BACKGROUND: Recent nosocomial outbreaks have raised concern about the risk of Mycobacterium tuberculosis transmission in United States hospitals.
METHODS: To determine current tuberculosis (TB) infection control practices, we surveyed a sample of approximately 3000 acute care facilities about the number of patients with drug-susceptible or multidrug-resistant TB (MDR-TB), health care worker (HCW) tuberculin skin test (TST) results, and compliance with the 1990 Centers for Disease Control and Prevention (CDC) TB guidelines. Analyses were restricted to one response per hospital.
RESULTS: Personnel at 1494 (49.8%) hospitals returned a completed survey. Respondent hospitals had a mean of 881 HCWs (range 8 to 10,000) and 196 (range 6 to 2450) beds; 62% percent were community nonteaching hospitals. Of respondent hospitals providing data for 1989 through 1992, the proportion that cared for patients with TB or MDR-TB increased from 46.4% to 56.6% and 0.8% to 4.5%, respectively. The pooled mean HCW TST positivity rate at hire rose from 0.95% to 1.14%, and the pooled mean HCW TST conversion rate increased from 0.40% to 0.51%. In 1992, when we compared hospitals with zero, one to five, or six or greater patients with TB, the risk of a positive HCW TST result at hire or at routine testing significantly increased with increasing number of patients with TB. From 1989 through 1992, the number of hospitals reporting the use of surgical masks for HCW respiratory protection decreased from 96.8% to 66.8%. In 1992, 66% of the hospitals reported compliance with four or more of the AFB isolation room criteria specified in the 1990 CDC TB guidelines.
CONCLUSIONS: Contrary to prior surveys, this study shows that many U.S. community hospitals admit patients with TB less frequently than do teaching hospitals, and infrequently admit patients with MDR-TB. Because the risk of HCW TST conversion varies with hospital characteristics, these data show the importance of performing a risk assessment, as recommended in the CDC TB guidelines, for each ward and hospital so that TB control measures can be individualized.
Sinkowitz RL, Fridkin SK, Manangan L, Wenger PN, Jarvis WR
Am J Infect Control 1996 Aug;24(4):226-34