{ "id": "AIBD-0579", "title": "COMPOSER sepsis alerts are associated with lower mortality at UC San Diego Health", "summary": "At UC San Diego, mortality after COMPOSER deployment was 9.49%, versus a modeled 11.39% without it.", "date_observed": "2024-01-23", "date_basis": "source publication", "date_precision": "day", "benefit_status": "reported", "source_collection": null, "source_entry": null, "account": "UC San Diego Health evaluated COMPOSER in two emergency departments among 6,217 adults with sepsis. The model went live on December 7, 2022, with follow-up through April 30, 2023. Post-deployment mortality was 9.49%, compared with a modeled counterfactual of 11.39%: an adjusted reduction of 1.9 percentage points (95% confidence interval 0.3–3.5). Raw pre-deployment mortality was 10.3%. The estimated benefit was statistically significant at one hospital and not the other. The study used historical controls, without randomization or a concurrent untreated group.", "category": "general benefit", "unexpected": false, "purpose": "Identify emerging sepsis so that clinical teams can assess and treat patients sooner.", "system": "COMPOSER deep-learning sepsis prediction and nurse alert system", "model_version": null, "developer": "University of California San Diego research team", "deployer": "UC San Diego Health", "location": "San Diego, California, United States", "domain": "Healthcare - sepsis", "segment": "Electronic-health-record early warning system", "benefit_type": "Improved Physical Health", "taxonomy_note": null, "registry_ids": {}, "ratings": [ { "group": "personal", "depth": 5, "reason": "The reported benefit is lower mortality. Its magnitude is estimated against a historical model rather than a randomized control." } ], "ai_role": { "status": "assessed", "statement": "COMPOSER analyzed patient data and alerted nurses to possible sepsis. Nurses reviewed patients with physicians, who selected treatment. The outcome concerns AI-assisted clinical care." }, "evidence": [ { "kind": "peer-reviewed study", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). npj Digital Medicine 7:14. doi:10.1038/s41746-023-00986-6. Evaluation by COMPOSER's developers and deploying health system. Results; corrected Table 3; Supplementary Notes 2–8 and hospital-specific analyses.", "title": "Impact of a deep learning sepsis prediction model on quality of care and survival", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "date": "2024-01-23", "written_by": "makers", "quote": null }, { "kind": "peer-reviewed study", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). Author Correction: Impact of a deep learning sepsis prediction model on quality of care and survival. npj Digital Medicine 7:153. doi:10.1038/s41746-024-01149-x.", "title": "Author Correction: Impact of a deep learning sepsis prediction model on quality of care and survival", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "date": "2024-06-12", "written_by": "makers", "quote": null }, { "kind": "peer-reviewed study", "citation": "Supreeth P. Shashikumar et al., “Artificial Intelligence Sepsis Prediction Algorithm Learns to Say ‘I Don’t Know,’” npj Digital Medicine 4 (2021): 134, https://doi.org/10.1038/s41746-021-00504-6.", "title": "Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "date": "2021-09-09", "written_by": "makers", "quote": null } ], "replication": { "produced_by": [ "Continuous analysis of electronic patient data and alerts delivered to nursing staff.", "Nurse review, physician assessment, and treatment at UC San Diego Medical Center and Jacobs Medical Center." ], "must_hold": [ "Reliable data and prompt clinical assessment are likely requirements. Local evaluation is needed to establish whether the benefit recurs." ], "echoes": [] }, "paired_incident": null, "headline": "COMPOSER sepsis alerts are linked to lower mortality", "metric": "1.9 percentage points", "comparator": { "against": "Modeled post-deployment mortality without COMPOSER, using historical data and adjustment for confounders", "before": "11.39% modeled counterfactual mortality", "after": "9.49% observed post-deployment mortality", "basis": "assumed", "quote": "the expected counterfactual mortality rate would have been 11.39% with a 95% confidence interval of [9.79%, 13.00%]", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/" }, "later": null, "value": null, "developer_country": "United States", "deployer_country": "United States", "unrated_because": null, "brief_notes": null, "brief": null, "revisions": [ { "date": "2026-10-23", "change": "Added the case.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" }, { "date": "2026-10-24", "change": "Followed COMPOSER's outcome paper to its supplement, model-development paper and author correction. The correction restores missing Table 3 and its citation. Removed the institutional press release because the current case's results and workflow are already documented by the original study. Preserved the historical-control design and modeled counterfactual.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" }, { "date": "2026-10-25", "change": "Identified the AI model, software, configuration, and human controls from the original paper, supplement, or author documentation.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" }, { "date": "2026-10-25", "change": "Reviewed the original evidence and supporting material; updated system details, methods, outcomes, limitations, costs, and source lineage.