Online Issues

Journal Vol 49 (4)

A Guide to Risk Assessment in Animal Care and Use Programs: The Metaphor of the 3-Legged Stool

Richard C. Van Sluyters

Richard C. Van Sluyters, OD, PhD, is Professor and Associate Dean for Student Affairs of the School of Optometry as well as IACUC Chair at the University of California in Berkeley.

Address correspondence and reprint requests to Dr. Richard C. Van Sluyters, 690 Minor Hall, University of California, Berkeley, CA 94720-2020 or email rcvs@berkeley.edu.

Abstract

A high-quality contemporary animal care and use program (ACUP) is a multifaceted dynamic system with three distinct organizational entities—the institutional official (IO), attending veterinarian (AV), and institutional animal care and use committee (IACUC)—each of whose authorities and responsibilities are mandated by a complex set of laws, regulations, policies, and guidelines. The 3-legged stool is presented here as a metaphor for the properly constituted and functioning ACUP, with the legs of the stool representing the IO, AV, and IACUC: when one component is weakened, the stool remains standing but tilts—the ACUP remains in compliance with animal welfare standards but is at risk of failure. Mechanisms for assessing an ACUP's strengths and weaknesses include both external evaluators, such as the US Department of Agriculture, Office of Laboratory Animal Welfare, and Association for Assessment and Accreditation of Laboratory Animal Care, and internal evaluators, such as the IACUC, AV, and IO. However, each of these evaluators primarily assesses whether aspects of the ACUP are in compliance with current standards; they do not necessarily cite weaknesses in a compliant program. This article stresses the need for ACUPs to undertake a self-assessment of risk and outlines a number of ways programs can recognize their risks, with examples for each of the three components, which are characterized as either "weak" or "overzealous" in meeting their mandates. I caution against the use of a legacy of compliance as the sole means for evaluating an ACUP's strength, and instead promote the value of rigorous risk assessment/mitigation to ensure that a program is both strong and resilient.

Key Words: animal care and use program; attending veterinarian; compliance; institutional animal care and use committee; institutional official; 3-legged stool; postapproval monitoring; risk assessment

Background

Awell-functioning animal care and use program (ACUP¹) is a dynamic system in which three distinct organizational entities, each staffed by highly trained and skilled personnel, work synergistically within a complex framework of animal welfare laws, regulations, policies, and guidelines to ensure the health and proper care of live vertebrate animals for humane use in essential biomedical research and teaching. The ACUP's direct clients or beneficiaries are the animals in its care and the instructors and researchers who use them, and its indirect beneficiaries are the program's host institution and society as a whole.

The three organizational units that constitute an ACUP are the institutional official (IO¹), attending veterinarian (AV¹), and institutional animal care and use committee (IACUC¹). Like the executive, judiciary, and legislative branches of the US government, each has its own mandated authorities and responsibilities (but they also have partially overlapping roles) and their actions are restrained by a system of checks and balances (for reviews, see Bayne and Garnett 2008; Silverman et al. 2007). When each organizational unit is properly constituted and functioning correctly, the ACUP is strong and resilient, with the capacity to detect and resolve problems before they compromise animal welfare, research findings, or instructional quality.

Risk Assessment in Animal Care and Use Programs

Institutions seeking to establish and maintain quality ACUPs must be capable of assessing the strengths and weakness of their programs, with the goal of building on the former and mitigating the latter. At first glance, such efforts may seem fairly easy as the animal welfare laws, regulations, policies, and guidelines that govern an ACUP include an oversight system for monitoring compliance.

Internally, the IO, the AV, and especially the IACUC are each charged with monitoring the ACUP's performance (Bayne and Garnett 2008; Silverman et al. 2007). There are also a number of external evaluators of the ACUP. Veterinary medical officers from the US Department of Agriculture (USDA) conduct unannounced inspections of registered research facilities for compliance with the US Animal Welfare Act and its implementing Regulations (AWAR). In addition, the National Institutes of Health's Office of Laboratory Animal Welfare (OLAW) requires institutions that receive Public Health Service (PHS) funding to submit an "assurance" that the institution will comply with the PHS Policy on Humane Care and Use of Laboratory Animals (NIH 2002), the National Research Council's Guide for the Care and Use of Laboratory Animals (NRC 1996), and the AWAR. The USDA and OLAW also each require institutions to regularly submit reports documenting their compliance with relevant animal welfare standards and to report significant instances of noncompliance. Finally, institutions that seek accreditation from the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) are subject to triennial site visit evaluations by teams of experts in ACUP compliance.

