Downloaded from http://jme.bmj.com/ on November 17, 2015 - Published by group.bmj.com
Brief report
Trends in animal use at US research facilities Justin Goodman, Alka Chandna, Katherine Roe Laboratory Investigations Department, People for the Ethical Treatment of Animals, Norfolk, Virginia, USA Correspondence to Dr Alka Chandna, Laboratory Investigations Department, People for the Ethical Treatment of Animals, Norfolk, VA 23510, USA; [email protected] Received 29 July 2014 Revised 16 December 2014 Accepted 5 January 2015 Published Online First 25 February 2015
▸ http://dx.doi.org/10.1136/ medethics-2012-101043
To cite: Goodman J, Chandna A, Roe K. J Med Ethics 2015;41:567–569.
ABSTRACT Minimising the use of animals in experiments is universally recognised by scientists, governments and advocates as an ethical cornerstone of research. Yet, despite growing public opposition to animal experimentation, mounting evidence that animal studies often do not translate to humans, and the development of new research technologies, a number of countries have reported increased animal use in recent years. In the USA—one of the world’s largest users of animals in experiments—a lack of published data on the species most commonly used in laboratories (eg, mice, rats and fish) has prevented such assessments. The current study aimed to fill this gap by analysing the use of all vertebrate animals by the top institutional recipients of National Institutes of Health research funds over a 15-year period. These data show a statistically significant 72.7% increase in the use of animals at these US facilities during this time period—driven primarily by increases in the use of mice. Our results highlight a need for greater efforts to reduce animal use. We discuss technical, institutional, sociological and psychological explanations for this trend.
limited figures do not account for the use of the excluded species noted above. Thus, estimates for the total number of animals used in laboratories vary widely from 17 million to as high as 100 million,14–17 as do approximations about patterns in their use. Some analyses suggest that there has been sizeable growth in animal use over the past several decades due to increased use of genetically modified (GM) mice,3 16 17 while others claim that US animal use has decreased by as much as 50% over the past 25 years.18 While the AWA’s animal use reporting requirements make it impossible to reconcile these figures due to the exclusion of key species, institutions funded by the National Institutes of Health (NIH) are required to submit a report that includes the average numbers of all vertebrate animals (mice, rats, birds, fish, reptiles and amphibians included) held and used for experimental purposes. The NIH does not analyse or publish these data, but the documents can be requested through federal and state open records laws. We used these reports to examine trends in overall animal use at the top institutional recipients of NIH grants.
INTRODUCTION
DATA COLLECTION, ANALYSIS AND RESULTS
Due to ethical concerns about the use of animals in potentially painful, distressful, or otherwise harmful experiments, in the USA and throughout the world, the ‘3Rs’ principle of replacing, reducing and refining the use of animals in experimentation has been embraced by government agencies, the scientific community and animal advocates.1 The need to minimise animal use is codified in the text of US federal regulations and guidelines that govern animal experimentation and comports with growing moral concern regarding the practice,2–4 increasing evidence that animal models do not faithfully translate to humans,5 6 and the development of technologies that supplant animal use.7 In recent years, there have been shifts away from animal use in certain areas, such as chemical toxicity testing and medical education,8 9 and away from species to whom greater moral consideration has been accorded for reasons of familiarity, charisma or likeness to humans—like chimpanzees, cats and dogs.5 10 However, in the USA it has been difficult to ascertain whether the 3Rs have resulted in overall reductions in animal use in experiments due to the exclusion of mice, rats, birds and cold-blooded animals—estimated by industry to constitute 95 to 98 per cent of all animals in US laboratories—from protection and reporting requirements under the federal Animal Welfare Act (AWA).2 11 US government data show that slightly more than a million AWA-regulated animals are used annually in experiments12 and that the number of regulated animals has decreased.13 However, these
Animal use data were obtained from species inventories—reporting the ‘approximate average daily inventory’ of vertebrate animals held at the institution—contained in the Animal Welfare Assurances filed at least once every 4 years by NIH-funded research institutions. Using open records laws, we acquired copies of the three most recent Assurances for a purposive sample of the top 25 largest public and private recipients of NIH funds for 2011. Because all facilities are not on the same reporting schedule, we looked at the three most recent time points for each rather than at specific years. As the NIH does not employ a standardised system for reporting the use of different species, we used a modified version of the categorisation system employed by the Canadian Council on Animal Care, resulting in 10 distinct species categories of interest (table 1). The average daily inventory of vertebrate animals over the three time periods studied was averaged across the 21 institutions for which three inventories were acquired (table 1). Total animal use increased 72.7% over the 15-year period studied, from an average of 74 619 at time point 1 to 128 846 at time point 3. A one-way repeated measures Analysis of Variance (ANOVA) indicated this increase was statistically significant, F (1,20)=6.7, p=0.018. This overall increase in animal use was not consistent across the different species categories. For the 19 institutions that reported individual species use in all reports, a two-way repeated measures
