The TBN-ST was read by one of the four nurse practitioners who made rounds each morning in the maternity care unit, each of whom had been trained in reading the TBN-ST. The palpation technique was used. The number of millimeters of induration transverse to die long axis of the arm was measured and the time and date of reading was recorded. In an effort to minimize the number of false-negative results due to early interpretation, a positive TBN-ST was defined as any palpable induration present at a reading before 48 h or 10 mm or greater of induration between 48 and 72 h.
All women who had a positive TBN-ST were to receive a chest roentgenogram and further workup as appropriate. Additionally, each woman with a positive TBN-ST or a chest roentgenogram consistent with tuberculosis was referred to the San Diego County Tuberculosis Control office for a confirmatory TBN-ST (if the TBN-ST had been read prior to 48 h) and appropriate therapy, both of which were free to the patient. At the end of the study period, we reviewed the medical records at the San Diego County Tuberculosis Control office to determine how many of the women had obtained follow-up care there and what their clinical course had been. We also determined whether there were any reported cases of tuberculosis among the infants delivered during the study. In order to establish a control population, the medical records of 116 women who delivered between June 26, 1991, and July 15, 1991, who had received adequate prenatal care were reviewed, and the results of all documented TBN-STs were recorded.