With all of these internal and external oversight systems for monitoring compliance, it might seem there is no shortage of ways for institutions to detect risks in their ACUPs. However, a defining characteristic of each of these forms of oversight is that their purpose is to point out demonstrable instances of noncompliance, not to warn institutions of weaknesses in the ACUP that could lead to noncompliance. Thus IACUCs are charged with reporting on significant animal welfare issues that have occurred or have not been corrected in a timely fashion. USDA inspectors cite the section of the Animal Welfare Regulations that has been violated, and institutions report to the USDA significant animal welfare issues that have not been corrected in a timely fashion. OLAW must be notified promptly of significant problems that have occurred, and AAALAC site visitors are trained not to mention their observations of conditions that could lead to noncompliance with the Guide but only to report those that have resulted in citable noncompliance. Institutions that believe these traditional methods of monitoring their ACUPs fulfill their needs for risk assessment are engaging in risky behavior!

Ideally, each of the three internal components of the ACUP should conduct its own form of risk assessment continuously, for example by training its personnel, availing itself of the latest information at meetings and in the literature, and developing best practices and constantly striving to improve standards of care. Top-notch ACUPs do indeed subscribe to such practices, but even they can find it difficult to perform cross-component risk assessment. For example, although the IACUC is charged with monitoring the performance of the overall ACUP—including the AV, the IO, and even the IACUC itself—its focus, as stated above, is usually the detection of compliance problems after the fact rather than the mitigation of risks before they cause failure. In the absence of a serious breakdown, how many IACUCs are willing to cite the IO for a perceived failure to understand his/her overall authority and responsibility in the ACUP or provide adequate institutional support? How many IACUCs would cite the AV for appearing to provide inadequate assistance to investigators during protocol preparation? Or cite themselves for not devoting enough time to conduct a thorough protocol review?

Programs that recognize these shortcomings seek outside expert advice on the strengths and weaknesses of their ACUPs (often just before they are due to be evaluated by a team of AAALAC site visitors!). Consultants who are acknowledged experts in various aspects of animal care and use are brought in to provide their opinions not just on observed compliance failures but also on practices that weaken the ACUP and impair its ability to detect and resolve problems before they compromise animal welfare.

The goal of this article is to help institutions perform this kind of effective risk assessment on their own, by providing guidance—through examples of situations or practices—for detecting weaknesses in their ACUP before they result in failed compliance. The examples cited are drawn from observations during my 12-year tenure as a member of AAALAC's Council on Accreditation and almost 23 years as chair of my institution's IACUC. Although my perspective is drawn largely from experience with academic settings, most if not all of what follows also applies to ACUPs in corporate or governmental institutions. How (and of course whether!) institutions choose to implement the suggestions presented here for detecting risks in their ACUPs may vary. One possibility is to keep them in mind when preparing the reports that OLAW and the USDA require institutions to submit at regular intervals.

Metaphor of the 3-Legged Stool

Just as America's founders exercised wisdom in creating the three branches of the US government, with their partially overlapping authorities and responsibilities and a strong system of checks and balances, so did the designers of the animal welfare laws, regulations, policies, and guidelines that govern the contemporary ACUP. When the three major components of each system are properly constituted and functioning, the system is "in balance."

A useful way to represent this balance is the metaphor of the 3-legged stool (Figure 1). Each leg of the stool represents a component of the ACUP and together they support the program. When each leg is strong and tall, the program is held high and level and is capable of withstanding unforeseen challenges (Figure 1A). A program is "imbalanced" when there is an inherent weakness in one of its components—one of the stool's legs is either too long or too short (Figure 1B). Although the stool remains standing (i.e., the ACUP is still compliant), it suffers from a fundamental flaw. The weakened component's compromised ability to detect and resolve problems before they lead to failure threatens the integrity of the entire program, and unforeseen challenges that target the weakened component may not be withstood.

