Improving Critical Care Documentation Accuracy in a Pediatric Emergency Department

Purpose/Objectives: Clinician documentation is highly variable. Many physicians are not given adequate, thorough education in billing, coding, and compliance during training. Awareness of documentation requirements remains low despite post-training clinical experience. Consequently, inadequate documentation can lead to underbilling where the level of expertise and clinical services provided are not appropriately billed. Especially at academic, not-for-profit institutions, maintaining appropriate staffing and resources depends on adequate reimbursement for care provided. At our hospital, critical care documentation was highly variable in the emergency department, so we implemented a project to increase appropriate critical care attestations from 51% to 90% for status asthmaticus, anaphylaxis, and diabetic ketoacidosis in the pediatric emergency department by April 2022.

Design/Methods: A physician team was formed in January 2020 and developed a key driver diagram (Figure 1). Retrospective baseline data using ICD-9 and ICD-10 codes were obtained from January 2018 to September 2020 after which data were followed prospectively in consecutive groups of 20 encounters. Statistical process control charts were used to analyze data. Nelson rules were utilized to detect special cause variation. The primary outcome was the inclusion of a critical care attestation. Lower (Level 1) reliability interventions surrounding awareness and education were employed starting in September 2020. Revised templated language for critical care encounters, a higher reliability intervention (Level 2), was also introduced in the electronic medical record in September 2020 and its use was monitored as a process measure.

Results: P-charts were used to analyze primary outcome and process measure. Baseline data represented 263 encounters with inclusion of critical care documentation in 51% of encounters. Following clinician education and release of the critical care template, special cause variation was detected, and the centerline shifted to 88.1% (Figure 2). A second centerline shift was again detected in December 2020 from 88.1% to 94.4% with a similar temporal shift noted with the process measure of template usage from 34.4% to 51.1%. Other interventions during these improvements included monthly feedback and weekly provider notifications of documentation deficiency. These levels of performance for the primary outcome and process measure have been sustained for 6 months.

Conclusion/Discussion: Through quality improvement methodology, we successfully achieved improvements in critical care documentation in the three diagnoses of interest. Improvements after Level 1 reliability interventions support the need for documentation requirement education and feedback to clinicians. Further improvements in documentation with increased usage of a critical care documentation template, a Level 2 reliability intervention, helped clinicians by simplifying and standardizing documentation workflow and requirements. Ongoing efforts are focused on maintaining this level of accurate documentation.