”
Donna Matthews Jarrell, from the Massachusetts General Hospital, said she appreciated the panelists’ example of working with institutions to embrace this new approach which emphasizes performance standards. She wondered, though, what these institutions have done after they have evaluated a performance standard, particularly with regard to the schedule on which they continue reviewing these standards. Clarke responded that in the case of the laminar flow hood example, once her office had agreed there were no adverse effects, it became the IACUC’s responsibility to monitor the situation on a six-month basis. Silk said OLAW takes the same position – it expects programs to be evaluated semi-annually while deferring to the IACUC’s authority to make that decision. If complications did arise during surgery and sepsis developed,
Diane Gaertner from the University of Pennsylvania added, the veterinarians and veterinary technicians who would be monitoring the animals daily would be reporting to the IACUC if there was any doubt that aseptic conditions were being maintained.
MacArthur Clark stressed the importance of the individuals who work in an institution. If the animal care staff is well-trained, well-qualified, and empowered to speak up, there will be adequate protection of animal welfare. Even in the U.K., which has a national inspection system with inspectors going regularly to visit institutions, the strength of the system lies with the individuals within those institutions. Inspections, in fact, are called visits because although there is an inspection component, the inspectors are in essence monitoring the performance of monitoring framework built around the animal care staff.
Steven Niemi, from Harvard University, said he was intrigued by the flexibility USDA affords institutions, and he asked Clarke how the community can find out about these ad hoc interpretations so they may be applicable in similar circumstances at other institutions. Clarke replied these decisions are based on two factors: what is in the literature and what works. If there are no adverse effects to the animals, her office encourages investigators to publish their findings to get the information out to the broader community. Her office has held meetings with research facility veterinary medicine officers at which this topic was discussed and she hopes there will be more such meetings at which information can be shared and discussed.
Saverio Capuano, from the Wisconsin National Primate Research Center, noted that his institution has 194 standard operating procedures, and the University of Wisconsin has a plethora of animal care policies
approved by the IACUC. As a result, his facility’s performance standards are not compiled in one place or are even written down in many cases. “When you come to visit me and when I tell you that some of my performance standards are in my head or in the heads of everyone who works at the Primate Center, how do you respond? How does USDA respond? How does OLAW respond to the fact that my performance standards are just everywhere?” he asked. Silk replied that the IACUC determines where performance standards are kept. Clarke added that when a USDA inspector comes and asks why there is a deviation from the regulations, they expect an answer. If the answer is logical, if there is documentation showing the IACUC has approved the modification, and if there are no adverse effects, then there will not be a problem.