Figure 1 This figure illustrates the metaphor of the 3-legged stool for the animal care and use program (ACUP). Each of the three mandated components of the ACUP—the institutional official (IO), attending veterinarian (Vet), and institutional animal care and use committee (IACUC)—is represented as a leg of the stool. (A) When all three components are properly constituted and functioning, the stool's legs are of equal length and strength, and the top of the stool, which represents the ACUP, is held high and level. Such a balanced program is maximally able to withstand unforeseen problems. (B) When one of the components malfunctions, either by usurping the responsibilities and authorities of one of the others or by failing to fulfill its own, the corresponding leg of the stool becomes either too long or too short and the program is imbalanced. This figure illustrates the impact of an overzealous IO. Such a program is weakened and at much greater risk of failing under pressure.

Recognizing Risks in an Animal Care and Use Program

Just like the US government, an ACUP is at risk of failure when any of its components usurps another's authority or responsibility or fails to fulfill its own. Viewed in this way, each of the ACUP's three major components—the institutional official, attending veterinarian, and IACUC—can be either too weak or too zealous, with the result that the program becomes structurally imbalanced and inherently at risk. Recognizing the telltale signs of a weak or overzealous IO, AV, or IACUC and taking steps to correct them can prevent programmatic failures. The following sections provide common examples of weak or overzealous behavior by the IO, AV, and IACUC.

Institutional Official

The Weak IO

The two most common reasons for weakness in the position of the IO are inadequate empowerment of the person chosen for this critical role or the individual's failure to grasp the authority and responsibilities of the position. Naturally, it is particularly worrisome when both of these weaknesses are present in the same IO.

With respect to the IO's empowerment, Silverman and colleagues (2007, 34) provide a good description of the position's purview:

The IO must be authorized to legally commit on behalf of the institution that the requirements of the AWAR and PHS Policy will be met. To be effective, the IO must have sufficient administrative authority to promulgate, implement, and enforce policies across departmental lines. In addition, the IO must have sufficient fiscal authority to approve and fund a level of staffing that is adequate to meet the needs of the program, as well as any needed program improvements, facility repairs, and renovations.

An IO whose position does not include these institutionwide authorities is too weak to execute the responsibilities mandated by animal welfare laws, regulations, policies, and guidelines. Such an IO cannot respond effectively to emergencies or programmatic needs without seeking approval from a higher authority, and in times of crisis will not be able to act decisively and expeditiously for the ACUP and the institution.

Of equal importance is knowledge of the authority vested in the position of IO and of the responsibilities that come with it. In most institutions the appointment of an individual with sufficient overall authority to be the IO almost certainly guarantees that this person has many other challenging administrative responsibilities at the institution. Getting such a person to take the time to become familiar with the animal welfare laws, regulations, policies, and guidelines that relate to the position of IO can be difficult. However, without such knowledge the IO will have to rely on others—usually the IACUC chair or the AV—for advice on when and how to act. Such reliance significantly weakens one "leg" of the stool and increases the influence of another, throwing the ACUP out of balance. In a well-run ACUP, the IO may not be called on to do much more than receive favorable reports and sign documents prepared by others; but this superficial effectiveness may mask the individual's unsuitability for the position, and at the first sign of serious trouble an insufficiently empowered, inadequately informed IO can quickly jeopardize the entire ACUP.

The lack of authority and ignorance that characterize these two common IO weaknesses often give rise to other programmatic issues that are usually just symptoms of these root problems. Insufficient funding for the IACUC is a common problem. Failure to provide the IACUC with adequate staff, space, equipment, training, support to attend educational meetings, and relief time and/or a stipend for its chair are all common symptoms of this overarching problem at the level of the IO. Insufficient funding for the animal care unit is another symptom, as are persistent major deficiencies in the institution's physical plant, chronic staff shortages, high turnover rates for key members of the ACUP (such as the AV and the IACUC chair), and failure to provide an adequate occupational health and safety program for everyone who has contact with the animals, their tissues, or their fluids.

The Overzealous IO

The IO who acts overzealously, exceeding the position's authorities and responsibilities, can be just as great a failing for an ACUP as a weak IO, although such instances are rarer. They tend to occur in smaller institutions, particularly if the IO is also the leading researcher and/or has a degree in veterinary medicine. In such situations, the IO may usurp the role of the AV or the IACUC, throwing the program out of balance.

