Using the NC3Rs EDA service
We recommend you sign up for and use the NC3Rs Experimental Design Assistant to design your studies: https://eda.nc3rs.org.uk/
It is free to use and has all the tutorials and best practice guidance you need to ensure your research and experimental design is rigorous, statistically sound and well designed.
Our Study solution then ensures you can implement your design correctly, plan it effectively, schedule it efficiently and then execute it (i.e. collect, record and analyse results data) accurately and efficiently, ensuring data transparency, reproducibility and eliminating bias.
The EDA file must be created in the https://eda.nc3rs.org.uk/ system and then saved locally and uploaded to the Study using the EDA attachment function in Studies, shown above.
This enables research oversight groups, such as your Protocol or Licence holder or IACUC to be able to view the EDA file and the study design and compare this to how the Study experiment has been constructed, and complies with the EDA design recommendations.
The Phase feature of Studies enables you to define different stages of an experiment one of which is often “acclimatisation” or what is sometimes referred to as a “pre-study” phase.
This phase may be used to:
- Record that certain animal models are being bred in preparation for study, or
- Record that your animals must achieve certain parameters before being accepted for study, such as:
- A minimum tumour size
- A certain body weight range
- Recovery from a particular surgery
If you wish to randomise your study, in simple terms, this means randomly allocated animals in your study to particular experimental groups.
First, calculate the total number of animals needed per study / experimental group
- Then sum the totals to give you your total animal count for the study
- Use the EDA tool to calculate this based on your planned experiment design
Second, you need to have all the animal records in the animal database (which you can create using the Import Animal function, or you can add manually via the Add animal button). You need to have assigned an individual identification code to each animal – which should be unique for each animal within your study. This can be visual or our RFID tag code, or both.
Next you have a choice: do you allocate your animals to cages before you randomise your animals or after?
If you have a pre-acclimatisation phase, you may wish to allocate your animals to cages first (e.g. animal 1-4 in cage 1, animals 5-8 in cage 2 etc.)
However, if you do not have a pre-study phase, you can run your randomisation process to determine which number animal belongs to which study group and then you can decide how you want to allocate your animals to which cages:
- You may wish to group animals in the same study group into the same cage(s), for example
- Cage 1 houses 7, 21, 23,36 and 60 or
- You may wish to house animals in an sequence so that animals 1-4 go into cage 1; 5-8 go into cage 2 etc.
Which approach is better?
- From a work management point of view, it makes sense to group animals into cages where the treatment / observation can be conducted in a streamlined efficient manner and, therefore, you should randomise your animals first, and then group them into cages so the test methods can be performed with groups of animals in the same cage
You should use the Workbench to allocate your animals to the required cages and racks.
Once the animals are in cages, you must use the Workbench to allocate your animals to the protocol to which your study is approved.
There are many methods – Excel has one; flipping a coin is a method etc.
- Please refer to: https://eda.nc3rs.org.uk/experimental-design-allocation for more information
First you must establish how many animals in total your study requires:
- The EDA tool has an inbuilt sample size calculator which will help you determine this across all groups
- The EDA tool will produce an Excel file that randomly allocates the animals across the number of study groups your experiment requires
Control and treatment groups
Study uses the general term “Experiments” in the study function to refer to any control or treatment group that is needed to separate groups of animals according to the research parameters that are to be applied and tracked.
You can define as many Experiment or experimental groups as required. For example, you may have control groups defined by gender, model, age or any combination required.
For treatment groups, you may wish to define groups according to the dose level or other intervention variable, such as tumour cell line injected.
If you are not blinding, you may name each Experiment with specific variable information so that the groups are easily distinguished by the group name given.
There are different approaches to blinding:
- Treatment application
Study supports blinding for each of these stages by enabling you to name your study groups and your treatments without any obvious indications of meaning.
If you wish to blind your technicians to the study group the animals belong to, you may wish to name the study groups without any meaningful labels, e.g.
- Study group 1
- Study group 2
If you wish to blind the technician to whether the treatment is a high level dose / low dose, you can name the test methods without meaningful labels, such as:
- Dose 1
- Dose 2
You will have to keep an offline record that maps the study group names and test method names to your actual experiment and treatment parameters. If you are analysing your study data by exporting the data to Excel, you can simply add columns into the spreadsheet with your study parameters.