An analysis suggests science really is trying to reduce animal experiments.
Most animal studies use rodents
A comparison of almost 3,000 research papers published over 30 years in major biomedical journals found a 30% fall in the number of studies using animals.
The analysis by Dr Hans-Erik Carlsson and colleagues also showed increasing use of alternative testing methods, such as experiments on cultured cells.
The team told the Veterinary Record there was now better reporting of the welfare of the animals kept in labs.
The Uppsala University researchers conducted their investigation because they wanted to get a clearer idea of the extent to which the international "scientific culture" was adopting the principles of the so-called "three R's" - the replacement, reduction and refinement of the use of animals in experiments.
Dr Carlsson's team says the changes over time are encouraging.
"The reduction in the number of animals described in each article, the more frequent reporting of the use of defined, high-quality animals and the improved reporting of animal welfare factors indicate that animal experimentation has become more refined," they write in the UK veterinary magazine.
The Uppsala information is published as the British government prepares to announce new measures on the welfare of animals used in laboratories.
A House of Lords committee in 2002 was highly critical of some of the attitudes and practices of the scientific community and urged the government to make real progress on the three R's.
One of the ideas being proposed is for a new centre to be established that would develop alternative experimental methods.
Dr Peter Kohl, a physiologist at the University of Oxford, told BBC News Online that one of the most promising alternatives to animal research was computer modelling, which mimicked systems inside the body.
"The definition of a model is a simplified representation of reality. But this highlights that a model always has to be incomplete. A model cannot accomplish all the aspects of the original," he said.
Dr Kohl has been developing models for his work on the effects of trauma to the heart.
UK ANIMAL EXPERIMENTS
2.73 million experiments in the 12 months of 2002
Total number of procedures rose by 4.2% on 2001
About 80% are for research and drug development
Safety testing accounts for most of the rest
Great apes such as chimpanzees cannot be used in experiments
"The theoretical tools, mathematical ones in particular, have allowed us to integrate a vast amount of biological data gathered in labs all over the world. And so much so that in recent years models of the heart have gained predictive power," he explained.
"With the help of mathematical models, we can perform fewer and better informed experiments."
Timothy Morris, of pharmaceutical company GlaxoSmithKline, added: "The three R's have changed with science and I think it is very important that scientists are allowed to engage with the three R's."
The issue of animal experimentation has become a very live one in Britain in the past year with the controversy over plans to build a primate research lab in Cambridge.
The city's university eventually pulled out of the project after a protracted planning process and strong words used by both sides in the vivisection debate.
There were just over 2.73 million regulated animal experiments in UK labs in 2002 - according to the latest statistics provided by the Home Office.
Rodents accounted for 84% of this total. The vast majority of experiments are for research and drug development; safety testing accounts for most of rest.
These figures have been broadly static for a number of years now and they are about half what they were in the 1970s.
They do not tell the whole story, of course. In many cases, the DNA alterations in GM animals fail. These animals have no use in the lab and are simply put down and their deaths not recorded in the official statistics.
The use of transgenic animals in experiments is expected to rise significantly in the coming years as scientists probe the causes of disease using information from the recently completed human, mouse and rat genomes.
The Fund for the Replacement of Animals in Medical Experiments (Frame) says this is a worrying development.
"The problem is that this technology is in its early development, but one would hope that over time there would be refinements in the techniques used to generate GM animals," said Frame scientific officer Sylvia Vaughan.
The lobby group has grave doubts about the idea of a special centre for the three R's.
"We believe there is a lot of work already being done in this area, by our group and others," Ms Vaughan said.
"What is really needed is more money to fund research - not a new centre to coordinate work."