Tuberculosis exposure and control in an urban emergency department

To measure tuberculosis (TB) conversion rates among staff of an urban emergency department compared with other hospital employees (OHEs) and to evaluate the effectiveness of new TB control measures.

A prospective interventional cohort study was performed in an academic adult 1,000-bed urban hospital, with more than 6,000 employees, and an annual ED census of 43,000. As part of a hospital-wide program, all employees were screened for tuberculin reactivity (Siebert purified protein derivative [PPD] of tuberculin) annually during a 3 1/2-year period. Additional information collected on each employee included country of birth, ethnicity, history of bacille Calmette-Gúerin (BCG) vaccination, length and site of employment, and age. At the end of the second year, TB control measures including construction of a new ED facility with TB respiratory isolation rooms, nonrecirculated air, and droplet shields for registrars were implemented. Relative risk (RR) and 95% confidence intervals (CIs) were calculated.

During the first screening cycle, PPD status was obtained on 5,697 hospital employees, 88 of whom worked primarily in the ED. Baseline status was 81% PPD- (induration <5 mm), 9% PPD+ (induration > or =10 mm), and 10% refused skin testing. ED staff did not differ from OHEs with regard to PPD status, age, ethnicity, BCG history, foreign birth, residing in a county with high TB prevalence, or length of employment. During the second cycle, 6 of 50 (12%) previously PPD- ED staff and 51 of 2,514 (2%) previously PPD OHEs converted to PPD+ status (RR=5.9; 95% CI 2.7,13.1). After implementing TB control measures, the conversion rate during the third year dropped to 0 for ED staff, whereas the OHEs had a conversion rate of 1.2% during that cycle.

TB conversion rates were 5.9 times greater in ED staff members than OHEs before engineering controls were implemented. The ED TB conversion rate fell dramatically in the year following new TB controls. Many EDs are high-risk sites for TB exposure and may benefit from similar measures.