A good record-keeping system is important for the efficient operation of an animal research facility (ARF). Records that must be kept by an ARF are of three general types, namely, animal records, financial management records, and compliance records. Animal records contain such information as the source of the animal; the animal’s species, strain, gender, and any other pertinent characteristics; the date of receipt of the animal; and the date and nature of the animal’s final disposition. Animal records must also identify protocols on which the animal is used and diagnostic and medical procedures used on the animal. To reduce the labor requirement and cost of animal record-keeping, a single record may cover homogeneous groups of animals. For example, a group of animals from the same source, of the same strain, received on the same date, housed in the same room, subject to the same diagnostic and medical procedures, and used on the same protocol can be covered by a single record with a notation of the number of animals involved. Basic to animal records is accurate animal identification. Animal facility management and investigators should evaluate and agree on appropriate animal identification methods and see that they are implemented consistently and conscientiously. Inaccurately identified animals can lead to inaccurate data, which can lead to the costly need to repeat experiments. Financial management records are necessary for cost analysis and the recovery of ARF costs through fees for services. These records include census records on the number of animals per day assigned to an investigator or protocol. They must also include the billing and payment records of investigators or protocols. Cost analysis records include personnel activity reports or other data for allocating salaries and wages to animal categories, space use records by animal category, cage-washing schedules and the number of cages washed by animal category, the quantities and costs of supplies used by animal category, and the cost of animals procured. Additional records might be necessary for accurate cost analysis, and the reader is referred to the Cost Analysis and Rate Setting Manual for Animal Research Facilities (CARS Manual) (NIH 2000 or http://www.ncrr.nih.gov/) for such information. Data collected for cost analysis should be examined to see whether they reveal opportunities for cost containment. For example, personnel activity reports could reveal inefficient assignment of personnel, and revision of assignments could lead to cost savings. Compliance records are those required for compliance with the Animal Welfare Act, the Public Health Service Policy on the Humane Care and Use of Laboratory Animals, and any other applicable laws and regulations. Included in these records are those of protocol reviews and approvals, numbers of animals and species approved and used for a protocol, and reviews of animal care and use programs and facilities. Also pertinent are occupational health, faculty and staff training, and facility security records. Records are essential but can be a substantial cost item for an animal research facility. The institution must give thought to the type and format of records and the intended uses of the data collected. Data should not be collected and recorded unless the institution foresees a need for the information. Similarly, records should not be retained beyond their useful life. Note that some compliance records must be retained for 3 years after termination of the research project. There is a large amount of interrelationship among the records kept by an animal research facility. For example, the number of animals procured and assigned to a protocol needs to be entered into animal records, financial management records, and compliance records. Because of this interrelatedness, the institution should set up a system of interrelated databases to minimize data entry. Cost accounting is very important for the efficient and cost-effective operation of an ARF. The facility should have a system of cost accounting like that described in the CARS Manual. This manual sets forth a method whereby an ARF can allocate its costs to specific animal categories and service activities. The total costs associated with an animal category or service divided by the number of animal days or service units yields per diem or service unit costs. The manual contains a discussion of how such unit costs can be used to determine fees charged to users. Fees determined by these methods can be explained to any interested investigator. Investigator understanding of the costs involved in the care of their research animals generally leads to a greater acceptance of those fees. Fees based on cost accounting are more readily justified to sponsors of research. The cost analysis and related statistical data also will assist an institution in comparing the costs and benefits of various services and activities and have the potential for identifying how cost savings might be achieved. Cost records can also be used to develop cost consciousness in the entire staff. A sense of pride in being part of an efficient facility is a useful element in