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" } ], "confidence": "claimed", "shade": "low", "claim_status": "The makers' claim", "claim_basis": "Sources: three peer-reviewed studies from the makers. The cited evidence includes no independent source and no finding by a regulator or court.", "profile": { "case_id": "AIBD-0579", "catalog_version": "1.0.0", "source_fingerprint": "d1042baa617ae7aee7a350d7cf443da623b9049505100e1bf422360e3f4fba5c", "collection_refs": [ { "collection_id": "aibd", "version": "snapshot-8bf7cfec9cb185aa" }, { "collection_id": "aibd", "version": "2026-10-25" } ], "source_ids": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "families": { "strategic assessments": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "evaluations": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "UCSD IRB 805726" ], "role": "evaluations" }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The evaluation described in the cited original sources." }, "record_ids": [ "UCSD IRB 805726" ] }, "monetary observations": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." }, "record_ids": [] }, "institutional decisions": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "UCSD ACQUIRE 609" ], "role": "institutional decisions" }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UCSD approval and implementation of the clinical study." }, "record_ids": [ "UCSD ACQUIRE 609" ] } }, "fields": { "actors[]": { "state": "value_present", "value": { "kind": "text", "text": "UC San Diego researchers developed COMPOSER. Nurses and physicians at UC San Diego Health acted on its alerts." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UC San Diego researchers developed COMPOSER. Nurses and physicians at UC San Diego Health acted on its alerts." } }, "actors[].ownership_control": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed clinical sources identify the organizations involved, but do not establish their complete ownership or control arrangements.", "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The reviewed clinical sources identify the organizations involved, but do not establish their complete ownership or control arrangements." } }, "ai_contribution": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ai_role": { "status": "assessed", "statement": "COMPOSER analyzed patient data and alerted nurses to possible sepsis. Nurses reviewed patients with physicians, who selected treatment. The outcome concerns AI-assisted clinical care." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "ai_relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "category": "general benefit", "ai_role": { "status": "assessed", "statement": "COMPOSER analyzed patient data and alerted nurses to possible sepsis. Nurses reviewed patients with physicians, who selected treatment. The outcome concerns AI-assisted clinical care." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role", "category" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].alternatives": { "state": "value_present", "value": { "kind": "text", "text": "The grade describes the reported benefit; it does not convert the historical comparison into a causal trial." }, "provenance": { "kind": "curatorial", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The grade describes the reported benefit; it does not convert the historical comparison into a causal trial." } }, "assessments[].assessor_date": { "state": "value_present", "value": { "kind": "structured", "data": { "assessor": "AIBD, AI-assisted source review", "date": "2026-10-25" } }, "provenance": { "kind": "curatorial", "basis": "Review of the sources listed in the lineage assessment.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "AIBD, 25 October 2026." } }, "assessments[].depth": { "state": "value_present", "value": { "kind": "structured", "data": { "ratings": [ { "group": "personal", "depth": 5, "reason": "The reported benefit is lower mortality. Its magnitude is estimated against a historical model rather than a randomized control." } ] } }, "provenance": { "kind": "curatorial", "basis": "Local five-level depth rubric; Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Depth 5 for the reported survival benefit." } }, "assessments[].display_confidence": { "state": "value_present", "value": { "kind": "text", "text": "A peer-reviewed study by the developers and deploying health system. The outline is retained." }, "provenance": { "kind": "curatorial", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "A peer-reviewed study by the developers and deploying health system. The outline is retained." } }, "assessments[].rationale": { "state": "value_present", "value": { "kind": "text", "text": "Depth 5 for the reported survival benefit." }, "provenance": { "kind": "curatorial", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Depth 5 for the reported survival benefit." } }, "assessments[].rubric_version": { "state": "value_present", "value": { "kind": "text", "text": "AIBD’s five-level depth rubric, reviewed on 25 October 2026." }, "provenance": { "kind": "curatorial", "basis": "Local methodology; no separately numbered rubric version is published.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "AIBD’s five-level depth rubric, reviewed on 25 October 2026." } }, "assessments[].status_basis": { "state": "value_present", "value": { "kind": "text", "text": "The depth describes the reported result. Evidence strength is assessed separately." }, "provenance": { "kind": "curatorial", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The depth describes the reported result. Evidence strength is assessed separately." } }, "assessments[].target": { "state": "value_present", "value": { "kind": "text", "text": "The benefit described here, for the beneficiary groups shown." }, "provenance": { "kind": "curatorial", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The benefit described here, for the beneficiary groups shown." } }, "availability": { "state": "value_present", "value": { "kind": "text", "text": "The paper describes clinical use at UC San Diego and a startup formed by several authors. It does not establish general public access to the deployed system." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The paper describes clinical use at UC San Diego and a startup formed by several authors. It does not establish general public access to the deployed system." } }, "brief": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "case_kind": { "state": "value_present", "value": { "kind": "text", "text": "A prospective comparison before and after clinical deployment." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "A prospective comparison before and after clinical deployment." } }, "case_links[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].evidence_links": { "state": "value_present", "value": { "kind": "structured", "data": { "clinical_results": { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "locator": "Results; corrected Table 3" }, "correction": { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "locator": "Correction notice" }, "model": { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "locator": "Original model methods" } } }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The deployment paper supplies clinical outcomes. The 2021 paper describes the model, and the correction restores an omitted table." } }, "claims[].evidence_strength": { "state": "value_present", "value": { "kind": "text", "text": "An original, peer-reviewed study by the developers and deploying health system, using a historical counterfactual." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "An original, peer-reviewed study by the developers and deploying health system, using a historical counterfactual." } }, "claims[].missingness": { "state": "value_present", "value": { "kind": "text", "text": "The study did not measure harms from false-positive alerts in patients without sepsis, financial returns, or an independently replicated mortality effect." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The study did not measure harms from false-positive alerts in patients without sepsis, financial returns, or an independently replicated mortality effect." } }, "claims[].proposition": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER deployment was associated with lower sepsis mortality and more frequent completion of the sepsis care bundle." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "COMPOSER deployment was associated with lower sepsis mortality and more frequent completion of the sepsis care bundle." } }, "cross_border_flows[]": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not document international transfers of patient data or quantify cross-border financial flows.", "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The reviewed sources do not document international transfers of patient data or quantify cross-border financial flows." } }, "decisions[].action_outcome": { "state": "value_present", "value": { "kind": "text", "text": "The team trained staff, deployed the alerts, and monitored the system during clinical use." }, "provenance": { "kind": "source_review", "basis": "Methods: Study design; deployment and monitoring.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The team trained staff, deployed the alerts, and monitored the system during clinical use." } }, "decisions[].grounds": { "state": "value_present", "value": { "kind": "text", "text": "The project received IRB approval 805726 and ACQUIRE approval 609, with a waiver of informed consent." }, "provenance": { "kind": "source_review", "basis": "Methods: Study design; deployment and monitoring.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The project received IRB approval 805726 and ACQUIRE approval 609, with a waiver of informed consent." } }, "decisions[].institution_task": { "state": "value_present", "value": { "kind": "text", "text": "UC San Diego Health introduced a nurse alert for possible sepsis." }, "provenance": { "kind": "source_review", "basis": "Methods: Study design; deployment and monitoring.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UC San Diego Health introduced a nurse alert for possible sepsis." } }, "decisions[].success_criterion": { "state": "value_present", "value": { "kind": "text", "text": "The evaluation examined mortality, sepsis treatment, and organ dysfunction rather than a single formal pass-or-fail threshold." }, "provenance": { "kind": "source_review", "basis": "Methods: Study design; deployment and monitoring.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The evaluation examined mortality, sepsis treatment, and organ dysfunction rather than a single formal pass-or-fail threshold." } }, "deployment_stage": { "state": "value_present", "value": { "kind": "text", "text": "Routine use at two UC San Diego emergency departments from 7 December 2022." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Routine use at two UC San Diego emergency departments from 7 December 2022." } }, "economics[].amount": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].costs[]": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The paper reports about six months of preparation, staff training, cloud operation, and monitoring, but no monetary cost or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The paper reports about six months of preparation, staff training, cloud operation, and monitoring, but no monetary cost or financial return." } }, "economics[].counterfactual": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].currency_price_basis": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].gross_net": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].measure_type": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].period_unit": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].realization": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].recipient_payer": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].return_payback": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "economics[].valuation_method": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report a monetary valuation or financial return.", "review_state": "reviewed", "coverage": "complete", "family": "monetary observations", "display": { "answer": "The reviewed sources do not report a monetary valuation or financial return." } }, "evaluations[].ai_uplift": { "state": "value_present", "value": { "kind": "text", "text": "The estimate concerns COMPOSER, nurse alerts, preparation, and subsequent care together. It does not isolate the prediction model from those changes." