133 
An examination of the experience with laboratory-acquired infections 
at Fort Detrick provides a basis for demonstrating the value of 
biological safety cabinets. The frequency of infections for four periods 
of operation are summarized in a text table of this EIS. The periods 
characterize four levels of physical containment that are similar to 
levels PI, P i, P3, and P4 with respect to required use of biological 
safety cabinets. That is, in the period 1943-45 all research was conducted 
on the open bench (9). Class I and class III biological safety cabinets 
were introduced at Fort Detrick after 1950 and were assigned on a 
risk-priority basis to programs where considerable aerosol exposures 
were likely to occur. After 1960 class I and class III biological 
safety cabinets were generally available for most research programs. 
The example period 1960-69 provides the experience for the only 
program at Fort Detrick that was conducted in accordance with the 
requirements for P4-level physical containment (10). It is this 10-year 
experience, where only one infection occurred, on which our confidence 
for the capability of P4 physical containment is based. 
An absolute correlation between the reduction in rate of laboratory- 
acquired infections and the use of biological safety cabinets is admittedly 
difficult to establish. During the 1960-69 research at Fort Detrick, a 
variety of etiologic agents were employed and the use of biological safety 
cabinets was dependent on their availability and assignment. Also, the 
the introduction of effective vaccines for anthrax in 1954, tularemia 
in 1959, and Venezuelan encephalitis in 1962 may have influenced this 
experience [2]. (It should be noted that the detection of an increase in 
antibody titre, regardless of immunization status, was recorded as a 
laboratory infection. ) Nevertheless, the performance of biological 
safety cabinets in containing test aerosols attests to their importance 
in reducing potential inhalation exposures. 
It is recognized that human error and accidents contribute to 
laboratory-acquired infections [3,12,17]. The Fort Detrick experience 
demonstrates that even the best possible safety equipment cannot prevent 
self-infection through human error. The one infection that occurred 
during the period 1960-69 in Building 1412B was caused by self-inoculation 
with a contaminated needle through the attached glove of a class III 
cabinet. Pike (11) has shown that most accidents that have been correlated 
with laboratory-acquired infections involve (a) the use of needle and 
syringe, (b) overt contact with spills, (c) injury from contaminated 
broken glassware, (d) improper use of pipettes, and (e) bites and 
scratches from infected animals. Occurrence of these accidents, 
however, can be significantly reduced by diligent adherence to the 
NIH Guidelines. 
The text includes further description and evaluation of class III 
cabinets. 