controlling costs. The major cost components of animal care are listed in Table 1. Almost all institutions have a system of charges for services to support their animal research facilities. As noted in the CIC study, nearly all institutions provide supplemental support from institutional funds. Per diem charges for animal care generally include housing, husbandry, cage sanitation, and maintenance of census records; in most institutions, routine medical care is also included. Routine veterinary medical care includes rodent health surveillance (sentinel animals, bedding transfer to sentinels, serology, and necropsy), disease diagnosis, physical examination of nonrodent mammals on arrival, response to medical emergencies, clinical and anatomic pathology support of diagnosis, and pharmacy stocking and maintenance. At the University of Michigan, each of these activities is attributed to an animal species in proportion to use of the activity for the purpose of setting the veterinary service fee (VSF) portion of the per diem. Every investigator at the university pays the daily VSF for his or her animals no matter who provides the daily care. For example, in the 1999 ARS, 74% of the institutions included support for routine rodent medical care in their per diems, and the remainder had a special fee (Table 14a, Appendix C). However, 63% of the institutions had a special fee for therapy of protocol-related disease. Institutions frequently provide a range of technical services on a fee-for-service basis: 42% had special fees for rodent euthanasia, 49% for rodent identification, 55% for rodent special diets, 56% for rodent breeding, 76% for rodent restraint, 89% for specimen collection, 88% for compound administration, and 76% for rodent rederivation (Table 12a, b, & c, Appendix C). A mixture of per diem and direct service charges makes good sense in that the user pays for special services. Animals of the appropriate species, genetic makeup, and quality must be procured for research purposes. Purchase of animals with uncertain health and unknown genetic background constitutes false economy in that their use can lead to inaccurate and invalid data or the necessity to repeat experiments. The decision to breed animals inhouse or to obtain them from commercial sources can be made after a careful analysis of all relevant factors. These include the purchase and shipping costs for commercial animals, the cost of inhouse breeding (including space costs), and the reliability of animal supply and quality. For efficient animal research, an institution can provide central core laboratories for a number of services rather than having individual laboratories duplicate services. These can be “free-standing” core laboratories or be provided by a laboratory otherwise heavily engaged in that activity. An example of one such service is cryopreservation of embryos. It is expensive to maintain breeding colonies of mutant mice or mice whose genome has been genetically manipulated unless there is an immediate need for them. It is often desirable to maintain unique genetic material or protect it against loss; at present, this can be done most economically by cryopreservation of embryos, but methods for the cryopreservation of rodent semen are also under development and might be applicable to some models. In the 1999 ARS, many institutions reported making cryopreservation of embryos or sperm available (Table 16e, Appendix C). In particular, 78% of group 3 institutions (730,000 mouse average daily census) reported making cryopreservation available through the animal resource program or other internal source. In this group, 43% of the institutions asked the investigators to bear the expense. It also might be desirable to establish specialized core laboratories for other activities, including monoclonal antibody production, production of transgenic or gene-knockout animals, characterization (by organ system or clinical specialty) of the phenotype of induced mutations in mammals, behavioral testing, histopathologic analysis, and experimental surgery (Tables 16a-h, Appendix C). Experimental surgery and, in larger programs, histopathology services are generally provided by the ARF, whereas other core services are generally provided by other internal sources or an external vendor. The physical plant of an animal facility must be designed to maintain the proper environment for the animals and to facilitate the investigative use of the animals. A well-designed physical plant with low maintenance costs, providing for efficient animal care and effective use of the animals by investigators, is an important element in controlling costs. Admittedly, there can be tradeoffs among low maintenance, efficient animal care, investigators’ convenience, and the initial cost of construction; these factors will vary institution by institution, and careful analysis should be given in each situation. There is a clear economy of scale in animal research facilities. The CIC study findings (Appendix B) indicated that labor productivity was the prime driver of animal care costs. Labor productivity was better in larger facilities. For example, caretaker productivity