institute a ''universal precautions" policy. All patients were to 
be considered infectious. For example, hospital workers would act 
as if they couldn't tell the difference between HIV -infected and 
other patients; they couldn't stamp an "A" on an HIV carrier or on 
HIV specimens because others might then be treated as less 
hazardous or in a more relaxed manner. While the "universal 
precautions" policy appeared rational, it became applied too 
widely. Because the policy stated that every specimen would be 
treated as infectious if it "may reasonably be thought to contain" 
infectious agents, all were assumed to carry the most dangerous 
infectious agents. Therefore, for simplicity, all biomedical 
materials should be packaged in leak-proof containers. 
Dr. Atlas suggested that the Subcommittee delete "only as an 
etiologic agent" from Appendix H; this would remove such a 
definition from certain recombinant DNA products. Perhaps the 
"etiologic agents" label could be limited to host organisms that 
were classified as "etiologic agents" in Appendix B, unless 
etiologic characteristics are expressed in the recombinant 
organism. The Subcommittee must also consider recombinant DNA 
molecules that can transfer the expression of certain 
characteristics to other organisms. 
Dr. Langston added that the criterion of "infectious" was missing. 
Moreover, Dr. Atlas' suggested definitions might exclude whole 
animals, she noted. Dr. McVicar believed that the Subcommittee 
should consider whole animals as a separate issue later. He 
returned to his suggestion of a single shipping method for 
recombinant DNA products, which all shippers would use. He 
proposed the following new definition for the Federal Regulations 
for Interstate Shipment of Etiologic Agents in 42 CFR, Section 
72.6: "etiologic agents refer to a microbiological agent or its 
toxin that causes or may cause human disease." 
Dr. Atlas asked whether "expression" would fall under that 
definition or would require the Subcommittee to make additional 
recommendations. Normally, a non-pathogenic host with a 
recombinant DNA molecule that was coated for toxin would fall under 
the "etiologic agent" definition because it is an organism that 
could cause disease in humans. Such a definition makes reference 
to a single list for the characteristics of the host. This broad 
definition has worked for Dr. Atlas. Dr. McVicar stated that if 
the recombinant DNA product comes under that definition then it 
should be on the list. Otherwise, it could be removed from the 
list. 
Dr. Stevenson asked who would determine what is on the list, what 
is excluded, and by what authority. He was concerned that 
bureaucrats might lack required scientific expertise. Rather than 
approving an action, they might not act in order to avoid mistakes. 
For example, once a bureaucrat called him to ask whether yeast was 
a microoraanism. and thus subiect to the NIH Guidelines; he felt 
[291] 
Recombinant DNA Research, Volume 13 
