Survey Finds Most Pathology Diagnostic Errors Occur in Pre-Examination Phase

In a study surveying 43 pathologists to characterize diagnostic errors across the anatomical pathology workflow, errors were found to be most prevalent during the pre-examination phase. The investigation evaluated commonly encountered diagnostic breakdowns across laboratory practice to identify where procedural vulnerabilities concentrate.

Diagnostic errors in pathology continue to represent a significant contributor to patient harm, persisting despite substantial technological advances in laboratory automation and quality management systems.

To evaluate these workflow risks systematically, the researchers utilized Failure Mode and Effects Analysis, which provides a structured approach for the prospective identification and prioritization of operational vulnerabilities. The survey findings from the participating pathologists directly informed an FMEA-based risk assessment that pinpointed high-priority failure modes across the workflow.

Drawing upon the failure modes identified through the risk assessment, the authors proposed a conceptual framework that integrates agentic artificial intelligence with FMEA-driven quality management. Within this conceptual structure, artificial intelligence is outlined as offering capabilities in image analysis, data validation, and workflow monitoring to assist in error detection and failure reduction.

Under the proposed model, artificial intelligence systems are conceptualized to enable the continuous recalibration of risk through dynamic data feedback.

The authors explicitly emphasized that the publication represents a conceptual and hypothesis-generating contribution rather than a clinically validated intervention. Prospective validation, the establishment of standardized performance metrics, and real-world implementation studies remain necessary to evaluate the clinical and operational impact of such integrated systems.

Subject of Research: Diagnostic error reduction in anatomical pathology using artificial intelligence and failure mode analysis

Article Title: A conceptual framework integrating artificial intelligence with failure mode and effects analysis for diagnostic error reduction in anatomical pathology

Article References: Poombal, Lail, R. A., Satturwar, S., Singh, R., Ullah, E., & Parwani, A. V. (2026). A conceptual framework integrating artificial intelligence with failure mode and effects analysis for diagnostic error reduction in anatomical pathology. Diagnostic Pathology, 21(1), Article 82. https://doi.org/10.1186/s13000-026-01816-5

Image Credits: AI Generated

DOI: 10.1186/s13000-026-01816-5

Keywords: anatomical pathology, diagnostic errors, failure mode and effects analysis, artificial intelligence, agentic AI, pre-examination phase, quality management, workflow monitoring, data validation, image analysis, risk assessment, patient harm, laboratory automation, pathologist survey, dynamic data feedback, error detection, prospective validation, diagnostic pathology