doubled when the labor-weighted volume (adjusting for the labor component of care across different species) increased fivefold. When an institution had more centralized facilities, labor productivity increased. For example, institutions with one or two facilities had a labor-productivity index about 1.5 times greater than institutions with 14 or more sites. Analysis of 1999 ARS confirmed and extended findings of the CIC study. There were 44 respondents who provided sufficient information to compute total operating cost of the facility and who listed the number of sites in their facility by size category (<5,000, 5,000–10,000, 10,000–20,000, and >20,000 ft2, Table 4, Appendix C). Total facility costs were regressed on amount of space (in square feet) in each category. Costs in dollars per square foot dropped from $93/ft2 in the second category (5,000–10,000) to $36/ft2 in the third and $28/ft2 in the fourth (Figure 2). The differences between those values were statistically significant at p < 0.0001; the coefficient for the smallest category was not statistically significant. Labor productivity also increased as caretaker hours per room increased. For example, labor productivity doubled when annual caretaker hours per room increased from 100 to 400 (CIC study, Appendix B). Those findings support the recommendation that animal care operations be concentrated, whenever possible, into fewer larger sites. Concentration of animal facilities must be weighed against investigator convenience in having animals readily available. Security is a major concern for animal research facilities and can constitute a substantial cost item. The 1999 ARS indicated that institutions had 46% of their sites protected by locks and keys, 17% by electronics, and 37% by a combination of electronics and locks and keys. Institutions should give careful attention to the risk of intrusion and the costs and benefits of various security systems. In addition to the economic costs, institutions should recognize that the public relations and psychologic costs of unwanted intrusions into an animal facility or research laboratory can be substantial. It must be recognized that the physical plant of an animal facility is a “hard use” area. The sanitizing materials, high traffic, heavy rolling equipment, the active nature of animal care and use, and some animal species themselves all exact a toll on the physical facility. That toll and the requirement to maintain reliable heating, ventilation, and air conditioning, electric systems, and sanitation and sterilization equipment dictate the need for constant maintenance. It is frustrating to the animal care staff, inefficient for operations, and a detriment to quality research when aspects of the physical plant underperform or require frequent maintenance. A well-maintained physical plant in which all systems operate reliably contributes to cost-efficient animal care. Nearly all institutions use 100% outside air with no recirculation. Because the air is conditioned (heated and humidified or cooled), not recirculating the conditioned air is expensive. According to the Guide, some recirculation is possible if the recirculated air is appropriately treated to remove microbial and chemical contaminants. Another method of energy recovery is to use heat exchangers to partially heat or cool the incoming outside air. The current methods of mouse husbandry were given considerable attention in the 1999 ARS in acknowledgment of the emerging prominence of mouse models in contemporary biomedical research. Nearly all institutions (98%) were housing some mice in microbarrier cages. Only a single small institution had not implemented microbarrier housing. Most institutions (67%) were using some individually ventilated cages. More large institutions (79%) were using these cages, whereas only 52% of smaller institutions were using this newer labor-saving cage system. Table 11a–11d, Appendix C, contains information on the 53 institutions with a mouse average daily census of more than 1,000. Automatic watering systems for mice have been controversial both because some mice develop dehydration if unable or untrained to manipulate the valves properly and because cages can be flooded if an automatic valve leaks or is continuously manipulated by the mice. Only 41% of the surveyed institutions had any automatically watered cages (Table 11a, Appendix C). Fewer group 1 (32%) and group 3 (29%) institutions used automatic watering systems than group 2 (67%) institutions. The 1999 ARS did not explore the role of such factors as cost, customer satisfaction, criteria for selecting a particular system, ease of sanitation, efficiency of operation, and intensity of oversight necessary to ensure proper function in the decision to deploy these systems. Some institutions house mice in microbarrier cages but do not use HEPA-filtered change hoods for transferring mice to clean cages. The percentage of mice changed in HEPA-filtered change hoods averaged 55% in small institutions, 76% in medium institutions, and 61% in large institutions (Table 11a, Appendix C). Microbarrier cages are changed more frequently than open-top cages because