Lessons Learned from Sepsis Microlearning Intervention

Background Improving early recognition and treatment of sepsis is key to decreasing patient mortality. A large academic health system implemented several quality improvement initiatives, yet monthly compliance with sepsis best practices remained low. Objective Develop and evaluate an electronic health record (EHR)-embedded microlearning intervention to address suboptimal adherence to sepsis care best practices. Methods We conducted a randomized stepped-wedge trial of our microlearning intervention with randomization done at the nursing block level. Antibiotic delay and secondary outcomes extracted from the EHR were analyzed using mixed models to account for intracluster correlation. Results The microlearning intervention did not reduce antibiotic delay (mean difference = 0.71 hours; p = 0.49). Despite the alert firing over 30,000 times during the study period, the microlearning intervention was viewed only a total of 30 times. Conclusion Our microlearning intervention did not improve sepsis care outcomes. We believe that although the content addressed key knowledge gaps, delivering the intervention through disruptive EHR alerts was not an accessible delivery channel to the nursing staff we targeted.