Electronic Health Record-Embedded Structural Interventions to Improve Sepsis Bundle Quality Measure Performance: A Quality Improvement Initiative at a Veterans Affairs Medical Center
Abstract
Background: Though emphasized as a quality metric, sepsis quality measure (SEP-1) compliance remains low nationwide. Electronic health record (EHR)–based interventions show promise, but few have examined EHR laboratory reference value misconfiguration as a distinct driver of noncompliance. Local Problem: The San Francisco Veterans Affairs, a 110-bed tertiary care center, recorded 3 consecutive quarters of SEP-1 composite performance in the lowest national quintile, nadiring at 31.5%. Methods: A multidisciplinary team applied LEAN-A3 methodology to identify root causes of low SEP-1 performance and design targeted countermeasures. The primary outcome was the percentage of suspected sepsis patients with a serum lactate >2.0 mmol/L who received a repeat lactate within 6 hours. Balancing measures included unnecessary repeat lactates and sepsis mortality. Interventions: Three EHR-based interventions were implemented between August and November 2025: (1) an Emergency Department Sepsis order-set embedding dual time-spaced lactate orders; (2) a structured Sepsis Evaluation Note template; and (3) a clinical decision support alert for patients with serum lactate >2.0 mmol/L, implemented to address a laboratory reference range misconfiguration. Results: Postintervention lactate reassessment compliance rose 14 percentage points (54%–68%), meeting Institute for Healthcare Improvement criteria for special cause variation. Facility 12-month rolling performance rose from 31.5% (lowest quintile) to 64.7% (middle quintile), reaching a quarterly high of 74.3%. No increase in unnecessary repeat lactates or sepsis mortality was detected. Conclusions: EHR-embedded structural interventions produced sustained SEP-1 improvement without impacting laboratory utilization or mortality, offering a reproducible model for improvement through systems redesign rather than education alone.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Cognitive Load During Problem Solving: Effects on Learning
Sweller, J. (1988) · Cognitive Science · doi:10.1016/0364-0213(88)90023-7 · Cited by 11,048 · see citing works
Flagged first, then by citations
- Authors
- Zhang, J. C., Javaherian, K., Zapata, J., Duffy, R., Lu, C. M., Price, E., Snowden, H., Agarwal, A., & Mihara, K.
- Affiliations
- San Francisco VA Medical Center (US)
- Department of Medicine, University of California San Francisco, San Francisco, CA
- Duke University (US)
- Venue
- American Journal of Medical Quality
- Published
- 2026-10-06
- ISSN
- 1062-8606 (ISSN-L)
- 1555-824X
- OpenAlex
- W7220441282
- Cited by
- 0
- References
- 1
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- American Journal of Medical Quality (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Epidemiology › Sepsis Diagnosis and Treatment
- Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
- Health Sciences › Nursing › Issues, ethics and legal aspects › Nursing Diagnosis and Documentation
- Keywords
- sepsis, electronic health record interventions, clinical decision support, order sets, veterans, quality improvement, sepsis bundle compliance
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review