of ammonia accumulation. The average interval between changes in microbarrier cages was 5.4 days in small institutions, 4.6 days in medium institutions, and 5.9 days in large institutions, with a range of 3–7 days (Table 11a, Appendix C). The survey showed that cage-changing was less frequent in individually ventilated cages; however, the mean interval between cage changes was much smaller in practice than commonly advertised for these systems. The mean interval between cage changes in individually ventilated cages averaged 8.2 days in small institutions and 8.9 days in medium and large institutions; the range for all institutions was 3.5–14 days. A summary of the CIC study findings (Appendix B) indicated that the cost of animal care is lower in rooms that house larger numbers of animals. In the 1999 ARS, institutions were asked about the maximal number of adult mice that were permitted in their standard shoebox cages presumed to provide about 70–75 in.2 of floor space. Most institutions (66%) permitted five mice per cage; 29% permitted only four mice per cage (Table 11b, Appendix C). The average number of mouse-cage racks in a room was 4.1; the range was 2–8 (Table 11b, Appendix C). In these institutions, respondents were asked about the minimal aisle width that they recommended between racks. The average of the responses was 3.1 ft; the range was 0.5–8 feet. In the experience of the members of this committee, few animal care technicians or research technicians are comfortable in performing animal room duties in aisle widths below the mean reported in the ARS; this is also reported in the case study in this report (Chapter 4). The consensus of the committee was that room design, ergonomic considerations, heating, ventilation, and air conditioning capacity to maintain appropriate ambient air conditions should be evaluated by each institution to preserve a high-quality work and research animal environment before pursuing higher room capacities as a strategy for cost containment. To reduce expenses, facility managers are exploring different methods of sanitizing mouse cages. The 1999 ARS (Table 11c, Appendix C) indicated that 81% of the institutions were autoclaving their microbarrier cages, 49% were autoclaving their individually ventilated cages, and only 8% were autoclaving their open-top conventional cages; 19% only autoclaved cages used for immunodeficient mice; and a few (8%) used hot water without detergent to clean cages before autoclaving them. The type of cage washing and autoclaving used by an institution will depend on the microbiologic status of the mice housed in the facility. Various methods of bedding disposal were used (Table 11d, Appendix C). Most institutions (75%) disposed of soiled bedding in a landfill, 26% disposed of soiled bedding by incineration, and 21% disposed of some soiled bedding in the sanitary sewer. Nearly all institutions disposed of animal carcasses by incineration; only 8% reported landfill disposal. The scope of animal-husbandry activities required to support biomedical research is extremely diverse because of the wide variety of animal species used and the requirements of the varied research being performed. Those factors make it difficult to identify cost-saving measures that will apply universally. Some general observations regarding cost considerations and potential savings are presented here with respect to common areas of animal husbandry, such as cage sanitizing, watering, environmental enrichment, purchasing supplies, and acquiring animals. Transferring animals to clean cages and sanitizing primary enclosures constitute the bulk of physical labor required to support research facilities that have large rodent populations. It is important to schedule these activities carefully so that staff changing cages have clean cages and equipment (water bottles, cage tops, card holders, and so on) as they are needed and staff washing cages can plan activities in the wash room. Several innovations show promise for minimizing costs associated with these husbandry requirements. Individually ventilated cage (IVC) systems provide cost savings by decreasing the frequency of cage changing (Perkins and Lipman 1996; Reeb and others 1998) and by increasing the number of cages of animals housed per square foot of facility floor space. These systems are increasingly popular and are now widely used. Of the 63 institutions participating in the 1999 ARS, 30 reported experience with the use of IVCs for laboratory mice. Of those 30, 21 reported that IVCs permitted an extension of the cage-changing interval (Table 11a, Appendix 3). In most cases, this was from twice a week to once a week, but nine institutions were able to achieve an interval of 10–14 days. Thus, IVCs appear to have the ability to reduce labor costs by increasing the total number of cages that a technician can service over a given interval by a factor of 2–3. In a case study provided by Emory University, where changing frequency for a cage of nonbreeding mice went from four times in 2 weeks to once in 2 weeks, the number of cage units serviced per worker per week increased from 780 to 2,000 (personal