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The estimate concerns COMPOSER, nurse alerts, preparation, and subsequent care together. It does not isolate the prediction model from those changes." } }, "evaluations[].comparator": { "state": "value_present", "value": { "kind": "text", "text": "A modeled post-deployment counterfactual based on the historical period, with adjustment for measured differences." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "A modeled post-deployment counterfactual based on the historical period, with adjustment for measured differences." } }, "evaluations[].design": { "state": "value_present", "value": { "kind": "text", "text": "A prospective, nonrandomized before-and-after study. A Bayesian time-series model estimated what would have happened without deployment." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "A prospective, nonrandomized before-and-after study. A Bayesian time-series model estimated what would have happened without deployment." } }, "evaluations[].effect": { "state": "value_present", "value": { "kind": "structured", "data": { "observed_mortality_percent": 9.49, "modeled_mortality_percent": 11.39, "modeled_95CI": [ 9.79, 13.0 ], "absolute_reduction_percentage_points": 1.9, "absolute_reduction_95CI": [ 0.3, 3.5 ], "relative_reduction_percent": 17, "bundle_compliance_percent": { "observed": 53.42, "modeled": 48.38 }, "bundle_difference_percentage_points": 5.0, "bundle_difference_95CI": [ 2.4, 8.0 ] } }, "provenance": { "kind": "source_review", "basis": "Results; corrected Table 3.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Mortality was 9.49%, versus 11.39% in the modeled comparison: a difference of 1.9 percentage points (95% CI, 0.3–3.5)." } }, "evaluations[].exposure": { "state": "value_present", "value": { "kind": "text", "text": "The historical period ran from 1 January 2021 to 6 December 2022. Deployment began on 7 December 2022." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The historical period ran from 1 January 2021 to 6 December 2022. Deployment began on 7 December 2022." } }, "evaluations[].id": { "state": "value_present", "value": { "kind": "text", "text": "UCSD IRB 805726" }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UCSD IRB 805726" } }, "evaluations[].measurement_method": { "state": "value_present", "value": { "kind": "text", "text": "Researchers used electronic records to identify Sepsis-3 cases, inpatient deaths, organ dysfunction, and completion of the sepsis care bundle." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Researchers used electronic records to identify Sepsis-3 cases, inpatient deaths, organ dysfunction, and completion of the sepsis care bundle." } }, "evaluations[].metric": { "state": "value_present", "value": { "kind": "text", "text": "In-hospital mortality, completion of the sepsis care bundle, and change in organ dysfunction over 72 hours." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "In-hospital mortality, completion of the sepsis care bundle, and change in organ dysfunction over 72 hours." } }, "evaluations[].reproduction_materials": { "state": "value_present", "value": { "kind": "text", "text": "The supplement describes implementation and statistical methods. De-identified data and analysis code are available only through the authors and institutional approval." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The supplement describes implementation and statistical methods. De-identified data and analysis code are available only through the authors and institutional approval." } }, "evaluations[].setting_sample": { "state": "value_present", "value": { "kind": "text", "text": "The analysis included 6,217 eligible sepsis encounters: 5,065 before deployment and 1,152 afterward." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The analysis included 6,217 eligible sepsis encounters: 5,065 before deployment and 1,152 afterward." } }, "evaluations[].uncertainty_limitations": { "state": "value_present", "value": { "kind": "text", "text": "There was no randomized or concurrent untreated group. Unmeasured changes in care could affect the comparison. Results covered five months at one health system." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "There was no randomized or concurrent untreated group. Unmeasured changes in care could affect the comparison. Results covered five months at one health system." } }, "external_references[]": { "state": "value_present", "value": { "kind": "structured", "data": { "ethics_protocol": "UCSD IRB 805726", "institutional_project": "ACQUIRE 609" } }, "provenance": { "kind": "source_review", "basis": "Methods: Study design.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UCSD IRB 805726 and ACQUIRE project 609." } }, "follow_up[]": { "state": "value_present", "value": { "kind": "text", "text": "The June 2024 correction restored Table 3 and its citation. It did not announce a new trial or change the reported mortality estimate." }, "provenance": { "kind": "source_review", "basis": "Author Correction, 12 June 2024.