Goodman J, et al. J Med Ethics 2015;41:567–569. doi:10.1136/medethics-2014-102404
567
Downloaded from http://jme.bmj.com/ on November 17, 2015 - Published by group.bmj.com
Brief report Table 1 Number of animals used† (average and total) across three time periods from 1997 to 2012 AVERAGE
All vertebrates Amphibians and reptiles* Birds* Cats Dogs Farm animals* Fish* Mice* Non-human primates Rats* Miscellaneous mammals
TOTAL
Time 1 1997–2003
Time 2 2000–2008
Time 3 2008–2012
Time 1 1997–2003
Time 2 2000–2008
Time 3 2008–2012
74 618.76 849.11 149.32 17.16 34.58 69.53 11 212.00 60 790.11 383.79 3245.89 381.89
121 417.95 531.00 334.47 11.00 28.84 261.05 23 050.32 94 267.47 540.11 2561.74 337.84
128 846.29 2731.00 332.95 7.26 32.63 288.26 22 676.74 99 303.00 587.74 4003.79 709.11
1 566 994 16 133 2837 326 657 1321 213 028 1 155 012 7292 61 672 7256
2 549 777 10 089 6355 209 548 4960 437 956 1 791 082 10 262 48 673 6419
2 705 772 51 889 6326 138 620 5477 430 858 1 886 757 11 167 76 072 13 473
Due to differences in reporting schedules across the institutes sampled, there is overlap in years covered across the three time periods. *Species not covered by the Animal Welfare Act. Some categories straddle covered and non-covered categories; for example, farm animals used in agricultural research are excluded from the Animal Welfare Act, while farm animals used in ‘non-production’ research are included. †Four institutions only had two obtainable species inventories and were excluded from these values and analyses of change over time. Two institutions did not provide complete individual species use data and were excluded from species values provided here and analysis of individual species use over time.
ANOVA was performed, with 3 levels of Time and 10 levels of Species treated as within-subject variables. This analysis also revealed a main effect of Time, F(2,36)=4.713, p=0.015, with total animal use (collapsed across species) increasing reliably across the three time periods. There was also a significant effect of Species Type, F(9,162)=44.3, p
Brief report
Trends in animal use at US research facilities Justin Goodman, Alka Chandna, Katherine Roe Laboratory Investigations Department, People for the Ethical Treatment of Animals, Norfolk, Virginia, USA Correspondence to Dr Alka Chandna, Laboratory Investigations Department, People for the Ethical Treatment of Animals, Norfolk, VA 23510, USA; [email protected] Received 29 July 2014 Revised 16 December 2014 Accepted 5 January 2015 Published Online First 25 February 2015
▸ http://dx.doi.org/10.1136/ medethics-2012-101043
To cite: Goodman J, Chandna A, Roe K. J Med Ethics 2015;41:567–569.
ABSTRACT Minimising the use of animals in experiments is universally recognised by scientists, governments and advocates as an ethical cornerstone of research. Yet, despite growing public opposition to animal experimentation, mounting evidence that animal studies often do not translate to humans, and the development of new research technologies, a number of countries have reported increased animal use in recent years. In the USA—one of the world’s largest users of animals in experiments—a lack of published data on the species most commonly used in laboratories (eg, mice, rats and fish) has prevented such assessments. The current study aimed to fill this gap by analysing the use of all vertebrate animals by the top institutional recipients of National Institutes of Health research funds over a 15-year period. These data show a statistically significant 72.7% increase in the use of animals at these US facilities during this time period—driven primarily by increases in the use of mice. Our results highlight a need for greater efforts to reduce animal use. We discuss technical, institutional, sociological and psychological explanations for this trend.
limited figures do not account for the use of the excluded species noted above. Thus, estimates for the total number of animals used in laboratories vary widely from 17 million to as high as 100 million,14–17 as do approximations about patterns in their use. Some analyses suggest that there has been sizeable growth in animal use over the past several decades due to increased use of genetically modified (GM) mice,3 16 17 while others claim that US animal use has decreased by as much as 50% over the past 25 years.18 While the AWA’s animal use reporting requirements make it impossible to reconcile these figures due to the exclusion of key species, institutions funded by the National Institutes of Health (NIH) are required to submit a report that includes the average numbers of all vertebrate animals (mice, rats, birds, fish, reptiles and amphibians included) held and used for experimental purposes. The NIH does not analyse or publish these data, but the documents can be requested through federal and state open records laws. We used these reports to examine trends in overall animal use at the top institutional recipients of NIH grants.