communication, M.J. Huerkamp). This ratio excludes workers dedicated to cage washing and excludes supervisory personnel. However, an appreciable cost savings in labor, material, and cage replacement resulted that was reflected in lower per diem charges to investigators. IVCs appear to be suitable for many facility settings and warrant consideration as a method of cost containment in programs that deal with large populations of laboratory mice. The type of contact bedding used in static isolator cages can affect the microenvironment; some bedding types show a significant difference in how long it takes ammonia to reach unacceptable levels (Perkins and Lipman 1995). Use of IVCs and certain types of bedding might allow animal husbandry programs to decrease the cage-changing frequency and still have an acceptable microenvironment. The reduction in labor required to process cages can be significantly reduced and have a major cost saving impact on facilities housing large numbers of rodents. IVCs can also house many more rodents per square foot of facility space than the traditional method of using shelf racks and standard shoe-box cages. This can provide considerable savings in construction costs by reducing the area of the vivarium needed to house a particular number of rodents and the ensuing costs of operating the physical plant (Lipman 1999). The cost of sanitizing caging and accessories involves more than just wages of personnel who perform the labor. The cage-washing area has inherent liabilities associated with the presence of various chemicals, steam, and conditions that can lead to repetitive-motion injuries of personnel. Introduction of robotics to handle repetitive procedures in the cage-washing area is a recent advance in long-term cost-saving measures. Robotic arms have been designed to process polycarbonate rodent cages through an indexed tunnel washer, working on both dirty and clean sides of the cage-washing apparatus. Robotic technology offers the possibility of substantial long-term cost savings for biomedical research facilities because of its long service life, low maintenance requirements, and the elimination of disability claims in connection with cage washing-related injuries. Whether or not robotics are used for cage washing, automatic dispensers to refill cages with bedding are a useful labor-saving device. As with any investment in labor-saving equipment the institution should compare the labor savings with the cost of the equipment to determine whether the investment is justified. Newer, more durable polymer plastics are available for rodent cages, with a cost that increases as the strength and durability of the plastic at high temperature increase. High-temperature-resistant plastic rodent cages are superior to standard polycarbonate cages in maintaining transparency and resisting formation of microfissures under conditions of frequent autoclaving (Agee and Swearengen 1995). Facilities that require frequent autoclaving of rodent cages, such as biohazard facilities or rodent barriers, might find that the more durable high-temperature plastic cages, which cost more, would result in savings over time. Although automatic watering systems are a labor-saving device, most mice housed in a variety of cage types in biomedical research facilities are provided water via bottles. When water bottles are used, steps can be taken to maximize efficiency and minimize repetitive-motion injuries associated with manipulating large numbers of the traditional water bottles, sipper tubes, and stoppers. Ergonomically designed tools are available to remove sipper tubes from rubber stoppers and reinsert them later. Water bottles with screw caps or with weep holes (drilled bottles) eliminate the need for rubber stoppers and the effort needed to insert them into bottles, which is considerable. The use of bottle holders with retainer lids that hold several water bottles at once makes dumping and handling of water bottles easier and reduces operation time. Supplies purchased for use in animal care and use programs should comply with the provisions of the Guide for the Care and Use of Laboratory Animals. Although supplies account for only a relatively small portion of the budget of an animal facility (about 11% at one major research institution), some cost savings are possible. Here we describe some of the alternative strategies that can be used by animal facility managers to contain supply costs. Food and bedding are likely to account for a high proportion of the supply costs. Specialty foods can be expensive and should be clearly identified and used only for the purpose defined. Bulk ordering of food and bedding permits obtaining bids on these items and can substantially reduce per-unit cost. However, bulk ordering presents problems for some institutions. Storage of items in bulk requires space, which often must be specifically designed for the items being stored, such as food. Many institutions have multiple animal facilities; in these cases, distribution costs will need to be considered. Cleaning supplies might also be appropriate for bulk purchase but are subject to the same considerations as food and bedding. Newer