745 - Evaluating a Clinical Decision Support Tool for Inpatient RSV prophylaxis in the Neonatal Intensive Care Unit

Poster Number: 745
Publication Number: 745.245

Emerald L. Hanson, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa city, IA, United States; Thomas Feldman, University of Iowa Stead Family Children's Hospital, Iowa City, IA, United States; James M. Blum, University of Iowa, Iowa City, IA, United States; Lindsey Knake, University of Iowa Stead Family Children's Hospital, Iowa City, IA, United States

Background: Respiratory syncytial virus (RSV) causes severe respiratory infection in high-risk infants. The American Academy of Pediatrics recommends Palivizumab as RSV prophylaxis for these patients.

Objective: To evaluate a novel electronic health record (EHR) clinical decision support (CDS) tool (Figure 1) that identifies admitted patients who qualify for inpatient RSV prophylaxis administration

Design/Methods: A baseline needs assessment was sent to practitioners in the NICU prior to implementation of the CDS tool. A repeat survey will be sent after the 2022–2023 season to evaluate effectiveness of the tool.
Based on our unit protocol, patients qualified for monthly inpatient RSV prophylaxis if they had a high-risk diagnosis, gestational age < 29 weeks, or gestational age between 29 and 32 weeks with respiratory support of >21% FiO2 for first 28 days of life. Our CDS tool was built using these criteria and included high-risk patients based on ICD codes in the hospital problem list. Baseline data were obtained by retrospectively applying the tool to birth admissions during the 2021-2022 RSV season (Oct-Feb). Readmitted and deceased prior to discharge patients were excluded from the analysis. The two-by-two table was based on an EHR-generated report and manual chart review.

Results: 39 NICU faculty and staff responded to the survey (Figure 2). Many practitioners were confident in their ability to determine patient eligibility for inpatient prophylaxis administration (67%), but also felt they cannot easily identify patients using current EHR tools (54%).
There were 380 patients who met inclusion criteria. Of those, 111 qualified for RSV prophylaxis (29%) and 269 did not qualify (71%). The CDS tool had a specificity of 97% and sensitivity of 80% (Table 1). The overall agreement of our CDS tool and clinicians’ decision making was 350 of the 380 patients (92%). The tool suggested a patient qualified for RSV prophylaxis 8 times and the patient did not receive RSV prophylaxis. Based on chart review, 4 patients were transferred to other facilities and the other 4 had criteria documented that may have qualified them for RSV prophylaxis. Our CDS tool missed 22 patients who received RSV prophylaxis due to congenital heart disease and neuromuscular abnormalities that were not documented or not included in the CDS list of diagnoses.

Conclusion(s): The survey highlighted the need to improve the ease by which clinicians identify patients for RSV prophylaxis. The CDS tool overall had high agreement with clinicians ordering RSV prophylaxis with only a small number of patients missed due to inaccurate problem list documentation.