Characteristics and Outcomes of Sepsis Presenting in Inpatient Pediatric Settings.

OBJECTIVE: The pediatric sepsis literature lacks studies examining the inpatient setting, yet sepsis remains a leading cause of death in children's hospitals. More information is needed about sepsis arising in patients already hospitalized to improve morbidity and mortality outcomes. This study describes the clinical characteristics, process measures, and outcomes of inpatient sepsis cases compared with emergency department (ED) sepsis cases within the Improving Pediatric Sepsis Outcomes data registry from 46 hospitals that care for children. METHODS: This retrospective cohort study included Improving Pediatric Sepsis Outcomes sepsis cases from January 2017 to December 2019 with onset in inpatient or ED. We used descriptive statistics to compare inpatient and ED sepsis metrics and describe inpatient sepsis outcomes. RESULTS: The cohort included 26 855 cases; 8.4% were inpatient and 91.6% were ED. Inpatient cases had higher sepsis-attributable mortality (2.0% vs 1.4%, P = .025), longer length of stay after sepsis recognition (9 vs 5 days, P <.001), more intensive care admissions (57.6% vs 54.1%, P = .002), and greater average vasopressor use (18.0% vs 13.6%, P <.001) compared with ED. In the inpatient cohort, >40% of cases had a time from arrival to recognition within 12 hours. In 21% of cases, this time was >96 hours. Improved adherence to sepsis treatment bundles over time was associated with improved 30-day sepsis-attributable mortality for inpatients with sepsis. CONCLUSIONS: Inpatient sepsis cases had longer lengths of stay, more need for intensive care, and higher vasopressor use. Sepsis-attributable mortality was significantly higher in inpatient cases compared with ED cases and improved with improved sepsis bundle adherence.

Using Quality Improvement Science to Promote Reliable Communication During Family-Centered Rounds.

BACKGROUND AND OBJECTIVES: Family-centered rounds (FCR) can lead to improved communication, satisfaction, and care delivery. However, FCR are variable in practice. Our primary goal was to implement and sustain consistent communication practices during FCR (a subset of all rounds in which parents were present) for patients on a pediatric hospital medicine service. We aimed to achieve 80% reliability for the following FCR practices: (1) discussion of risk factors and prevention strategies for hospital-acquired conditions (HACs), (2) discussion of discharge planning, and (3) asking families for questions. METHODS: Research assistants observed FCR on a pediatric acute care unit at an academic medical center and recorded if the rounding team discussed HAC risk factors, discussed discharge, or asked families for questions. Using the Model for Improvement, we performed multiple plan-do-study-act cycles to test and implement interventions, including (1) standardized note templates, (2) education via peer-led group discussions and team e-mails, and (3) routine provider feedback about performance. Data were analyzed by using statistical process control charts. RESULTS: From October 2017 to April 2019, reliability increased to >80% and sustained for (1) discussion of HAC risk factors (increased from 11% to 89%), (2) discussion of discharge planning (from 60% to 92%), and (3) asking families for questions (from 61% to 87%). Peer-led physician education, reminder e-mails, and physician engagement were the most impactful interventions corresponding to centerline shifts. CONCLUSIONS: Using multiple interventions, we achieved and sustained improvements in key communication-related elements of FCR. Future work will focus on determining if improved practices impact clinical outcomes.

Implementation of Febrile Infant Management Guidelines Reduces Hospitalization.

The clinical management of well-appearing febrile infants 7-60 days of age remains variable due in part to multiple criteria differentiating the risk of a serious bacterial infection. The purpose of this quality improvement study was to standardize risk stratification in the emergency department and length of stay in the inpatient unit by implementing an evidence-based clinical practice guideline (CPG). METHODS: The Model for Improvement was used to implement a CPG for the management of well-appearing febrile infants, with collaboration between pediatric emergency medicine and pediatric hospital medicine physicians. Interventions included physician education, process audit/feedback, and development of an electronic orderset. We used statistical process control charts to assess the primary aims of appropriate risk stratification and length of stay. RESULTS: Over a 34-month period, 168 unique encounters (baseline n = 65, intervention n = 103) were included. There was strong adherence for appropriate risk stratification in both periods: the proportion of low-risk patients admitted inappropriately decreased from 14.8% to 10.8%. Among admitted high-risk patients, the mean length of stay decreased from 49.4 to 38.2 hours, sustained for 18 months. CONCLUSION: CPG implementation using quality improvement methodology can increase the delivery of evidence-based care for febrile infants, leading to a reduction in length of stay for high-risk infants.