facilities have 400-gallon tanks for cage-washing detergent to take advantage of this cost-saving opportunity. Where permitted by facility design and available space, existing facilities might consider retrofitting with equipment that provides greater storage capacity to achieve cost savings. In addition, rack washers are now available with holding tanks that use smaller quantities of chemicals and water. At present, the Animal Welfare Act regulations only mandate environmental enrichment for nonhuman primate species and thus afford institutions the opportunity to contain costs by limiting the application of enrichment strategies to these species. However, there is a growing body of literature on environmental-enrichment strategies for many of the common laboratory animal species, and the Guide (NRC 1996a) provides an impetus for institutions to evaluate and incorporate enrichment measures into their animal care and use program for all species where appropriate. Indeed, many biomedical research facilities now provide environmental enrichment to many species of research animals, including rodents. To minimize the resultant increase in the amount of personnel time and facility resources dedicated to these activities, the most labor-efficient devices should be incorporated. For example, if tunnels and other similar devices are used in rodent cages, they should be colorless, nonopaque materials that allow easy visualization of all the animals in the cage. This provision will eliminate the need for additional time and effort to manipulate the devices to permit all the animals in the cage to be seen during observation periods. The 1999 ARS did not provide any information on the magnitude of costs borne by institutions providing environmental enrichment. Eighteen of 52 institutions (see Table 26b, Appendix C) indicated that they subsidized program development costs, such as environmental enrichment, but further details were not given. A wide variety of enrichment devices and supplies are available as specialty items from commercial sources. However, very good inexpensive alternatives can often be made from other items on hand or from ordinary supplies and materials that are available locally over the counter. Taking that approach potentially carries the dual benefit of involving the animal care staff in a creative, innovative enterprise that contributes to animal well-being and reducing the supply costs associated with this effort. Health and other safety factors should be considered during the design and use of enrichment devices to ensure that neither animals nor personnel are exposed to additional risks. For animals that are commercially available, inhouse breeding for general animal use is usually more expensive than purchasing animals as they are needed for research studies. Inhouse breeding is required, however, for some studies, such as research on reproductive processes and production of knockout or transgenic animals. About 55% of the mice used in research are purchased from vendors (Table 2). Larger institutions purchase a smaller proportion of mice than smaller institutions presumably because of their more extensive use of transgenic, knockout, or other unique mouse strains in research studies that necessitate inhouse breeding. Grouping orders can be an effective way to reduce handling and transportation costs. This requires coordination between the principal investigator and the animal facility management to ensure that the animals are available as needed for the research program. Depending on the volume of products used and other institutional circumstances, it might be beneficial to purchase veterinary supplies in bulk or through an institutional pharmacy to achieve cost savings. Drugs and biologics should be stored centrally under appropriately controlled and secure conditions. The Guide (NRC 1996a) calls for an extensive occupational health and safety program that includes considerable administrative time to establish and maintain the program, to track employees, to train personnel to establish guidelines for the use of personal protective equipment, and to provide for periodic medical evaluation and practice of preventive medicine. Anecdotal evidence suggests that such a program is expensive, but there are no studies of the cost of such programs. This is a subject for further research to ascertain the total cost of such a program and its components so that methods can be devised for cost containment. Protective clothing and other personal protective equipment, such as gloves, face masks, bonnets, booties, and eye-protective devices, can also be purchased in quantity and provided to staff members as needed. Because the cost of these disposable items can be large, some programs are considering purchasing more durable laboratory coats, jump suits, or coveralls, the most expensive components. In some cases, these items can be repeatedly autoclaved and recycled for reuse to reduce the overall cost. However, the cost of personnel time to collect and autoclave these items needs to be taken into account. In summary, the major findings and opinions expressed in this chapter are as follows:
|