Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Promise or Precaution: An Exploratory Study of the Performance of OpenEvidence (OE) in Clinical Decision Making in the Global Context.

Created on 16 Nov 2026

Authors

Jennifer Benjamin, Ken Masters, Brenda Kunga, Jen-Feng Liang, Anoop Agrawal, Alvaro Margolis, Shelley Kumar, Adhari AlZaabi, Heather MacNeill, Sudha J Rajan, Sofia Valanci

Published in

Cureus. Volume 18. Issue 10. Pages e118022. Epub Oct 16, 2026.

Abstract

Background Clinicians face the challenging task of keeping up with evidence-based literature to inform clinical decision-making. OpenEvidence (OE) (OpenEvidence, Miami, USA) holds great promise by providing the latest evidence to inform patient care management globally.  Objective To explore the performance of OE in a global context.  Methods Eleven clinicians across six countries, including the United States, Canada, Uruguay, Kenya, Taiwan, and Oman, performed standardized searches for 10 common clinical conditions, with each query identifying the clinician's institution and location. OE search outputs (80 evaluations) were evaluated using the Rubric Assessment Scale (RAS). Additional questions compared US-based results versus other countries. Data were analyzed using descriptive statistics. Differences between US based results and the global group were assessed using the Wilcoxon rank-sum test to compare median values as part of an exploratory proof of concept study.  Results OE performed well on core quality metrics. About 88% of the searches were rated as "completely" or "almost entirely" accurate, and 89% were "efficient and succinct" or "mostly efficient," as rated by the authors, with lower scores of 73% for comprehensiveness, 75% for argument and evidence, and 65% for creativity and originality.  US-based searches demonstrated alignment with management guidelines (p=0.02), with higher perceived overall value (p=0.02) and more frequent inclusion of locally authored references by experts (p<0.0001). Greater variability in differential diagnosis was observed in the global group [interquartile ranges (IQR) 1-5 vs 1-2: p=0.03], reflecting wider rater disagreement rather than less relevant differentials, alongside a notable lack of region-specific management recommendations. Of the search results, most references originated in the US, with no Global South references, lacking contextualized management based on regional society guidelines. OE results lacked epidemiologically relevant differentials such as malaria for a febrile child in Kenya. Accessibility to OE was variable, with India lacking access and several Global South countries experiencing variable access.  Conclusion In this exploratory study, OE returned an evidence base that was overwhelmingly US-derived, lacking Global South references, and did not vary with geographic origin of query even when the location was explicitly requested. There was output variability in management suggestions, with US-based searches showing greater alignment with clinical guidelines and locally authored references compared to global searches. These findings are hypothesis-generating and cannot establish causal mechanisms; given the exploratory design and small sample size with convenience sampling, the results should be interpreted cautiously. Clinicians need to critically analyze and evaluate artificial intelligence (AI) outputs and be aware of inherent biases before applying them to clinical practice, especially when caring for patients in global contexts.

PMID:
43036754
Bibliographic data and abstract were imported from PubMed on 16 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 12
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement