65 
an internal infection, to a blood infection; and there is a great deal of 
experience, and I will say a great deal of knowledge, and a great problem 
in hospitals with people getting E. col i infection from cuts. The lack of 
the discussion of the route of infection, cuts, and penetration leads to a 
raisfocus of the biohazards discussion, which focuses on ingestion, and has 
no serious discussion of the problem of infection through a wound. For ex- 
ample, I did not hear data at the Falmouth Conference presented, or at least 
in any reasonable quality, of what happens when K-12, for example, or vari- 
ous other E. coli strains, are introduced internally via a cut or a wound. 
It may not colonize your gastrointestinal tract, but in the blood stream it 
is not competing with wild strains of E. coli . So this discussion of physi- 
cally how do people pick up infections in the laboratory is central data for 
formulating the further data that need to be developed. 
Also, my last point, how does one monitor performance of a PI , P2, P3, 
?U lab? My feeling is that you have to have, one, mandatory reporting of 
illness, and setting up of a central registry, preferably in the Center for 
Disease Control, to monitor what is the incidence of people in these labs 
getting sick? Not necessarily, again, from the organisms that they are 
working with. They may be getting the flu, but you want to keep track of 
that. If you don't have data on how often they do get an infection, you 
have no way of assessing whether the physical containment is effective or 
not. You are just hoping it is effective. Now, I believe in hope. I 
believe in doing all you can in the absence of data. But where one can 
have data, where one can require mandatory reporting, if the institute 
medical committee cannot find out what the person is sick from, they can 
report "bacteria unidentifiable, sensitive to penicillin." Okay, fine. 
That lets us know that somebody did get an infection and we don't know what 
it is. We will keep an eye open for other infections that we can't identify 
sensitive to penicillin in similar laboratories. 
I would require institutional medical departments to collaborate with 
local departments of public health, because a small institution may not have 
the medical facilities to do this, in trying to identify the infectious 
agent in laboratory-acquired infections, and then reporting that to the cen- 
tral registry. 
DR. FREDRICKSON: Thank you, Dr. King. You may stand there, please, for 
comment s . 
Dr . Sh aw . 
DR. SHAW: I would like to ask Dr. King if you have some suggestions 
on getting the baseline or laboratory infections as controls. It seems to 
me that if we were to use microbiology labs that are not doing recombinant 
DNA experiments as controls we would have a bag of worms to try to analyze 
and sort out, and I am wondering if you have thought through this problem 
as to how we might get a baseline? 
( 269 ] 