The signs of an overzealous IO include ACUPs in which the IO also serves as the IACUC chair or the AV. (I recall one institution in which the IO served as both the IACUC chair and the AV. One suspects that discussions in this ACUP were brief and problems unheard of!) A more subtle sign of overzealousness is when the IO makes it a point to attend the IACUC meetings. The IACUC reports to the IO, and in many institutions, especially larger ones, the IO has the authority to appoint the IACUC members. It's a fundamental principle of administrative policy that leaders do not serve on committees that report to them, let alone one that they appointed. Such behavior diminishes the autonomy of the IACUC and exaggerates the IO's impact on the program. It's hard to imagine how the regular presence of the IO at IACUC meetings could not have a chilling effect on the candor and breadth of the committee's discussions. Furthermore, the IO who attends IACUC meetings in order to ensure that the committee "behaves itself" is creating a second problem by attempting to resolve the first one incorrectly.

Other signs of an overzealous IO are somewhat less obvious (and perhaps less compelling), and are probably more applicable in large academic institutions than in other settings. They include ACUPs in which the IO sets the per diem rates for animal housing and/or assigns all animal housing space. The cost and availability of space for housing animals are naturally of prime importance to researchers, for whom costs are always too high and space too scarce. Although the IO should have the final authority over these two factors that directly affect both research productivity and the quality of the ACUP, that authority should be limited to situations in which all other efforts have failed to resolve an issue.

Per diem rates are also critical to the AV, as most animal care programs are run at least partially on a cost-recovery basis. Similarly, the assignment of housing space, and its relationship to other elements of the animal care program (e.g., proximity to other species, assignment of animal care staff, vector control, environmental control), is critical to the AV. By setting per diem rates the IO takes fiscal control and responsibility for the animal care unit out of the hands of the AV, and by assigning housing space usurps the AV's role in managing the animal care program. (The foregoing does not, however, mean I believe the AV should unilaterally set per diem rates or assign animal housing space; see the AV discussion below.) The IO who sets per diem rates and assigns housing space is also vulnerable to confrontations with disgruntled faculty members, instead of remaining impartial and available for appeals.

The last sign of an overzealous IO is somewhat more subtle. Under current animal welfare laws, regulations, policies, and guidelines, only the IO has the authority to override the IACUC's approval of a proposed animal activity. Whereas it takes a majority vote at a meeting of a quorum of the IACUC members to suspend a previously approved activity, the IO can stop such a project instantly. Not even the AV has such far-reaching authority, being able to halt only those animal activities that pose an immediate significant risk to human or animal health or welfare that was not anticipated by the investigator and approved by the IACUC. The IO's mandated authority to quickly suspend an animal activity can be critical to ensuring human or animal health or welfare and can protect the integrity of the ACUP and the entire institution in a time of crisis. However, as is always the case when such sweeping powers are granted, there is the potential for abuse. An IO who second-guesses the IACUC's reviews of proposed animal activities and frequently overrides its approvals is usually acting overzealously and always inappropriately. In the rare cases where an IACUC is truly "broken," such actions are not the way to fix it. An IO's overzealousness in blocking IACUC-approved animal activities circumvents the mandated review/approval process, weakens the IACUC "leg" of the stool, and jeopardizes the integrity of the ACUP.

Attending Veterinarian

The Weak AV

Veterinarians have traditionally been the backbone of ACUPs. The AV's role in ensuring that an institution provides a high-quality animal care program is critical, so weaknesses on the part of the AV usually have adverse consequences for the entire ACUP. Signs of such weakness most often stem from a failure to participate fully in the program, with the result that the AV's "leg" of the stool is weakened and the program becomes vulnerable to animal care problems. In some programs, particularly those that are small and use a part-time veterinary consultant, the AV is rarely on site, with the result that there is a void in leadership and expertise in the animal care section of the ACUP. Even veterinarians who are full-time employees of an institution may be away traveling or consulting for extensive periods of time, or may choose not to attend IACUC meetings on a regular basis, not to participate in semiannual inspections of animal facilities and animal study areas, and not to advise researchers who are planning animal activities. Inadequate review of proposed animal activities that are on the agenda for discussion at an IACUC meeting is another sign of AV weakness. Each decision not to participate fully in the ACUP is a sign of such weakness, and each exposes the institution to unnecessary risk.

