Download PDFPDF

Effect of a sepsis prediction algorithm on patient mortality, length of stay and readmission: a prospective multicentre clinical outcomes evaluation of real-world patient data from US hospitals
Compose Response

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g.
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Statement of Competing Interests


  • A rapid response is a moderated but not peer reviewed online response to a published article in a BMJ journal; it will not receive a DOI and will not be indexed unless it is also republished as a Letter, Correspondence or as other content. Find out more about rapid responses.
  • We intend to post all responses which are approved by the Editor, within 14 days (BMJ Journals) or 24 hours (The BMJ), however timeframes cannot be guaranteed. Responses must comply with our requirements and should contribute substantially to the topic, but it is at our absolute discretion whether we publish a response, and we reserve the right to edit or remove responses before and after publication and also republish some or all in other BMJ publications, including third party local editions in other countries and languages
  • Our requirements are stated in our rapid response terms and conditions and must be read. These include ensuring that: i) you do not include any illustrative content including tables and graphs, ii) you do not include any information that includes specifics about any patients,iii) you do not include any original data, unless it has already been published in a peer reviewed journal and you have included a reference, iv) your response is lawful, not defamatory, original and accurate, v) you declare any competing interests, vi) you understand that your name and other personal details set out in our rapid response terms and conditions will be published with any responses we publish and vii) you understand that once a response is published, we may continue to publish your response and/or edit or remove it in the future.
  • By submitting this rapid response you are agreeing to our terms and conditions for rapid responses and understand that your personal data will be processed in accordance with those terms and our privacy notice.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

Other responses

Jump to comment:

  • Published on:
    Concern regarding the recently published sepsis prediction model
    • Christopher V. Cosgriff, Resident Physician Hospital of the University of Pennsylvania
    • Other Contributors:
      • Leo Anthony Celi, Intensive Care Physician, Research Scientist

    To the editor:

    We read the recent study from Burdick and colleagues with great interest as few machine learning models have been assessed prospectively, and even fewer have shown benefits on clinically-relevant outcomes such as length of stay or mortality.1 However, we are concerned about some of the reported data. The model they deploy, termed InSight across various publications, predicts the development of sepsis based on the systemic inflammatory response syndrome (SIRS) criteria, which has been replaced by the Sepsis-3 guidelines.2,3 Their model which is a gradient boosted classification tree built with XGBoost is trained by looking back at vital sign data and predicts if a patient will meet SIRS criteria in the next four hours. In the present study they define outcomes cohort as “sepsis-related” by including any patient who met 2/4 of the SIRS criteria. Using this definition, they report a pre-intervention in-hospital mortality of 3.86%.4 However, this is far below the expected sepsis related mortality, regardless of the criteria used. Furthermore, by reporting outcomes in patients who met SIRS >2, rather than everyone screened by the algorithm, the relative harm done to patients for whom this model falsely predicted sepsis would not be factored into the equation assessing the algorithm’s value. They report an estimate of the cost reduction, but this again does not account for the costs that could be associated with the adverse effects of unnecessarily tre...

    Show More
    Conflict of Interest:
    None declared.