Development and validation of an algorithm to identify severe sepsis onset from electronic medical records.

OBJECTIVE

To develop and evaluate an automated algorithm to identify sepsis onset, referred to as time-zero (t0), from the electronic medical records (EMR).

MATERIALS AND

METHODS

We developed an algorithm to construct a comprehensive timeline of systemic inflammatory response syndrome (SIRS) criteria and organ dysfunction (OD) using structured data, and documentation of infection (DOI) using both structured data and unstructured clinical notes. Algorithm performance was assessed using 2030 manually abstracted adult sepsis cases from a multicenter health system in southeast Michigan.

RESULTS

On average, the algorithm DOI time was significantly earlier than abstractors (mean: -0.33 hour, 95% Cl, -0.55 to -0.11), resulting in a significantly earlier t0 (mean: -0.60 hour, 95% CI, -0.74 to -0.45).

DISCUSSION: Automated approaches to analyzing EMR data offer a scalable framework for SEP-1 monitoring, research, and quality improvement.

CONCLUSION

Incorporating unstructured clinical notes improves DOI detection.