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0002" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The June 2024 correction restored Table 3 and its citation. It did not announce a new trial or change the reported mortality estimate." } }, "human_oversight": { "state": "value_present", "value": { "kind": "text", "text": "Nurses reviewed an alert and chose whether to notify a physician, note that care was already underway, or record that infection was unlikely. Physicians decided on treatment." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Nurses reviewed an alert and chose whether to notify a physician, note that care was already underway, or record that infection was unlikely. Physicians decided on treatment." } }, "id": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "id": "AIBD-0579" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "id" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "infrastructure_dependencies[]": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER ran on AWS with electronic-record feeds and an Epic nurse alert. The deployment also required clinical staffing and monitoring of data delivery." }, "provenance": { "kind": "source_review", "basis": "Methods: AI model deployment; Supplementary Notes 2–3.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "COMPOSER ran on AWS with electronic-record feeds and an Epic nurse alert. The deployment also required clinical staffing and monitoring of data delivery." } }, "outcomes[].beneficiaries": { "state": "value_present", "value": { "kind": "text", "text": "Adults with sepsis in the Hillcrest and La Jolla emergency departments." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Adults with sepsis in the Hillcrest and La Jolla emergency departments." } }, "outcomes[].beneficiary_count": { "state": "value_present", "value": { "kind": "structured", "data": { "Sepsis3_encounters": 6340, "comfort_care_exclusions": 123, "analyzed_encounters": 6217, "pre": 5065, "post": 1152, "estimated_additional_survivors": 22, "estimated_not_observed_individuals": true } }, "provenance": { "kind": "source_review", "basis": "Results: Study cohort and Mortality; Methods: study population and Causal Impact analysis.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The study analyzed 6,217 sepsis encounters. Its estimate of 22 additional survivors comes from the counterfactual model." } }, "outcomes[].benefit_type": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "benefit_type": "Improved Physical Health" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "benefit_type" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].claim": { "state": "value_present", "value": { "kind": "text", "text": "Sepsis mortality after deployment was lower than the study’s modeled estimate for care without COMPOSER." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Sepsis mortality after deployment was lower than the study’s modeled estimate for care without COMPOSER." } }, "outcomes[].distribution": { "state": "value_present", "value": { "kind": "text", "text": "The mortality difference was statistically significant at Hillcrest but not at La Jolla. The two hospitals served different patient populations." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The mortality difference was statistically significant at Hillcrest but not at La Jolla. The two hospitals served different patient populations." } }, "outcomes[].duration": { "state": "value_present", "value": { "kind": "text", "text": "The post-deployment evaluation ran from 7 December 2022 through 30 April 2023. Mortality was measured during hospitalization." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The post-deployment evaluation ran from 7 December 2022 through 30 April 2023. Mortality was measured during hospitalization." } }, "outcomes[].id": { "state": "value_present", "value": { "kind": "text", "text": "AIBD-0579-outcome-01" }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "AIBD-0579-outcome-01" } }, "outcomes[].stakeholder_group": { "state": "value_present", "value": { "kind": "structured", "data": { "groups": [ "personal" ] } }, "provenance": { "kind": "curatorial", "basis": "The beneficiary groups correspond to the reviewed depth assessments.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "People." } }, "outcomes[].status": { "state": "value_present", "value": { "kind": "text", "text": "Observed outcomes compared with a modeled counterfactual." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Observed outcomes compared with a modeled counterfactual." } }, "outcomes[].tradeoffs": { "state": "value_present", "value": { "kind": "text", "text": "Nurses received about 235 alerts per month across both sites. Unnecessary treatment among patients without sepsis, and its costs or harms, were not evaluated." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Nurses received about 235 alerts per month across both sites. Unnecessary treatment among patients without sepsis, and its costs or harms, were not evaluated." } }, "outcomes[].unexpected": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "unexpected": false, "purpose": "Identify emerging sepsis so that clinical teams can assess and treat patients sooner." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "unexpected", "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "places[]": { "state": "value_present", "value": { "kind": "text", "text": "UC San Diego’s Hillcrest and La Jolla hospitals, California, United States." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "UC San Diego’s Hillcrest and La Jolla hospitals, California, United States." } }, "purpose": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "purpose": "Identify emerging sepsis so that clinical teams can assess and treat patients sooner." