Impact of adverse events on patient outcomes in a Japanese intensive care unit: a retrospective observational study.

AIM: We investigated adverse events (AEs) in a Japanese intensive care unit (ICU) and evaluated the impact of cause-specific AEs on mortality and length of stay. DESIGN: A retrospective observational study in the ICU of an academic hospital. METHODS: We reviewed medical records with the Global Trigger Tool. RESULTS: Of the 246 patients, 126 (51%) experienced one or more AEs with an incidence of 201 per 1000 patient-days and 115 per 100 admissions. A total of 294 AEs were detected with 119 (42%) adverse drug events, 67 (24%) procedural complications, 63 (22%) surgical complications, 26 (9%) nosocomial infections, 5 (2%) therapeutic errors and 4 (1%) diagnostic errors. Adverse event (AE) presence was associated with length of ICU stay (ß = 2.85, 95% confidence interval [CI]: 1.09-4.61). Adverse drug events, procedural complications and nosocomial infections were strongly associated with length of ICU stay (ß = 2.38, 95% CI: 0.77-3.98; ß = 3.75, 95% CI: 2.03-5.48; ß = 6.52, 95% CI: 4.07-8.97 respectively).

Improving Antibiotic Administration Rate for Patients With Sepsis in the Emergency Department.

BACKGROUND: Previous studies have demonstrated that delayed antibiotic administration increases the risk of mortality in patients with sepsis. LOCAL PROBLEM: In the emergency department, the antibiotic administration rate within 1 and 3 hours for patients with suspected sepsis was low. METHODS/INTERVENTIONS: We implemented an educational intervention with the nursing staff, which included training sessions and feedback, to ensure early detection and management of patients with suspected sepsis. Antibiotic administration rates were compared before and after education. RESULTS: A total of 503 patients were included. The antibiotic administration rate improved as the phases continued (1 hour: from 5.2% to 15.6%, P = .004; 3 hours: from 35.6% to 49.7%, P = .04; 6 hours: from 74.1% to 89.1%, P = .002). The time to initial antibiotic administration also improved from 229 to 185 minutes (P < .001). CONCLUSIONS: Nurse-initiated quality improvement improved the early administration of antibiotics for patients with suspected sepsis.