Veterinary weaknesses may also become evident in the animal care program itself. For example, programs with a large number of investigator-maintained animal colonies, where husbandry and routine animal care are provided by members of the research team with little or no direct oversight by the AV and animal care staff, signal a weakness on the part of the AV. Maintaining high-quality, uniform standards of animal husbandry and health in such programs is extremely difficult. Programs in which many animals are housed for prolonged periods in investigators' laboratories exacerbate this problem. Except in those infrequent situations where such practices are justified by scientific need, the IACUC has to shoulder a good deal of the blame for approving the widespread use of such practices. But it's a weakness on the part of the AV to accept widespread laboratory housing of animals without working to help the institution develop equivalent options in an animal facility and under the AV's umbrella.

Other obvious signs of veterinary weakness include persistent physical plant problems; inadequate housekeeping, storage, and sanitation practices; and deficient record keeping. High mortality and morbidity in animal colonies are also obvious warning signs, the root cause for which may be yet another AV weakness—inadequate training of investigators and laboratory personnel by veterinary staff. Failure to provide adequate training can also reveal itself in the form of anesthesia or analgesia problems during procedures, or problems with poor surgical technique, including failure to maintain sterility during recovery procedures.

The Overzealous AV

As with the overzealous IO, an AV also can overstep the position's authorities and responsibilities. Obvious signs of such behavior include ACUPs where the AV also serves as the IACUC chair or the IO. A slightly more subtle manifestation of the problem is an AV who runs the IACUC meetings—setting the agenda, leading the discussions of proposed animal activities, calling for votes, taking the minutes, and so forth.

Another subtle weakness is the staffing and funding of the IACUC from the AV's budget. This practice is actually quite common and some of the blame for it falls to the IO, who, for want of a better place to allocate IACUC funds, places them in the AV's budget. The flaw here is revealed when the IACUC is asked to exercise its monitoring and oversight responsibilities on the veterinary care aspects of the ACUP. IACUC staff who owe their hiring, paychecks, promotions, travel budget, and new computers to the AV are in a compromised position when they are called on to assist in reviewing their supervisor's job performance. Similarly, IACUCs whose offices are located in the animal care unit and whose web pages are on the AV's website have surrendered their independent identities to the AV, unnecessarily weakening the IACUC "leg" of the stool. Small wonder that investigators in such programs usually bypass the IACUC and ask the AV about approval of animal activities, the addition of personnel, the status of pending research grants, and other issues that should be addressed to the IACUC.

Whereas a weak AV fails to provide investigators with sufficient help in drafting their proposed animal activities for IACUC review, the overzealous AV actually writes protocols for the investigators and then adamantly defends them during the IACUC's deliberations, which s/he may be leading. Similarly, the overzealous AV not only attends the semiannual inspections of animal facilities and study areas but also actually conducts them, determining when and where the team goes and what they record as their observations, thus usurping the authority and responsibility of the IACUC members. The overzealous AV writes the IACUC's semiannual reports to the IO as well as the entire AAALAC program description, including the sections that describe the functions of, and relationship between, the IO and IACUC.

While the AV is rightly responsible for managing the financial aspects of the animal care program, it is inappropriate for an overzealous AV to set per diem rates in a vacuum, especially in a large academic institution. Doing so inevitably drives a wedge between the AV and the investigators, who see themselves as subject to "taxation without representation," and thus unnecessarily stresses the ACUP. A better strategy is for the AV to appoint (or ask the IO to appoint) an advisory committee that is populated by investigators who represent the various "geopolitical" units in the institution and is charged with advising the AV and IO on setting appropriate per diem rates. Given the same fiscal information the AV would take into consideration in setting these rates, the advisory committee inevitably will come to the same recommendations; the difference is that the responsibility for these decisions now will be seen to rest with the investigators, not the AV. Furthermore, in those cases where the advisory committee does not support a rate increase that is clearly needed to fund essential animal care operations, such a committee can be a powerful ally for the AV in approaching the IO to provide additional core support for animal care.

