
Share
By pairing its curated medical literature with full patient chart data from its Wellsheet acquisition, Elsevier is wagering that clinical decision support tools win adoption through evidence traceability, not conversational polish.
Elsevier just closed a gap that's been nagging clinical decision support tools for years: the disconnect between "here's what the literature says" and "here's what's actually going on with this specific patient." The company's upgraded ClinicalKey AI now pulls a patient's full electronic health record into the loop, letting clinicians check real-time treatment plans against Elsevier's evidence-based library instead of typing symptoms into a search box and hoping the match is close enough.
That's a meaningful architectural shift for a tool that's already live at more than 1,200 hospitals.
ClinicalKey AI started as a HIPAA-ready, point-of-care search layer over roughly 1,000 peer-reviewed journals. Ask it a clinical question, get a cited answer pulled from curated content rather than the open web. Useful, but limited: the system only knew what a clinician typed in, which meant complex, multi-condition patients got generic answers unless someone manually fed in the relevant history.
The fix came through acquisition. In June, Elsevier bought Wellsheet, a health tech company with a decade of experience aggregating and synthesizing patient chart data, notes, labs, imaging, medication orders, across EHR systems. Bolting that onto ClinicalKey AI means the tool can now ingest the entirety of a patient's chart and reason over it alongside the literature, rather than treating patient context as an afterthought.
"The problem we're trying to solve here is that physicians are challenged by so many things when they're taking care of patients; top among them is patient complexity," Dr. Claudine Lott, physician executive for clinical solutions at Elsevier, told Fierce Healthcare at HLTH this week. Patients are stacking more conditions and more medications, she said, and "it can be hard to really understand the full picture and determine how to proceed."
Elsevier isn't shy about where it thinks its moat is. In a crowded field of AI-powered clinical decision support products, the company is leaning on two things competitors can't easily replicate: a wholly owned, proprietary library of peer-reviewed scientific, technical, and medical literature spanning millions of articles, journals, and books, and now a decade of patient-chart analysis experience via Wellsheet. The company recently added full-text content and visuals from the JAMA Network Family of Journals to the mix, and it's been stacking specialized tools around the core product, including ClinicalKey Nursing AI for nursing workflows and LeapSpace, a research assistant built on more than 15 million peer-reviewed articles and 100 million Scopus records.
Three months before this upgrade, Elsevier rolled out an "Ask AI" conversational feature across 1,500+ U.S. hospital sites on its flagship ClinicalKey platform. The company is explicit that Ask AI doesn't touch the open web and doesn't store or retain patient-identifiable information, a distinction that matters a lot when you're selling into health systems with compliance teams breathing down every vendor's neck.
San Juan Regional Medical Center in Farmington, New Mexico, is one of the earlier adopters of the patient-context functionality, and its experience is a useful stress test for whether this actually works outside a flagship academic medical center.
San Juan is a nonprofit, community-owned Level III trauma center serving the Four Corners area, with smaller clinical and quality teams than big academic systems and a state where nearly 40% of residents rely on Medicaid. Resource constraints aren't theoretical there.

Carlo Hallak, M.D., the hospital's physician executive for information services, told Fierce Healthcare that Elsevier's positioning as a primary knowledge source, not a "wrapper" sitting on top of someone else's model, was a key factor in adoption. He also emphasized that ClinicalKey AI is built for the entire care team. "The hospital is run by nurses, mid-level providers, other caregiver teams," he said.
The results he's citing are concrete:
Physicians are also using the tool during weekly complex case discussions to pull trending medications, imaging, and test results into one summarized view instead of hunting across systems.
Worth noting: San Juan is also working with neighboring health systems on interoperability, since frequent ED patients tend to bounce between facilities and no single chart tells the whole story. ClinicalKey AI helps inside San Juan's walls, but the bigger fragmented-data problem across organizations is a separate, harder fight.
On governance, San Juan is taking a risk-based approach: source verification and peer review for AI-generated summaries, with tighter scrutiny reserved for anything predictive or suggestive rather than purely summarative. That review loop pulls in clinical experts alongside compliance, legal, and cybersecurity teams, on top of standard vendor security vetting. Hallak's view is that oversight needs to scale up as these tools move from summarizing information to recommending or taking action, which tracks with how most health systems are thinking about AI risk tiers right now.
Elsevier designed the tool with that scrutiny in mind: citations tracing every claim back to source material, and explicit guardrails against the tool prescribing courses of action. It surfaces evidence; the clinician still decides.
Tags
Original Sources
HLTH 2026: Elsevier upgrades ClinicalKey AI with full patient EHR context
↗ https://www.fiercehealthcare.com/health-tech/hlth-2026-elsevier-upgrades-clinicalkey-ai-full-patient-ehr-context
About the author
Kai built ML infrastructure at a Bay Area startup before developing an obsession with transformer architectures and inference optimisation that eventually pulled him out of product work entirely. A stint at a compute research lab sharpened his instinct for what actually matters in a model release versus what is marketing. He writes from the inside — from the perspective of someone who has debugged the systems he is describing at three in the morning. He is allergic to hype and instinctively drawn to the unglamorous plumbing questions that everyone else skips over.
More from The Engineer →This Week's Edition
20 November 2026
29 articles
Related Articles

BMO Claims Quantum Utility Milestone in AI Feature Selection, Using 100x Less Data
Products & Applications · 5 min

National Academies Report Calls for Faster, Tougher Retraction Standards
Models & Research · 5 min

Who Decides If Your Treatment Is Medically Necessary? Industry Leaders Say It Shouldn't Be AI Alone
Policy & Regulation · 6 min
Related Articles

BMO Claims Quantum Utility Milestone in AI Feature Selection, Using 100x Less Data
Products & Applications · 5 min

National Academies Report Calls for Faster, Tougher Retraction Standards
Models & Research · 5 min

Who Decides If Your Treatment Is Medically Necessary? Industry Leaders Say It Shouldn't Be AI Alone
Policy & Regulation · 6 min
More Stories
© 2026 Cedar & Bloom. All rights reserved.