83 - National Needs Assessment of Neonatal Care Providers' Utilization of Clinical Decision Support Tools

Poster Number: 83
Publication Number: 83.437

Kristyn Beam, Beth Israel Deaconess Medical Center, Waltham, MA, United States; Cindy Wang, Harvard University, Cambridge, MA, United States; Andrew Beam, Harvard School of Public Health - Boston, MA, Waltham, MA, United States; Veeral Tolia, Texas A&M Health Science Center College of Medicine, Dallas, TX, United States; Kaashif A. Ahmad, MEDNAX, Houston, TX, United States

Background: Clinical decision support tools (CDS) exist in neonatology for a variety of scenarios but few are implemented into clinical care. Reported barriers for CDS use and implementation include lack of availability at the point of care, difficulty with access, challenging user interfaces, difficulty interpreting the CDS output and recommendations, or a simple lack of recommendations. Prior to development of new CDS, understanding these barriers is critical.

Objective: To examine the barriers to use of four well-known CDS in newborn care. The tools examined included the Kaiser Permanente Early Onset Sepsis Risk Calculator (EOS), Bilitool.org, the Bronchopulmonary Dysplasia Outcome Estimator (BPD), and the NICHD Extremely Preterm Birth Outcomes Tool (ELBW).

Design/Methods: We developed a 72-field needs assessment questionnaire. After a pilot period, it was distributed to three listservs encompassing neonatal trainees, neonatal nurse practitioners, pediatric hospitalists, and neonatal attendings. The questionnaire was open for three months and several reminders were sent to increase responses. At the conclusion of data collection, responses were downloaded and deidentified. Data were analyzed with descriptive statistics using R.

Results: We received 426 questionnaires and 339 had complete demographic data (summarized in Table 1). Bilitool and the EOS tool were used by 411 (96.5%) and 395 (92.7%) of respondents, respectively. The ELBW tool by 305 (71.6%) and the BPD tool by 166 (39%) (Figure 1). Figure 2 summarizes how useful respondents found each tool. Individuals reported that the BPD tool did not impact their clinical decision-making due to a lack of integration into the electronic health record (21/55, 38.2%), lack of confidence in prediction accuracy (20/55, 36.4%), and unhelpful predictions (6/55, 10.9%). Similar reasons were cited for the infrequent incorporation of the ELBW tool into clinical scenarios. 31% (29/92) of respondents stated their institution has a guideline that uses the BPD tool output to guide treatment, with steroids as the most frequent treatment. 66% (41/62) of respondents use the BPD tool for clinical decisions in an individual capacity.Conclusion(s): From a large sample of neonatal care providers, there is variable use of four CDS spanning from 39% to 96%. Of the two tools that have lower use among neonatal providers, lack of integration into the EHR as well as lack of confidence in prediction accuracy rated as the most important factors. Utilizing this information in the development of future risk prediction tools will be important to ensure uptake into a clinical workflow.
Table 1. Demographics.
Figure 1. Do you use the following tools?Tool Usage by Position.jpegThe percentage of neonatal providers that used each tool grouped by position. The EOS tool and Bilitool were used the most by all position levels. Attending neonatologists and neonatal fellows used the ELBW Tool and BPD Tool more than other positions.