#87: Association Between Healthcare-Associated Respiratory Viral Infections and Length of Stay at a Pediatric Hospital

Background Healthcare-associated infections (HAI) are major preventable causes of morbidity and mortality. While there are fewer overall HAI in children, there is a greater potential impact in disability-adjusted life years. Healthcare-associated respiratory viral infections (HARVI) are not frequently tracked within institutions, yet the risk for such infections in pediatric hospitals is very high. Recent data demonstrate large inter-hospital variability of HARVI incidence that may depend on various factors including the number of immunocompromised patients in the hospital and the presence of shared rooms. We hypothesize that the burden of healthcare-associated respiratory viral infections and their impact on the length of stay (LOS) is substantial at a large urban pediatric hospital. Methods A cohort of all children with any HARVI admitted to a large urban pediatric hospital between July 2017 and June 2018 were included after obtaining IRB approval. We defined a HARVI as a respiratory infection with an onset of symptoms while the patient was hospitalized meeting three criteria: A positive microbiologic test for one of 8 viruses, presence of symptoms of a respiratory infection, and onset of symptoms after admission beyond the minimum incubation period for each virus. Infections with symptom onset after admission beyond the maximum incubation period were considered definite hospital onset whereas others were considered possible hospital onset. The electronic medical record provided data on demographics, underlying medical conditions, hospital length of stay prior to infection and hospital unit of infection, and consequences and outcome of HARVI. The at-risk population for calculation of the incidence of HARVI was all admitted patient-days at the hospital over this time period. Results Between July 2017 and June 2018 the incidence of HARVI (definite or possible hospital onset) was 1.2 infections per 1,000 admitted patient-days (60% due to rhinovirus/enterovirus, 12% due to respiratory syncytial virus, and 9% due to influenza). Overall, 48% of patients were under 2 years of age, 18% were between 2 and 5 years of age, and 34% were over 5 years of age. Twenty-one percent were immunocompromised and 35% had underlying lung disease. The median length of stay prior to symptom onset was 11 days (IQR 5–36 days) and the median total length of stay was 30 days (IQR 15–82.5 days). Eight individuals had more than one HARVI over this time period. Nineteen percent were transferred to the intensive care unit and 7% died during their hospital admission Conclusion HARVI occurs frequently in a pediatric hospital and often in patients with underlying comorbidities. The risk for HARVI increases substantially with increased length of stay. Such data support the need for tracking HARVI in high-risk institutions.