Implementation of a Real-Time Electronic Health Record Alert to Optimize Antimicrobial Prescribing for Select Acute Upper Respiratory Tract Infections in the Emergency Department and Ambulatory Clinics

Abstract

Background. Antimicrobial overprescribing for acute upper respiratory infection is common. The use of electronic health records (EHRs) offers clinicians opportunities to integrate real-time decision making tools upon order entry to support evidencebased antimicrobial prescribing. The objective of this project was to implement and assess the antibiotic prescribing impact of a real-time alert for acute bronchitis (AB) and chronic obstructive pulmonary disease (COPD) exacerbation patients. Methods. Patients diagnosed with AB or COPD exacerbation were evaluated in two phases. Phase I focused on emergency department (ED) patients at a tertiary care center. A year later, Phase II was implemented for patients presenting to three pilot clinics within the same integrated health system: internal medicine, family practice, and urgent care. Evidence-based treatment algorithms were designed and endorsed by the antimicrobial stewardship program and key stakeholders. Providers were educated about the algorithms using multiple methods. A real-time EHR alert based on ICD-9 and ICD-10 diagnosis codes was developed. The alert provided a hyperlink to the corresponding treatment algorithm. Post-intervention data were compared to retrospective site-based data for both project phases. Results. Phase I post-intervention data included 446 AB and 35 COPD exacerbation patients who presented to the ED over a 3 month time period. An 18% reduction in antimicrobial prescribing frequency for AB was observed (96% pre- versus 78% post-intervention). Antimicrobial selection shifted to align with the treatment algorithm recommendations yielding a 26% reduction in azithromycin prescription frequency. Phase II post-intervention data included 42 AB and 4 COPD exacerbation patients who presented to the pilot clinics over a 1 month time period. A 15% reduction in antimicrobial prescribing frequency for AB was observed (94% pre versus 79% post-intervention). There was a shift in antimicrobial selection for those who received an antimicrobial, with an 18% increase in doxycycline prescriptions and an 8% reduction in azithromycin prescriptions. Conclusion. An overall reduction in antimicrobial prescription frequency was observed in both study settings after implementation of an EHR alert. Further study is needed to determine sustainability of this program.