Share thisRegardless of how you record information or data derived from your animal colonies, you are using some form of data management process. It may be as basic as scribbling notes on a post-it pad or as elegant as scanning an animal and updating its records from a cellphone.
Lab animals are used in research to derive scientific information that would be difficult to obtain otherwise. The gathering, recording, and storage of this information is crucial to both the experimental results and the justification of the use of these animals.
Regardless of how you record information or data derived from your animal colonies, you are using some form of data management process. It may be as basic as scribbling notes on a post-it pad or as elegant as scanning an animal and updating its records from a cellphone. The focus of this article is to identify what type and extent of this data management is appropriate for your institution. Clearly, every animal facility is different in some way, whether it is the research that is being carried out or the management organization, so at best here we can discuss concerns and generalities that are common to all.
A common issue with all data management systems error rates and what degree of error the facility in question can tolerate. Because of the nature of animal research, most data collected from the facility is entered by human and not from automated or electronic systems. This human data entry is the source of most errors with the unfortunate result that the more data you collect, the more errors you will have. Similarly, the more types of data that you collect, the more opportunities for error are present. It would seem that the less data you collect, the more accurate your data management system is.
In reality, there is a balance between the amount and types of data collected and the span of exposure of that data to human error. This has to be considered when examining the first decision of system choice, that of a paper-based system or a computer-based system. Both systems have their advantages and disadvantages and we explore these to highlight the differences.
Paper or Electronic?
I will describe a very simple hypothetical animal research project where the vivarium purchases founder animals and breeds them to maintain a group of stock animals that are regularly distributed to a PI for certain tests. The data that is required from the project may be as follows:
• Mating and holding cage per-diem counts
• Animal weaned counts for investigator availability
• Animal use counts for IACUC
Managing this project’s data with a typical paper-based system may consist of two types of cage card, stock cards, and breeding cards. The founders are set up as mating pairs and pertinent information such as a line name is entered on a breeding card. This card would have space to record litters born to these parents that would be completed at the appropriate time. As litters are recorded, the birth dates are written down and used as a basis to know when to wean the animals. When a litter is weaned, the pups are transferred to separate sex holding cages and a stock card is completed to indicate the contents of the cage. At the same time, a paper ledger may also be employed to record more detailed information. If the pups will be genotyped prior to weaning, this ledger may contain reference to the cage card that recorded the litter and the sex and genotype of the individual pups. For identification, the pups may be ear-notched and this information is also entered into both the ledger and on the cage cards. As animals are weaned, this is also added to the ledger to indicate that animals are available for use in tests.
This seems quite simple and there is no need to resort to high technology, so what is the problem?
First, the entire time the data is valid and in use, it is exposed to human error. In this example, the vivarium staff is performing double entry of data thus doubling the error potential. We need to obtain the required data from the project, so first we need cage per-diems. To do this, typically, staff will periodically check all the breeding and stock cards to determine a count of cages in use for this project. These counts will then be used to calculate charges to be made to the investigator.
Next, we need to know when animals will be ready to hand over to the investigator for testing. A typical problem can be highlighted as follows:
The PI asks for a group of animals for testing, let’s say for example 150 gray females. The vivarium staff will pull out the stock ledger and count the number of gray females available finding only 130. Not a problem for the PI, he can wait a day or two, but how many will be weaned and available by then?
The problem is that the ledger in use does not track weaning dates. Vivarium staff must go into the animal room to find out this information, checking for weaning dates on each individual cage card to see if there will be sufficient animals weaned in two days’ time. With the paper system, data may not be readily accessible and retrieving needed information may incur cost and time.
Finally, some form of animal usage count is needed for IACUC reporting. Keeping a running total of animals transferred to the PI and deducting the counts from authorized numbers on the protocol may be sufficient. This is easy to manage, but requires continuous calculation by the vivarium staff.