urgent calls and identify requests that may be dealt with by other team members such as Infection control nurses or laboratory staff. The audit is based on the Royal college of pathologist Key perfor- mance indicators on availability of interpretative clinical advice and engagement with multidisciplinary teams. The aim was to deter- mine what percentage of all requests for advice received by the duty clinical microbiologist were inappropriate. METHODS: Prospective audit of all microbiology consultant advice requests received between 13 and 25 October 2016. A template was used to record clinical advice given by telephone both in and out of hours. RESULTS: General advice calls constituted 71% and 34% were in- appropriate. Result interpretation were 12% and 74% were inappropriate. Sensitivity requests represented 11% and 47% were inappropriate. Infection control calls were 3% and 20% were inap- propriate. Prophylaxis requests were 2% and all were appropriate. There was failure to look up the Hospital Antimicrobial guidance prior to calling the consultants. Junior staff were more likely to make inappropriate general advice requests. Poor indication for antibi- otic escalation and lack of using own clinical judgement was identified. Lack of generic result interpretation comment on some results or failure of requestors to read the comments was identi- fied. Calls for further sensitivities could have been directed to the laboratory staff. There was failure by some requestors to decipher which antibiotics to select where more than one isolate was re- ported from one sample. CONCLUSIONS: There were many inappropriate calls made by Junior staff to consultant Microbiologists. The laboratory policy should include risk stratification for the urgency of response to requests and which members of staff to deal with different levels of requests and for which requestors. Generic comments to accompany most results. Session EPOP-108 12:30-1:30 p.m. An Urban 40 Bed Level III Neonatal Intensive Care Unit reduced Catheter-Associated Bloodstream Infections by Implementing New Guidelines Roberta Glenn, BS, MT(ASCP), CIC, ICP, Thomas Jefferson University Hospital BACKGROUND: Catheter-associated bloodstream infection (CLABSI) in the neonatal population is a major source of morbidity and mor- tality. The disruption of skin and mucus membranes by the use of invasive devices contributes to the susceptibility of this popula- tion. In our Neonatal Intensive Care Unit (NICU), the CLABSI rate was continually increasing. It became evident that new practice guide- lines were necessary to reduce the CLABSI rate in this vulnerable population. METHODS: Initial interventions were instituted in 2010, includ- ing central line insertion and maintenance checklists, daily line necessity evaluation, introduction of a closed medication admin- istration system, and reinforcement of hand hygiene. Additionally in the following two years, clean gloves for all medication admin- istration was instituted along with a PICC team and changing all central line tubing every 96 hours using a two person aseptic tech- nique with sterile gloves and masks over a sterile field. Finally in 2014 - 2015, chlorhexidine gluconate (CHG) skin prep and CHG discs were implemented using strictly defined age appropriate proto- cols for all central line insertions and dressing changes and unused injection ports were covered with disinfecting port protectors. RESULTS: In 2010 the CLABSI rate was 6.52 (20 infections). In 2012, after the initial interventions were implemented, the rate dropped to 0.42 (1 infection) which is statistically significant by Chi-square analysis (P-value = .0001). The infection rate began to rise in 2013 and 2014 and prompted further initiatives. This resulted in zero CLABSIs for greater than 16 months. CONCLUSIONS: Teamwork and vigilance in following the various practice initiatives has resulted in sustained reduction of CLABSIs in the NICU, contributing significantly to patient safety and good outcomes. Session EPOP-109 12:30-1:30 p.m. Analysis of a Tuberculosis Post-Exposure Event: An Algorithm and Recommendations for TB Contact Investigations Emily Robbins, MPH, Graduate Student—Graduated Dec 2016, University of Pittsburgh, Graduate School of Public Health; Marian Pokrywka, MS, MT(ASCP), CIC, Infection Preventionist, UPMC Mercy Hospital; Linda Rose Frank, PhD, MSN, ACRN, FAAN, Professor, Public Health, Medicine, and Nursing, University of Pittsburgh Graduate School of Public Health; Mohamed Yassin, MD, PhD, Medical Director, Infection Control—Hospital Epi, UPMC Mercy—University of Pittsburgh BACKGROUND: Immunosuppressed patients, such as those under- going chemotherapy and radiation, are at greater risk of developing active tuberculosis (TB). Without prompt diagnosis, the risk of noso- comial transmission is increased for patients and staff alike. An oncology patient with active TB was undiagnosed for several weeks at a large, university-associated hospital due to the case’s clinical presentation and complex chest X-rays (CXRs). Other immunosup- pressed and immunocompetent patients were exposed to the index case. A contact investigation was completed by the hospital’s in- fection prevention (IP) department. METHODS: A retrospective analysis of the TB contact investiga- tion was conducted. Patient charts were reviewed to determine the rate of compliance of Tuberculin Skin Test (TST) among notified pa- tients as well as the TST conversion rate. Recommendations for an improved contact investigation protocol were formulated. A contact investigation algorithm for infection preventionists (IPs) was de- veloped for use in future exposure events. RESULTS: One hundred forty-two patients were identified as po- tential contacts. Oncology patients (67) represented 47% of the exposures. Of the 142 contacts, 120 were successfully notified of the exposure by certified mail. Thirty-two percent of the patient con- tacts completed the recommended TST. There were no detected conversions following exposure to the index patient. However, unknown latent tuberculosis infection (LTBI) was identified in 3 on- cology patients. CONCLUSIONS: Low TST compliance limited identification of all po- tential conversions but findings were consistent with other published studies. New notification methods should be considered such as phone calls to patients and PCP notifications to improve TST com- pliance. Three cases of LTBI were identified emphasizing the importance of TB screening in high-risk populations, such as on- cology patients. Improved protocols should be written for TB exposure events in hospital settings. A TB Contact Investigation Al- gorithm was developed as a result of this study as a tool for use by IPs conducting post-exposure investigations. S55 Poster Abstracts / American Journal of Infection Control 45 (2017) S16-S93 APIC 44th Annual Educational Conference & International Meeting | Portland, OR | June 14-16, 2017