US-based Elsevier Health & OpenEvidence Launch Generative AI Tool ClinicalKey

Elsevier launches ClinicalKey AI, integrating generative AI into its clinical decision support tool, aiming to streamline medical information access.

Prominent player in US healthcare information sector, Elsevier Health, has unveiled a generative AI-driven version of its flagship clinical decision support tool, ClinicalKey, in collaboration with AI company OpenEvidence. 

While the original tool remains accessible, ClinicalKey AI is now available as an optional supplementary purchase, offering users a more comprehensive solution for their clinical decision-making needs. 

The latest add-on module of ClinicalKey incorporates an innovative conversational search interface. It also facilitates a faster and more intuitive method for doctors to locate pertinent medical information. 

As reported by the company, over 30,000 physicians across the US have already tested the underlying AI technology, demonstrating a significant interest in its potential benefits. This move aims to enhance the efficiency of medical practitioners in accessing crucial information.

Enhanced Content Accessibility & Personalization

ClinicalKey AI leverages validated sources beyond Elsevier's existing point-of-care content, including journal abstracts, medical reference texts, and government publications. 

The tool summarises responses with proper citations, considering patient-specific factors such as comorbidities and current medications. 

Moreover, it tailors responses based on the clinician's profile or specialty and offers access to previous queries and responses. Notably, ClinicalKey AI adheres to HIPAA compliance standards, ensuring the security and confidentiality of patient information.

Jan Herzhoff, president, Elsevier Health, highlighted the significance of addressing the search challenges medical professionals face. He emphasized the collaborative efforts with OpenEvidence in developing this innovative solution to enhance the efficiency of clinical decision support systems.

Improving Clinical Decision-Making Efficiency

Based on user behaviour analysis and market research, Elsevier identified three primary categories of user queries: basic, medium, and hard complexity. Questions ranging from basic drug-related inquiries to complex issues such as interactions between drugs or comorbidities often pose challenges for clinicians. 

Additionally, ClinicalKey AI aims to mitigate these challenges by providing timely and accurate responses, thereby potentially reducing the percentage of unanswered queries across all complexity levels.

Elsevier's partnership with Cone Health and the University of New Mexico demonstrates its focus on improving the accuracy and usability of the tool. Through collaboration with subject matter experts and feedback from test users, Elsevier aims to continuously enhance ClinicalKey AI's capabilities and effectiveness in addressing clinicians' information needs.

Anticipated Impact & Future Integration

While ClinicalKey AI is currently not integrated with Electronic Health Records (EHRs), Herzhoff expressed optimism regarding future integration possibilities. 

Moreover, results from a recent survey conducted by Elsevier indicate a growing acceptance and desire among physicians for integrating generative AI tools into clinical decision-making processes. 

Despite some authors expressing concerns over potential copyright issues, the tool's reliance on licensed content ensures credibility and reduces the risk of inaccuracies.

In a similar development, US-based Data Kinetic, an applied AI research, development, and advisory company, launched its first suite of industry-focused applications designed specifically for healthcare workflows and solutions. The solution suite enables healthcare specialists to utilize their preferred platforms to run models securely, leveraging air-gapped security to isolate models and prevent data leakage.