Delirium AI model raises detection at second hospital; CMS proposes cutting hospital training rules; Medicare adds hospitalist quality pathway

• AI Nov 2, 2026
  Delirium machine learning model raises detection rate to 19.59% from 9.23% after deployment at a second Mount Sinai hospital
  A machine learning delirium risk model was deployed without retraining at Mount Sinai Morningside and compared with Mount Sinai Hospital across 33,095 non-intensive care admissions of patients 60 or older. The area under the receiver operating characteristic curve was 0.91 at Morningside and 0.94 at Mount Sinai Hospital, and Morningside's mean monthly delirium detection rate rose from 9.23% to 19.59% after deployment. Positive predictive value was 25.9% at Morningside, and the authors reported a high false positive rate, according to the JAMA Network Open study published Nov. 2.
  Sources: JAMA Network Open, PubMed
• Policy Nov 5, 2026
  CMS proposes cutting hospital training rules for emergency readiness and obstetric care, estimating $320 million in savings
  The Centers for Medicare & Medicaid Services (CMS) proposed on Nov. 5 to remove hospital staff training requirements for transfer protocols and emergency services readiness, and for obstetric care that would otherwise take effect Jan. 1, 2027, under its Cutting Administrative Requirements for Excellence in Patient Care proposal. CMS estimated the changes would save $320 million and return more than 3.26 million hours of staff time to providers.
  Source: CMS
• Payment Nov 4, 2026
  Medicare 2027 fee schedule adds a hospitalist quality pathway and cuts the conversion factor 2.31% to $32.63
  The 2027 physician fee schedule final rule from the Centers for Medicare & Medicaid Services (CMS) adds three Merit-based Incentive Payment System value pathways, focused on diabetes, hypertension and hospitalists, according to Medical Economics. The conversion factor falls 2.31% to $32.63 for clinicians outside qualifying advanced payment models. Medical Economics said the cut is deeper than the 1.68% proposed and that the gap sits in an adjustment for changes to work relative value units.
  Sources: CMS, Medical Economics
• Payment Nov 4, 2026
  CMS removes 637 more services from inpatient-only list in final 2027 hospital outpatient payment rule
  The Centers for Medicare & Medicaid Services (CMS) finalized removal of 637 more services from the inpatient-only list in its 2027 hospital outpatient payment rule, released Nov. 4, according to the American Hospital Association (AHA). CMS said the rule continues to phase out the list, which will allow more services to be delivered in outpatient settings, and expands the ambulatory surgical center covered procedures list by several hundred surgical procedures. Outpatient payment rates rise a net 2.3%, a 3.2% market basket update offset by a 0.9 percentage point productivity cut, the AHA said.
  Sources: CMS, American Hospital Association
• Research Nov 3, 2026
  Unplanned readmission within 90 days reaches 27.5% among 99,778 older veterans discharged home from 120 VA hospitals
  In a retrospective cohort of 99,778 community-dwelling patients 65 or older discharged home from 120 Department of Veterans Affairs hospitals, unplanned readmissions occurred in 14.4% at 30 days and 27.5% at 90 days. Mortality was 6.1% at 30 days among those with an unplanned readmission, and half of the 30-day deaths followed one. The authors proposed days in acute care within 90 days as a transitional care outcome, according to the Journal of Hospital Medicine report published Nov. 3.
  Sources: Journal of Hospital Medicine, PubMed
• Research Nov 6, 2026
  Later antibiotics do not change escalation of care for 758 children hospitalized with pneumonia, 3% versus 2%
  In a retrospective cohort of 758 children hospitalized with community-acquired pneumonia at four hospitals, escalation of care occurred in 3% of those given antibiotics within four hours of emergency department arrival versus 2% of those given them later (odds ratio 1.46, 95% CI 0.57 to 3.73), a difference that was not statistically significant. The authors said more time to observe children without antibiotics does not change outcomes, according to the Journal of Hospital Medicine report published Nov. 6.
  Sources: Journal of Hospital Medicine, PubMed

The weekly innovation roundup for hospital medicine, published every Sunday for the week before; every item links to its sources. Latest hospital medicine news · All specialties