1379. Comparison of Inpatient Tuberculosis Screening Methods and Their Effect on Patient Duration in Airborne Isolation

Abstract

BackgroundTuberculosis (TB) remains a significant public health concern, and exposure in healthcare settings is prevalent. Current guidelines recommend testing for TB by acid-fast bacilli (AFB) smear microscopy with 3 sputum samples and/or using nucleic acid amplification test (NAAT), and mycobacterium culture. The purpose of this project is to compare how different TB diagnostic tests affect the duration of stay in respiratory isolation.MethodsThis study was conducted at the Veteran Affairs South Texas hospital, which includes a total of 437 beds. Data were collected retrospectively from medical records. Eligibility included patients admitted to the hospital and placed in airborne isolation for TB screening and diagnosis, had 3 sputum samples collected 8 hours apart and/or had 2 PCR MTB/RIF. Patients were excluded if they had TB or were not undergoing evaluation for TB. Three time periods analyzed included, 3 AFB sputum samples analyzed in-house from December 2012 to January 2014 (Group A), 3 AFB sputum samples analyzed at outside facility during 2013 to 2014 as well as 2 months in 2012 (Group B), and 2 MTB PCR/RIF in house during 2017 and 2018 (Group C). Duration of isolation was compared between groups using the Kruskal–Wallis test. A total number of 815 patients were screened, leaving 105 patients for analysis after exclusion. There were 49 patients analyzed from Group A, 28 from Group B, and 28 from Group C.ResultsCrude analysis of the data showed numerical differences in the total number of days and hours in isolation between the 3 groups. The average (mean) days in isolation were 4.2 for Group A, 7.4 for Group B, and 5.5 for Group C. There was no statistically significant difference in either days or hours of airborne precautions by “rule out” method. Days of isolation in airborne precautions (median IQR) was 4 for all groups (P = 0.3313). Likewise, hours of airborne precautions had a median IQR of 96 for all groups P = 0.4347.ConclusionAlthough there was no statistical significance between the groups, crude analysis did show a numerical difference in the mean total airborne days and hours. Lack of statistical difference may be due to low number of patients, timing of order placement for in-house PCR, and longer than expected stay in airborne precautions.Disclosures All authors: No reported disclosures.