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "record_dates": { "state": "value_present", "value": { "kind": "structured", "data": { "historical_start": "2021-01-01", "deployment": "2022-12-07", "evaluation_end": "2023-04-30", "publication": "2024-01-23", "correction": "2024-06-12" } }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Deployment began on 7 December 2022. Follow-up ended on 30 April 2023." } }, "replication.attempts[]": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "reason": "The reviewed sources do not report an independent repetition of this clinical result.", "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The reviewed sources do not report an independent repetition of this clinical result." } }, "replication.conditions": { "state": "value_present", "value": { "kind": "text", "text": "The deployment required reliable electronic records, nurse training, physician response, and continuing checks of model performance and data delivery." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The deployment required reliable electronic records, nurse training, physician response, and continuing checks of model performance and data delivery." } }, "revision_history[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "revisions": [ { "date": "2026-10-23", "change": "Added the case.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" }, { "date": "2026-10-24", "change": "Followed COMPOSER's outcome paper to its supplement, model-development paper and author correction. The correction restores missing Table 3 and its citation. Removed the institutional press release because the current case's results and workflow are already documented by the original study. Preserved the historical-control design and modeled counterfactual.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" }, { "date": "2026-10-25", "change": "Identified the AI model, software, configuration, and human controls from the original paper, supplement, or author documentation.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "editor": "AIBD" } ] } }, "provenance": { "kind": "source_review", "basis": "Updated only the mapped fields affected by this source review. Other mapped content and review decisions are preserved.", "record_fields": [ "revisions" ], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002" ], "reviewed_on": "2026-10-24" }, "review_state": "reviewed", "coverage": "partial" }, "sector": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "domain": "Healthcare - sepsis", "segment": "Electronic-health-record early warning system" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "domain", "segment" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].access_version": { "state": "value_present", "value": { "kind": "structured", "data": { "reviewed_on": "2026-10-25", "versions_and_access": [ { "source_ref": "AIBD-0579-source-0001", "source_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "access_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "locator": "Results; corrected Table 3; Methods; Data and Code availability; Competing interests.", "status": "Original text read.", "reviewed_on": "2026-10-25" }, { "source_ref": "AIBD-0579-source-0001", "source_url": "https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41746-023-00986-6/MediaObjects/41746_2023_986_MOESM1_ESM.docx", "access_url": "https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41746-023-00986-6/MediaObjects/41746_2023_986_MOESM1_ESM.docx", "locator": "Supplementary Notes 2–8.", "status": "Original text read.", "reviewed_on": "2026-10-25" }, { "source_ref": "AIBD-0579-source-0002", "source_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "access_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "locator": "Entire correction notice.", "status": "Original text read.", "reviewed_on": "2026-10-25" }, { "source_ref": "AIBD-0579-source-0003", "source_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "access_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "locator": "Methods: conformal prediction and neural model.", "status": "Original text read.", "reviewed_on": "2026-10-25" } ] } }, "provenance": { "kind": "source_review", "basis": "Access records below distinguish full text, supplements, metadata, and inaccessible material.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Access and version details are listed for each reviewed source." } }, "sources[].citation_locator": { "state": "value_present", "value": { "kind": "structured", "data": { "citations": [ { "source_ref": "AIBD-0579-source-0001", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). npj Digital Medicine 7:14. doi:10.1038/s41746-023-00986-6. Evaluation by COMPOSER's developers and deploying health system. Results; corrected Table 3; Supplementary Notes 2–8 and hospital-specific analyses.", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "date": "2024-01-23" }, { "source_ref": "AIBD-0579-source-0002", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). Author Correction: Impact of a deep learning sepsis prediction model on quality of care and survival. npj Digital Medicine 7:153. doi:10.1038/s41746-024-01149-x.", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "date": "2024-06-12" }, { "source_ref": "AIBD-0579-source-0003", "citation": "Supreeth P. Shashikumar et al., “Artificial Intelligence Sepsis Prediction Algorithm Learns to Say ‘I Don’t Know,’” npj Digital Medicine 4 (2021): 134, https://doi.org/10.1038/s41746-021-00504-6.", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "date": "2021-09-09" } ], "locators": [ { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "locator": "Results; corrected Table 3; Methods; Data and Code availability; Competing interests.", "contribution": "Clinical comparison, counts, oversight, implementation, and study limitations.", "relationship": "Original evaluation by the developers