top of page
Writer's pictureMazen Kherallah

Unlocking the Potential of EHR-Based Sepsis Alerts: Insights from the SCREEN Trial

EHR-based Sepsis Alert
EHR-based Sepsis Alerts

The transition from paper to electronic health records (EHRs) promised to revolutionize healthcare by enabling real-time data analysis to identify and address care gaps proactively. One application of this vision has been the use of EHRs to generate automated alerts, particularly for sepsis, a life-threatening condition associated with organ dysfunction following infection. Sepsis is a leading cause of in-hospital mortality and readmissions, and timely identification and treatment are critical for improving outcomes. Despite these promises, the implementation of EHR-based sepsis alerts has often fallen short, with mixed results and ongoing challenges. The recently published SCREEN (Stepped-wedge Cluster Randomized Trial of Electronic Early Notification of Sepsis in Hospitalized Ward Patients) trial offers valuable insights into both the potential and limitations of these systems.


Do you have EHR-based sepsis alert in your hospital?

  • 0%Yes

  • 0%No

  • 0%I do not know



Understanding the Challenges of Sepsis Alerts

Sepsis alerts aim to improve early recognition and prompt treatment, yet their adoption has faced criticism for issues such as:

  • Delayed recognition and inappropriate antibiotic or resuscitation measures.

  • Alert fatigue due to frequent, non-specific notifications.

  • Bias in underrepresented populations and potential overtreatment.


Despite promising observational studies, the lack of randomized trials has limited confidence in these systems. The SCREEN trial addressed this gap, providing a robust evaluation of the effectiveness of an EHR-based sepsis alert.



The SCREEN Trial: A Landmark Randomized Evaluation

Conducted in 45 wards across five hospitals, the SCREEN trial compared outcomes among patients before and after the activation of a sepsis alert system. The alert, triggered by the quick Sequential Organ Failure Assessment (qSOFA) score, notified bedside nurses and on-call physicians when two of three criteria (hypotension, tachycardia, altered mental status) were met within a 12-hour period.


Key Components of the Intervention

  • Pre-launch training: Clinical staff were educated on sepsis guidelines and trained on the alert system.

  • Real-time notifications: Alerts were communicated via EHR pop-ups and handheld devices.

  • Audit and feedback: Dashboards tracked alert adherence to support quality improvement efforts.


Key Findings

  • High alert acknowledgment: Three-quarters of alerts were acknowledged, and one-third of these were confirmed as sepsis cases.

  • Improved care processes: Alerts increased the ordering of lactate tests and initiation of intravenous fluids.

  • Mortality reduction: The primary outcome of 90-day hospital mortality improved significantly (adjusted relative risk, 0.85; 95% CI, 0.77-0.93; P < .001).


Lessons and Limitations

The SCREEN trial demonstrated the potential of EHR-based alerts to improve patient outcomes, but it also highlighted important complexities:

  1. Intervention complexity: The effectiveness of sepsis alerts depends on the accuracy of prediction models, timeliness of notifications, and preparedness of clinical teams.

  2. Understanding mechanisms: While the trial showed improved mortality, it did not clarify which specific care changes were most impactful or which patients benefited most.

  3. Broader care effects: The trial observed no direct mortality benefit for patients receiving alerts, suggesting the intervention influenced broader care practices.

  4. Contextual factors: The COVID-19 pandemic—occurring mid-study—added a layer of complexity but also demonstrated the robustness of the trial design against such confounding factors.



Future Directions for EHR-Based Interventions

The SCREEN trial’s findings underscore the need for further research to optimize the implementation of EHR alerts. Key priorities include:

  • Mixed-methods evaluations: Incorporating qualitative assessments of clinician experiences to understand barriers and facilitators.

  • Customizing alerts: Tailoring interventions to specific settings and populations.

  • Scaling evaluations: Conducting additional randomized trials to confirm findings and refine best practices.


Leveraging EHRs for clinical decision support remains one of the most promising yet underutilized strategies for improving healthcare delivery. The SCREEN trial provides a significant step forward, but its results should serve as a catalyst for further investigation rather than immediate widespread adoption. With sustained research and evaluation, we can better harness the power of digital health tools to address sepsis and other critical conditions effectively.



REFERENCES


  1. Arabi  YM, Alsaawi  A, Alzahrani  M,  et al; SCREEN Trial Group; Saudi Critical Care Trials Group.  Electronic sepsis screening among patients admitted to hospital wards: a stepped-wedge cluster randomized trial.   JAMA. Published online December 10, 2024. doi:10.1001/jama.2024.25982

    Article Google Scholar

  2. Singer  M, Deutschman  CS, Seymour  CW,  et al.  The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3).   JAMA. 2016;315(8):801-810. doi:10.1001/jama.2016.0287

    Article PubMed Google Scholar Crossref

16 views0 comments

Comentarios


bottom of page