Assessment of Patient Specific Characteristics Influencing Antibiotic Prescribing in the Emergency Department among Patients with Positive Respiratory Viruses Multiplex Polymerase Chain Reaction (PCR) Results. (4th October 2017)

• Record Type: Journal Article
• Authors: Mathews, Samantha; Selent, Cameron; Li, Jeanne (Jing); Deacon, Kyle; Kang-Birken, Lena; Fitzgibbons, Lynn
• Is Part Of: Open forum infectious diseases. Volume 4(2017)Supplement 1
• Page Start: S583
• Page End: S583
• Publication Date: 2017-10-04
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofx163.1524
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: Respiratory viral illnesses (RVI) are a leading cause of emergency department (ED) visits. Despite advances such as rapid multiplex polymerase chain reaction for respiratory viruses (vPCR), empiric and often inappropriate antibiotic (AB) use remains ubiquitous. Our goal was to understand which ED patients with RVI received ABs despite a diagnosis (dx) of RVI by vPCR, with a focus on which factors may influence prescribing. Methods: We performed a retrospective chart review of discharged ED patients with positive (+) vPCR, from October 2016 to April 2017, focusing on demographics, vital signs, comorbidities, recent AB receipt, lab/imaging results, discharge (DC) dx and timing of vPCR results relative to DC. Chi-square analyses and a logistic regression model compared those who did vs. did not receive ABs at DC, in order to identify factors associated with AB treatment in setting of a confirmed viral illness. Results: 424 patients had a + vPCR, with 323 (76%) of these results available prior to DC and, of these, 68 (21%) patients were prescribed ABs. AB receipt correlated with age ( P = 0.04), duration of RVI ( P < 0.001), fever ( P = 0.08), white blood count (WBC) ( P < 0.001), abnormal Chest X-ray (CXR) ( P < 0.001) and DC dx (pneumonia (PNA), P < 0.001; otitis media (OM) P = 0.002). No correlation was seen with co-morbid conditions (immunocompromised P = 0.12; chronic lung disease P = 1), or procalcitonin ( P = 0.87). Dx of influenza had an inverse correlation with likelihood of ABs upon DC ( P < 0.001). Conclusion: For patients in the ED diagnosed with RVI by vPCR, several factors were associated with AB receipt. Older patients, and those with longer duration of illness, higher fevers, higher WBC and radiographic evidence of PNA were more often prescribed ABs. Providers may regard these findings as less likely to be explained by RVI alone, leading to the addition of ABs. A substantial proportion of vPCR results were not available at the time of DC, querying the vPCR role in decision making for these patients. Interestingly, patients with influenza received less ABs than those with other RVI dx. This potentially reflects the belief that influenza alone causes severe disease in the absence of a secondary bacterial process, or indicates the importance of antivirals for influenza as an active intervention, replacing ABs in this role. Disclosures: All authors: No reported disclosures.