←
Back
◀
Page
/
-
▶
-
Fit
+
Fit
Crying wolf: the growing fatigue around sepsis alerts
The Lancet Respiratory Medicine (Editorial)
▲
▼
H
C
Annotations
KB
No annotations yet. Use the highlight or comment tools to add some.
Related Articles
Alert Fatigue in Automated Clinical Decision Support
Notes
## Bibliographic Reference The Lancet Respiratory Medicine (2018) 'Crying wolf: the growing fatigue around sepsis alerts', *The Lancet Respiratory Medicine*, 6(3), p. 161. doi: 10.1016/S2213-2600(18)30072-9. `lancet2018cryingwolf` ## BibTeX ```bibtex @article{lancet2018cryingwolf, author = {{The Lancet Respiratory Medicine}}, title = {Crying wolf: the growing fatigue around sepsis alerts}, journal = {The Lancet Respiratory Medicine}, volume = {6}, number = {3}, pages = {161}, year = {2018}, doi = {10.1016/S2213-2600(18)30072-9} } ``` ## Routing - **Primary KB**: artificial-intelligence-kb (decision support — AI-assisted decision making and trust calibration; safety — failure modes of deployed automated decision support). **Note: borderline.** The source is a clinical/medical editorial; it is held here for its transferable content — the failure modes of an automated, algorithmic decision-support system in deployment (over-triggering, alert fatigue, poor positive predictive value, loss of trust) — not for its medical subject matter. It is a real-world case study of the cry-wolf / automation-disuse problem rather than a machine-learning methods source. Flagged for the user to reroute or drop if judged out of scope. - **Cross-links**: remote-operations-kb (the cry-wolf / multiple-alarm study `mcdonald1996spatial`, alarm management); human-centered-design-kb (decision-support and trust). Linked, not duplicated. - **Compiles to**: [[alert-fatigue-clinical-decision-support]] ## Notes One-page editorial (≤3 pp., so no page locators). Reports the 2018 NHS England admission that primary-care sepsis alerts were being over-triggered, the resulting alert fatigue (clinicians desensitised, alerts ignored/disabled, real cases missed), and a 2015 systematic review finding automated electronic sepsis alerts have poor positive predictive value and do not reduce mortality or length of stay. Concludes alerts must balance false vs. valid alarms and preserve their intended urgency. The title makes explicit the link to the cry-wolf / false-alarm effect and to Bainbridge's [[ironies-of-automation]].
Save Notes