Finally, although numerous important animal care considerations must be taken into account in assigning animal housing space to investigators, an overzealous AV fails to appoint an advisory committee of key investigators to offer advice on animal housing space assignment, especially when too little of it is available to meet research needs. Here again, the overzealous AV presumes that s/he alone should determine whose research should go forward and whose should be put on hold. In academic institutions, the rancor associated with making tough decisions of this kind can only be mitigated by taking the AV out of the bull's eye and making it clear that decisions are based on the advice of and supported by one's peers. Failure to employ such a committee inevitably leads to end runs by disappointed investigators to the IO, stressing and weakening the ACUP.

IACUC

The Weak IACUC

Problems traceable to a weak IACUC have for some time been the single most common source of negative findings during AAALAC accreditation site visits. It follows then that the IACUC is the first place to which institutions should turn when performing an internal risk assessment. While veterinarians must undergo a rigorous training program in laboratory animal medicine and even secure board certification before they can successfully compete for the AV's job, and while it can be argued that an equally challenging background is required for the administrators who rise to the level of IO, no such training or experience typically is required for the people appointed to serve on the IACUC. Even the chair of the IACUC may have had no special preparation for the job beyond having served as a committee member. Couple this with the fact that under current animal welfare laws, regulations, policies, and guidelines it falls to the IACUC to be the "conscience" of the ACUP—responsible for assuring the rest of the institution, general public, funding agencies, and regulating and accrediting bodies that the ACUP is in full compliance with the complex standards governing the use of live vertebrate animals in research and instruction—and one can begin to see why the IACUC is typically the weakest "leg" of the stool.

The most common cause of IACUC weakness is lack of training, experience, and/or commitment among its members and staff. Signs of this problem include annual rotation of the IACUC chair, failure to provide the chair with adequate support or relief from other institutional duties, appointment of IACUC members to serve for only 1 year, the appointment of few members who are active investigators, and the lack of a requirement that the IACUC chair, staff, or members attend national or even local meetings where specialized training is provided. Keeping the IACUC abreast of the complex, ever-changing regulatory environment typically requires a sufficient number of skilled administrators to coordinate and support the committee's activities. Failure to provide such staffing dangerously weakens the IACUC and indicates a lack of institutional commitment on the part of the IO.

Insufficient institutional commitment is probably the second most common source of weakness on the part of the IACUC. Weak IACUCs typically suffer from inadequate staffing, too little (or sometimes no) office space, an operating budget far less than required to properly fulfill their oversight and monitoring mandates, obsolete computers and software, and no access to programming or hardware expertise. The result is an IACUC that is administratively overburdened and disheartened and that takes months rather than days or weeks to review investigators' proposed animal activities.

A weak IACUC is unable to monitor ongoing animal activities, areas where animals are being studied, or the research personnel it has approved. It fails to compare animal use protocols to research grant proposals, performs little or no prereview of submitted protocols before the committee meeting, is inept at reviewing proposed animal activities for compliance with applicable standards, and loses (or keeps deficient) records (e.g., meeting minutes, protocols, reports). Desperate to keep from drowning in this sea of incompetence, the chairs of such IACUCs often turn to their staff or the AV to run their committee meetings. To save time, they virtually abandon the full committee method of reviewing proposed animal activities, turning instead to the designated member method or adopting an impermissible "expedited" review by the chair, AV, or IACUC staff. The committee's semiannual inspections of animal facilities are cursory and always delegated to the same two committee members, and the committee rarely if ever visits animal study areas located outside the facilities. Semiannual review of the remainder of the ACUP is either perfunctory or entirely lacking.

The weak IACUC, like the three monkeys who "see no evil, hear no evil, and speak no evil," leaves the institution blind, deaf, and mute with respect to its ACUP, turning the 3-legged stool into a 2-legged stool that is certain to topple the first time it's nudged.

The Overzealous IACUC

As the chair of an IACUC, I am tempted to say there is no such thing as an overzealous IACUC! But IACUCs do sometimes get carried away in attempting to exercise their authorities and meet their responsibilities. An overzealous IACUC ignores the AV during protocol review, overriding the veterinarian's recommendations or suggestions about drug doses or choices, the use of analgesics, the proposed justification for a physical method of euthanasia without anesthesia or sedation, the need for asepsis during a proposed procedure, the proposed justification for nonrecommended forms of husbandry, the housing of animals outside of an animal facility, and so forth. Overzealous IACUCs are sometimes characterized by a dictatorial chair, who seems to have a permanent lock on the position and exercises total control over both the committee's actions and the staff.

