Carnegie -Mellon University 
Department of Engineering and Public Policy 
Schenley Park 
Pittsburgh, Pennsylvania 15213 
[412] 578-2672 
May 16, 1978 
Dr. Donald S. Fredrickson 
Director 
National Institutes of Health 
9000 Rockville Pike 
Bethesda, Maryland 20014 
Dear Dr. Fredrickson: 
While I presently hold a visiting appointment in the Department of 
Engineering and Public Policy at Carnegi e-Mellon University, my Ph.D. is 
in biochemistry. This past semester I taught a course which I entitled ’’Public 
Policy Issues in Biochemistry," which examined how some of the issues and 
techniques traditionally associated with biological research are coming in- 
creasingly into the realm of public policy issues. An obvious topic that we 
discussed in the course was recombinant DNA. 
In preparing this section of the course, I reviewed a considerable por- 
tion of the literature, with an emphasis on risk evaluation and cost-benefit 
frameworks. My purpose in writing to you is to delineate some techniques 
and methods which might assist in future decision-making processes on this 
issue. 
I read the proposed guideline revisions which appeared in the Federal 
Register (September 27, 1977) and I discussed them recently with Dr. Kamely 
in the Office of Recombinant DNA Activities. She indicated a significant 
problem with finalizing the revised guidelines is that the field is changing 
so rapidly that it is difficult to write a set of guidelines to cover all contin- 
gencies. A review of the literature would also suggest a considerable diversity 
of thoughts on the nature of appropriate guidelines. 
Certainly the basic problem in assessing recombinant DNA activities 
is the uncertainty in the risk evaluation and the resulting subjectivity required 
in an evaluation of the proper containment for particular experiments. In 
addition, the series of events which might lead to an organism escape and 
environmental impact are probabilistic in nature. There appear to me a set 
of quantitative, analytical tools which would be useful, in a problem evalua- 
tion, depending on the nature of the question asked. For instance: 
1. Is the combination of physical containment X and biological contain- 
ment Y sufficient to reduce the probability of Z fatalities below 
probability P? 
[Appendix A — 316] 
