An outbreak of chickenpox, involving three patients and one nurse on a pediatric ward, necessitated rapid identification of susceptible employees in conjunction with standard epidemiologic intervention in order to prevent spread to other high-risk patients. Of 46 hospital personnel 15 (33%) gave a negative or unknown history of prior disease. Response to a varicella-zoster skin test was compared with antibody determination as measured by fluorescent antibody to membrane antigen. Correlation of these two screening methods was absolute. Of 46 hospital personnel four (9%) were susceptible to infection (negative skin test and antibody less than 1:4) requiring their removal from the ward. All those with positive histories for prior disease with the virus and 11/15 (73%) with negative or unknown histories were immune as indicated by both tests. A readily available varicella zoster skin test would be an extremely useful epidemiologic tool for screening hospital personnel.