and deploying health system.", "source_refs": [ "AIBD-0579-source-0001" ] }, { "url": "https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41746-023-00986-6/MediaObjects/41746_2023_986_MOESM1_ESM.docx", "locator": "Supplementary Notes 2–8.", "contribution": "Implementation, alerts, monitoring, causal model, and hospital-specific results.", "relationship": "Supplement to the same evaluation.", "source_refs": [ "AIBD-0579-source-0001" ] }, { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "locator": "Entire correction notice.", "contribution": "Restores Table 3 and its missing citation.", "relationship": "Correction to the same paper, not replication.", "source_refs": [ "AIBD-0579-source-0002" ] }, { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "locator": "Methods: conformal prediction and neural model.", "contribution": "Model construction and ability to abstain on unfamiliar inputs.", "relationship": "Earlier development study, not independent verification of the clinical effect.", "source_refs": [ "AIBD-0579-source-0003" ] } ] } }, "provenance": { "kind": "source_review", "basis": "Reviewed source metadata and contribution-specific locators.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "Original citations and the passages that support each finding." } }, "sources[].id": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "role": "Retained sources" }, "provenance": { "kind": "curatorial", "basis": "Identifiers correspond to the retained source order.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "3 retained sources." } }, "sources[].language": { "state": "value_present", "value": { "kind": "text", "text": "English. No translation was used." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "English. No translation was used." } }, "sources[].relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "evidence": [ { "kind": "peer-reviewed study", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). npj Digital Medicine 7:14. doi:10.1038/s41746-023-00986-6. Evaluation by COMPOSER's developers and deploying health system. Results; corrected Table 3; Supplementary Notes 2–8 and hospital-specific analyses.", "title": "Impact of a deep learning sepsis prediction model on quality of care and survival", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "date": "2024-01-23", "written_by": "makers" }, { "kind": "peer-reviewed study", "citation": "Boussina A, Shashikumar SP, Malhotra A, et al. (2024). Author Correction: Impact of a deep learning sepsis prediction model on quality of care and survival. npj Digital Medicine 7:153. doi:10.1038/s41746-024-01149-x.", "title": "Author Correction: Impact of a deep learning sepsis prediction model on quality of care and survival", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "date": "2024-06-12", "written_by": "makers" }, { "kind": "peer-reviewed study", "citation": "Supreeth P. Shashikumar et al., “Artificial Intelligence Sepsis Prediction Algorithm Learns to Say ‘I Don’t Know,’” npj Digital Medicine 4 (2021): 134, https://doi.org/10.1038/s41746-021-00504-6.", "title": "Artificial intelligence sepsis prediction algorithm learns to say “I don’t know”", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "date": "2021-09-09", "written_by": "makers" } ] } }, "provenance": { "kind": "source_review", "basis": "Updated only the mapped fields affected by this source review. Other mapped content and review decisions are preserved.", "record_fields": [ "evidence" ], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002" ], "reviewed_on": "2026-10-24" }, "review_state": "reviewed", "coverage": "partial" }, "sources[].upstream_sources": { "state": "value_present", "value": { "kind": "structured", "data": { "reviewed_on": "2026-10-25", "lead_source_ref": "AIBD-0579-source-0001", "retained_source_order": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "origins": [ { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10805720/", "locator": "Results; corrected Table 3; Methods; Data and Code availability; Competing interests.", "contribution": "Clinical comparison, counts, oversight, implementation, and study limitations.", "relationship": "Original evaluation by the developers and deploying health system.", "source_refs": [ "AIBD-0579-source-0001" ] }, { "url": "https://media.springernature.com/original/springer-static/esm/art%3A10.1038%2Fs41746-023-00986-6/MediaObjects/41746_2023_986_MOESM1_ESM.docx", "locator": "Supplementary Notes 2–8.", "contribution": "Implementation, alerts, monitoring, causal model, and hospital-specific results.", "relationship": "Supplement to the same evaluation.", "source_refs": [ "AIBD-0579-source-0001" ] }, { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11169238/", "locator": "Entire correction notice.", "contribution": "Restores Table 3 and its missing citation.", "relationship": "Correction to the same paper, not replication.", "source_refs": [ "AIBD-0579-source-0002" ] }, { "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8429719/", "locator": "Methods: conformal prediction and neural model.", "contribution": "Model construction and ability to abstain on unfamiliar inputs.", "relationship": "Earlier development study, not independent verification of the clinical effect.", "source_refs": [ "AIBD-0579-source-0003" ] } ], "excluded_or_consolidated_sources": [], "review_scope": "Methods, results, counts, chronology, AI identity and operation, human responsibilities, limitations, and economic evidence." } }, "provenance": { "kind": "source_review", "basis": "The original materials and exact locators are listed in this assessment.