Introduction
This document provides a formal correction to the research article titled "Utility of a patient similarity-based digital tool for risk communication to patients with type 2 diabetes mellitus: Perspectives from primary care physicians in ambulatory care," published in PLOS ONE. This update is necessary to ensure the completeness and transparency of the study's administrative disclosures, specifically regarding funding and competing interests.
Study Context and Objectives
The original research explored how primary care physicians utilize digital tools that leverage "patient similarity" to communicate health risks to individuals with type 2 diabetes. By comparing a patient's data with similar historical profiles, these tools aim to provide more personalized and persuasive risk assessments. The correction notice clarifies the institutional support behind this initiative, which is part of a larger effort to integrate artificial intelligence into chronic disease management.
Updated Funding Disclosure
The authors have updated the funding statement to acknowledge the support of the National Research Foundation, Singapore. The research was conducted under the AI Singapore Programme (Award No: AISG-GC-2019–001-2B).
It is important to note that the funding body had no role in the study design, data collection, analysis, or the decision to publish the findings. This disclosure reinforces the independence of the research and the objectivity of the results presented in the original article.
Competing Interests and Project Leadership
The correction also provides additional details regarding the authors' professional affiliations with related AI healthcare projects. Specifically, authors WH, MLL, and NCT are identified as Principal or Co-Principal Investigators of the “JARVIS-DHL” project.
The JARVIS-DHL initiative focuses on "Transforming Chronic Care for Diabetes, Hypertension and HyperLipidemia (DHL) with AI." Disclosing these leadership roles is essential for full transparency, allowing readers to understand the broader research context and any potential professional interests associated with the study of AI-driven clinical tools.
Conclusion
These administrative updates serve to maintain the high standards of transparency required in academic publishing. While the core scientific findings regarding the utility of patient similarity-based tools remain unchanged, this correction ensures that the public record accurately reflects the financial and institutional frameworks that made the research possible.
Summary updated