Such IACUCs also develop and promulgate unneeded policies that hamper investigators, the AV, and even the IO. They are inflexible in their deadlines and rule making. They conduct obsessive semiannual inspections and program reviews, documenting every piece of bedding found on the floor of animal housing rooms, every scratch or chip in the paint, and minor deviations from the Guide's recommendations for heating, ventilation, and air conditioning. Such IACUCs adamantly cling to engineering standards, refusing to adopt outcome-based performance standards when reviewing proposed animal activities, and they never (ever!) consider granting an investigator approval to perform an animal activity that requires an exception to the standards outlined in the Guide.

Summary

The structure of ACUPs in some ways resembles that of the US government, with three distinct branches each with its own mandate, and with a system of checks and balances designed to ensure that each fulfills its responsibilities without usurping those of the other two. As in government, there are significant areas of overlap in the responsibilities of the three components of the ACUP. The approach in this article has been to give examples of weakness or overzealousness in each component. A potential weakness of this approach is that it underemphasizes the importance of close cooperation and coordination among the three components. ACUPs in which the IO, AV, and IACUC work in harmony to fulfill their individual responsibilities are models for risk resistance.

Whereas the potential risks discussed here do not in themselves constitute failure to comply with animal welfare laws, regulations, policies, or guidelines, to an experienced observer they mark flaws in an ACUP that can cause the program, when subjected to an unexpected stress or tension, to unravel. A program without such flaws is much more likely to have the strength and flexibility to absorb an unexpected insult without compromising human or animal welfare or research findings.

Consider, for example, the potential problems caused by the sudden departure of a long-standing, highly experienced AV from a program in which the AV ran the IACUC meetings. If this AV also conducted the semiannual inspections of animal facilities and wrote both the AAALAC program description and the investigators' animal use protocols, then the potential for a major breakdown is very real. Alternatively, consider what could happen if a headstrong new investigator decided to challenge the animal care technicians about using his own nonstandard husbandry practices in an ACUP with an AV who is rarely on site, is unfamiliar with the protocols, and never attends the IACUC meetings or the semiannual facility inspections. Potentially significant vulnerabilities can develop in a program as a result of the AV's over- or underachievement, and it is easy to imagine similar scenarios for the IO or the IACUC.

Risk assessment is inherently more difficult than compliance monitoring, and its benefits are almost always less immediate, tangible, and rewarding than those of correcting instances of overt noncompliance. As a result, some may question the value of rigorous risk assessment/mitigation, especially in a program with an exemplary track record. People associated with such programs may say to themselves, "Well, I recognize some of these so-called risks in our ACUP, but our USDA inspections always come out well, we're AAALAC accredited, and in our experience we've just never had a serious problem." In response to this line of reasoning, it is worth noting that there is a risk associated with relying solely on a program's track record to evaluate its integrity. The flaw in this approach was perhaps most cleverly expressed late in his distinguished career by Vernon Sanders Law, a very experienced baseball player who pitched for the Pittsburgh Pirates from 1950 to 1967:

"Experience is a hard teacher because she gives the test first, the lesson afterwards."

Acknowledgments

I thank the many colleagues, most of them laboratory animal medicine veterinarians, who over the years helped me gain some insight into the complex relationships at the core of every well-functioning animal care and use program. Most important among these is Roy V. Henrickson, DVM, DACLAM, with whom I worked closely for over a decade of sometimes quite challenging times at UC Berkeley. I am also grateful to the four reviewers for their much-appreciated efforts to improve my manuscript. Any errors of fact or judgment that remain are solely my own.

Abbreviations used in this article: ACUP, animal care and use program; AV, attending veterinarian; IACUC, institutional animal care and use committee; IO, institutional official

References

Bayne KA, Garnett NL. 2008. Mitigating risk, facilitating research. ILAR J 49:369-371.

NIH [National Institutes of Health]. 2002. Public Health Service Policy on Humane Care and Use of Laboratory Animals. Bethesda: NIH Office of Laboratory Animal Welfare (OLAW).

NRC [National Research Council]. 1996. Guide for the Care and Use of Laboratory Animals. Washington: National Academy Press.

Silverman J, Suckow MA, Sreekant M. 2007. The IACUC Handbook, 2^nd ed. Boca Raton: CRC Press.