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0002", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The original studies and their supporting materials are linked to the findings that they support." } }, "strategic_assessments[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "strategic assessments" }, "summary": { "state": "value_present", "value": { "kind": "text", "text": "At UC San Diego, mortality after COMPOSER deployment was 9.49%, versus a modeled 11.39% without it." }, "provenance": { "kind": "source_review", "basis": "Boussina et al. (2024), corrected Results and Table 3; Methods; Supplementary Notes 2–8.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "At UC San Diego, mortality after COMPOSER deployment was 9.49%, versus a modeled 11.39% without it." } }, "systems[].autonomy_permissions": { "state": "value_present", "value": { "kind": "text", "text": "Nurses acknowledged the alert and could notify a physician. Clinicians decided on tests and treatment." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "systems[].component_role": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER estimated sepsis risk over the next four hours and reported the measurements that contributed to each alert." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "systems[].deployment_configuration": { "state": "value_present", "value": { "kind": "structured", "data": { "prediction_horizon_hours": 4, "data_refresh": "hourly", "sensitivity_target_percent": 80, "alert_lockout_hours": { "no_infection": 8, "notify_or_workup_underway": 12 }, "performance_review": "every two weeks", "retraining_trigger": "positive predictive value below 20%; no retraining reported by study end", "hosting": "AWS", "interface": "Epic Best Practice Advisory to nurses" } }, "provenance": { "kind": "source_review", "basis": "Methods: model integration and clinical workflow; Supplementary Notes 2–3.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "complete", "display": { "answer": "The system assessed risk over the next four hours and issued nurse alerts in Epic. Repeat alerts were suppressed for eight or twelve hours after a response." } }, "systems[].inputs_outputs": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER analyzed clinical measurements and their timing to estimate sepsis risk and identify the main contributing factors." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "systems[].model_name": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER, a feedforward neural network with conformal prediction." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "systems[].model_version": { "state": "not_reported", "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "reason": "The deployment report does not give a release or checkpoint identifier.", "review_state": "reviewed", "coverage": "partial", "display": { "answer": "The deployment report does not give a release or checkpoint identifier." } }, "systems[].product_name": { "state": "value_present", "value": { "kind": "text", "text": "COMPOSER with a nurse-facing alert in Epic." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "systems[].provider": { "state": "value_present", "value": { "kind": "text", "text": "University of California San Diego." }, "provenance": { "kind": "source_review", "basis": "Read the corrected deployment paper, “Sepsis algorithm and platform”, and supplement, Note 2 and Figure 1. Followed reference 15 to the original COMPOSER paper and read its Model description and Figure 1.", "record_fields": [], "source_refs": [ "AIBD-0579-source-0001", "AIBD-0579-source-0003" ], "reviewed_on": "2026-10-25" }, "review_state": "reviewed", "coverage": "partial" }, "taxonomy_crosswalks[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "title": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "title": "COMPOSER sepsis alerts are associated with lower mortality at UC San Diego Health", "headline": "COMPOSER sepsis alerts are linked to lower mortality" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "title", "headline" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "version": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "d1042baa617ae7aee7a350d7cf443da623b9049505100e1bf422360e3f4fba5c" ], "role": "SHA-256 of normalized original case data" }, "provenance": { "kind": "curatorial", "basis": "Content fingerprint refreshed after the October 25, 2026 source review.", "record_fields": [], "source_refs": [], "reviewed_on": "2026-10-25" }, "review_state": "record_mapped", "coverage": "complete" }, "workflow": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "account": "UC San Diego Health evaluated COMPOSER in two emergency departments among 6,217 adults with sepsis. The model went live on December 7, 2022, with follow-up through April 30, 2023. Post-deployment mortality was 9.49%, compared with a modeled counterfactual of 11.39%: an adjusted reduction of 1.9 percentage points (95% confidence interval 0.3–3.5). Raw pre-deployment mortality was 10.3%. The estimated benefit was statistically significant at one hospital and not the other. The study used historical controls, without randomization or a concurrent untreated group.", "replication": { "produced_by": [ "Continuous analysis of electronic patient data and alerts delivered to nursing staff.", "Nurse review, physician assessment, and treatment at UC San Diego Medical Center and Jacobs Medical Center." ], "must_hold": [ "Reliable data and prompt clinical assessment are likely requirements. Local evaluation is needed to establish whether the benefit recurs." ], "echoes": [] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "account", "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" } }, "summary": { "total": 91, "addressed": 91, "known": 72, "reviewed": 77, "record_mapped": 10, "not_assessed": 4, "complete": false, "by_state": { "value_present": 72, "not_reported": 15, "unresolved": 0, "not_assessed": 4, "not_applicable": 0, "withheld": 0, "source_unavailable": 0 } } } }