Physician Preferences: OpenEvidence Ranks Highest Among AI Chatbots in Study

OpenEvidence Surpasses All Other AI Chatbots in Independent Study

In an independent study conducted by Stanford and Harvard, significant findings have emerged regarding the use of AI in clinical decision-making. Physicians were given the freedom to consult any available AI support tool, and surprisingly, OpenEvidence was favored over all other AI chatbots combined, including recognized names like ChatGPT, Claude, and Gemini. This central finding of the NOHARM (Numerous Options Harm Assessment for Risk in Medicine) study underscores the trust that healthcare professionals place in OpenEvidence.

The NOHARM study, spearheaded by a network of clinical AI researchers from prestigious medical schools, evaluated numerous large language models and clinical AI systems, comprising 12,747 expert annotations across 4,249 potential clinical actions. Though the benchmarks set by the study are not peer-reviewed and warrant cautious interpretation, the insights into physicians' preferences provide a revealing picture of how real-world clinicians make decisions.

In a randomized segment involving 101 board-certified U.S. physicians, those given complete discretion to select their clinical decision support tools opted for OpenEvidence 22.3% of the time. This was notably more than the cumulative choices made for all other AI chatbots—19.8%. The significance of these numbers cannot be understated: they reflect both revealed preference and the inherent trust physicians have in OpenEvidence as a reliable resource.

Why Do Physicians Prefer OpenEvidence?

OpenEvidence has established itself not through incentives or compensation but through the trust engendered by its grounded, evidence-based answers. Each individual among the hundreds of thousands of verified clinicians who consult it daily does so independently, reaffirming the platform's integrity at the point of care. Moreover, the content delivered by OpenEvidence is consistently sourced, cited, and anchored in peer-reviewed medical literature, reinforcing its credibility.

The recent launch of EvidenceGrade™ further solidifies OpenEvidence’s commitment to transparency and quality. This new feature allows clinicians to view in real-time the strength of the evidence supporting each answer. Built upon the GRADE framework used by major health organizations, EvidenceGrade visually represents the quality of published research that informs each response. This empowers physicians to make informed decisions based on robust evidence rather than simply relying on post-fact verification processes.

OpenEvidence recognizes that the trustworthiness of medical answers is intrinsically linked to the strength of their underlying evidence. As Daniel Nadler, the founder and CEO of OpenEvidence, articulated: “The most important benchmark in clinical AI isn't determined by researchers; it is established by physicians with hundreds of thousands of decisions made every day.” This statement encapsulates the crux of evidence-based medicine: physicians opt for OpenEvidence because their trust is built on legitimate evidence, evaluated openly with each clinical interaction.

The NOHARM study underscores that its findings leave a compelling insight into the world of clinical AI: when given the choice, clinicians overwhelmingly gravitate towards resources that reflect their confidence in the underlying research and data. OpenEvidence's increasing adoption among U.S. physicians speaks volumes about its role in empowering informed clinical decisions.

Conclusion

To sum up, OpenEvidence has proven to be a trusted ally among physicians navigating the complexities of clinical care. As the study reveals, the trust in OpenEvidence is not a result of incentives but a recognition of its unwavering commitment to proof-based information. Moving forward, OpenEvidence aims to continue providing high-quality, evidence-based resources to support medical professionals in their daily practices, reinforcing the foundational principle of patient care and effective medical decision-making.

For more information about OpenEvidence and its offerings, visit OpenEvidence.com.