By Rebecca Morelle
BBC News science reporter
Scientists are carrying out a study to see if it is possible to report levels of suffering experienced by animals during scientific procedures.
About 2.85 million animals were used in experiments in 2004
The Home Office only issues statistics based on how severe a procedure is expected to be when it is licensed.
The study aims to see if suffering can be assessed and reported after the procedure has taken place.
A report setting out the preliminary findings of the investigation is due to be published.
The work is a collaboration between the Laboratory Animal Science Association (Lasa) and the Animal Procedures Committee (APC).
Levels of severity
Both the House of Lords select committee on animals in scientific procedures and the Nuffield Council for Bioethics have called for greater information to be provided on animal suffering during research.
At present, Home Office annual statistics on animal research detail the number of project licences that have been granted. In addition, each licence is categorised as mild, moderate, substantial, or unclassified (meaning the animal is under anaesthetic). This is based on the likely experience of the "average" animal in the experiment.
But critics say this information is not specific enough, and because each licence is classified before the experiments have taken place, offers no detail about the actual suffering of the animals used.
A working group from Lasa and APC, involving nine research institutions, looked at the feasibility of collecting and reporting information about the levels of severity of animal procedures.
"We looked at whether this data was readily available, if it could be collected easily, put together and then reported," said Dr David Smith, president of Lasa and chair of the working group.
They surveyed the current practices used in animal experiments in the nine establishments. They found the majority of those asked already keep a record of the suffering and severity experienced by individual animals.
The working party then examined ways this data could be usefully organised.
"We looked at a range of different systems to report this information, and concluded that to cover the full range of animal experiments, we need to report both the severity of the procedure and how long the animal spent in the experimental process," explained Dr Dominic Wells, a member of the working group from Imperial College.
"It became very clear to us that in many cases the severity would not be consistent for the whole of the experiment but, rather, would vary over time."
The scientists proposed a "two-grid" system, looking at the intensity of the adverse effects together with their duration.
"For example, a mouse that has undergone a vasectomy would experience a brief period of moderate severity followed by a long period of very mild severity," said Dr Wells.
The report did concede that providing this extra information could increase the bureaucratic burden upon scientists involved in animal research, but suggested there may be ways of reorganising the current system to offset this.
But it also suggested this would increase openness in animal research.
Dr Vicky Robinson, chief executive of the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs), commented:
"The potential for increased administrative burden is a concern and we would like to see this minimised wherever possible. However, scientists have a scientific, ethical and social responsibility to be as open about their work as possible."
Alistair Currie, campaigns director for the British Union for the Abolition of Vivisection (Buav) said he welcomed bringing more information into the public domain, but warned: "This is only of value if such assessments are conducted accurately and fairly and both experts and the public are allowed to scrutinise and test it."
The scientists involved in the study plan to further refine the proposals and test them on a wider range of organisations. When the study is complete it will be submitted to the Home Office.