INTRODUCTION
DATA COLLECTION, ANALYSIS AND RESULTS
Due to ethical concerns about the use of animals in potentially painful, distressful, or otherwise harmful experiments, in the USA and throughout the world, the ‘3Rs’ principle of replacing, reducing and refining the use of animals in experimentation has been embraced by government agencies, the scientific community and animal advocates.1 The need to minimise animal use is codified in the text of US federal regulations and guidelines that govern animal experimentation and comports with growing moral concern regarding the practice,2–4 increasing evidence that animal models do not faithfully translate to humans,5 6 and the development of technologies that supplant animal use.7 In recent years, there have been shifts away from animal use in certain areas, such as chemical toxicity testing and medical education,8 9 and away from species to whom greater moral consideration has been accorded for reasons of familiarity, charisma or likeness to humans—like chimpanzees, cats and dogs.5 10 However, in the USA it has been difficult to ascertain whether the 3Rs have resulted in overall reductions in animal use in experiments due to the exclusion of mice, rats, birds and cold-blooded animals—estimated by industry to constitute 95 to 98 per cent of all animals in US laboratories—from protection and reporting requirements under the federal Animal Welfare Act (AWA).2 11 US government data show that slightly more than a million AWA-regulated animals are used annually in experiments12 and that the number of regulated animals has decreased.13 However, these
Animal use data were obtained from species inventories—reporting the ‘approximate average daily inventory’ of vertebrate animals held at the institution—contained in the Animal Welfare Assurances filed at least once every 4 years by NIH-funded research institutions. Using open records laws, we acquired copies of the three most recent Assurances for a purposive sample of the top 25 largest public and private recipients of NIH funds for 2011. Because all facilities are not on the same reporting schedule, we looked at the three most recent time points for each rather than at specific years. As the NIH does not employ a standardised system for reporting the use of different species, we used a modified version of the categorisation system employed by the Canadian Council on Animal Care, resulting in 10 distinct species categories of interest (table 1). The average daily inventory of vertebrate animals over the three time periods studied was averaged across the 21 institutions for which three inventories were acquired (table 1). Total animal use increased 72.7% over the 15-year period studied, from an average of 74 619 at time point 1 to 128 846 at time point 3. A one-way repeated measures Analysis of Variance (ANOVA) indicated this increase was statistically significant, F (1,20)=6.7, p=0.018. This overall increase in animal use was not consistent across the different species categories. For the 19 institutions that reported individual species use in all reports, a two-way repeated measures
Goodman J, et al. J Med Ethics 2015;41:567–569. doi:10.1136/medethics-2014-102404
567
Downloaded from http://jme.bmj.com/ on November 17, 2015 - Published by group.bmj.com
Brief report Table 1 Number of animals used† (average and total) across three time periods from 1997 to 2012 AVERAGE
All vertebrates Amphibians and reptiles* Birds* Cats Dogs Farm animals* Fish* Mice* Non-human primates Rats* Miscellaneous mammals
TOTAL
Time 1 1997–2003
Time 2 2000–2008
Time 3 2008–2012
Time 1 1997–2003
Time 2 2000–2008
Time 3 2008–2012
74 618.76 849.11 149.32 17.16 34.58 69.53 11 212.00 60 790.11 383.79 3245.89 381.89
121 417.95 531.00 334.47 11.00 28.84 261.05 23 050.32 94 267.47 540.11 2561.74 337.84
128 846.29 2731.00 332.95 7.26 32.63 288.26 22 676.74 99 303.00 587.74 4003.79 709.11
1 566 994 16 133 2837 326 657 1321 213 028 1 155 012 7292 61 672 7256
2 549 777 10 089 6355 209 548 4960 437 956 1 791 082 10 262 48 673 6419
2 705 772 51 889 6326 138 620 5477 430 858 1 886 757 11 167 76 072 13 473
Due to differences in reporting schedules across the institutes sampled, there is overlap in years covered across the three time periods. *Species not covered by the Animal Welfare Act. Some categories straddle covered and non-covered categories; for example, farm animals used in agricultural research are excluded from the Animal Welfare Act, while farm animals used in ‘non-production’ research are included. †Four institutions only had two obtainable species inventories and were excluded from these values and analyses of change over time. Two institutions did not provide complete individual species use data and were excluded from species values provided here and analysis of individual species use over time.
ANOVA was performed, with 3 levels of Time and 10 levels of Species treated as within-subject variables. This analysis also revealed a main effect of Time, F(2,36)=4.713, p=0.015, with total animal use (collapsed across species) increasing reliably across the three time periods. There was also a significant effect of Species Type, F(9,162)=44.3, p
