Adjudicating evolving inpatient diagnoses with a clinical coding agent: a retrospective multicentre study
- 1Zhongshan Hospital, Fudan University
Abstract
BackgroundInternational Classification of Diseases (ICD) codes are widely used for reimbursement, health-service planning, and disease surveillance. Automated systems can identify diagnoses and supporting evidence across clinical documents. Yet coding is more than finding diagnoses. Decisions based on isolated information may not fully reflect the evolving clinical picture during admission. Provisional, superseded, or conflicting diagnoses can therefore persist in the final code set, complicating review and correction. We therefore developed and evaluated Clinico, a coding agent that organises evidence by time and clinical relationships to adjudicate diagnoses and update their status during admission.
MethodsWe did a retrospective multicentre study with internal and external validation using inpatient records from two Shanghai hospitals. Routine coder-finalised codes served as references. Using 5,000 development stays, we refined Clinicos prompts and workflow without fine-tuning model parameters. We compared Clinico with three prespecified comparators, two trained on 80,340 stays. Clinico maintained an evidence-linked diagnostic ledger across each admission and reconciled related or conflicting claims before finalising the code set and selecting the principal diagnosis. The primary outcome was principal-diagnosis exact-match agreement; complete code-set micro-F1 was the key secondary outcome.
FindingsEvaluation included 5,000 internal and 4,109 external test stays. Principal-diagnosis exact-match agreement was 83.6% (95% CI 82.6-84.6) internally and 52.0% (50.5-53.5) externally; the next-highest observed estimates were 74.7% and 31.5%. Complete code-set micro-F1 was 90.6% (95% CI 90.2-91.0) internally and 42.6% (42.0-43.2) externally; the next-highest observed estimates were 90.3% and 25.5%.
InterpretationIn both hospital test sets, Clinico had the highest observed estimates of principal-diagnosis agreement and complete code-set micro-F1 among evaluated methods. Its longitudinal adjudication process links decisions to source evidence, allowing them to be checked against the record. Prospective studies should assess its effects in clinical workflows using independently adjudicated references.
FundingNone.
Related
- Retrospective Review of Suspected Melanoma Diagnoses Made by a Single-Centre Teledermoscopy Service in New Zealand.
- Changes in hepatic functional reserve and renal function during hospitalization in patients with chronic liver disease : A retrospective observational study of inpatient rehabilitation
- A Structured, Auditable Taxonomy for Pharmacist-Identifiable Medication-Related Risk Signals in Routine Admission Data: A Retrospective Medication Safety Audit in a High-Risk Inpatient Cohort
- Postoperative Risks in Thoracolumbar Fractures Patients with Mental Disorders: A Retrospective Cohort Study Using a Nationwide Inpatient Database
- Oral Cavity Cancer and the Subsequent Incidence of Other Cancer Diagnoses: A Retrospective Cohort Study in German General Practices
BibTeX
@misc{li2026adjudicating,
title = {{Adjudicating evolving inpatient diagnoses with a clinical coding agent: a retrospective multicentre study}},
author = {Y. Li and Y. Shi and Y. Sun and T. Suo and P. Wang and Z. Zhang and J. Xian and L. Gu and W. Ma and J. Gu and J. Fan},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.31.26361789},
url = {https://www.medrxiv.org/content/10.64898/2026.08.31.26361789v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Adjudicating evolving inpatient diagnoses with a clinical coding agent: a retrospective multicentre study
AU - Y. Li
AU - Y. Shi
AU - Y. Sun
AU - T. Suo
AU - P. Wang
AU - Z. Zhang
AU - J. Xian
AU - L. Gu
AU - W. Ma
AU - J. Gu
AU - J. Fan
PY - 2026
DA - 2026/08/31
PB - medRxiv
DO - 10.64898/2026.08.31.26361789
UR - https://www.medrxiv.org/content/10.64898/2026.08.31.26361789v1
AB - BackgroundInternational Classification of Diseases (ICD) codes are widely used for reimbursement, health-service planning, and disease surveillance. Automated systems can identify diagnoses and supporting evidence across clinical documents. Yet coding is more than finding diagnoses. Decisions based on isolated information may not fully reflect the evolving clinical picture during admission. Provisional, superseded, or conflicting diagnoses can therefore persist in the final code set, complicating review and correction. We therefore developed and evaluated Clinico, a coding agent that organises evidence by time and clinical relationships to adjudicate diagnoses and update their status during admission. MethodsWe did a retrospective multicentre study with internal and external validation using inpatient records from two Shanghai hospitals. Routine coder-finalised codes served as references. Using 5,000 development stays, we refined Clinicos prompts and workflow without fine-tuning model parameters. We compared Clinico with three prespecified comparators, two trained on 80,340 stays. Clinico maintained an evidence-linked diagnostic ledger across each admission and reconciled related or conflicting claims before finalising the code set and selecting the principal diagnosis. The primary outcome was principal-diagnosis exact-match agreement; complete code-set micro-F1 was the key secondary outcome. FindingsEvaluation included 5,000 internal and 4,109 external test stays. Principal-diagnosis exact-match agreement was 83.6% (95% CI 82.6-84.6) internally and 52.0% (50.5-53.5) externally; the next-highest observed estimates were 74.7% and 31.5%. Complete code-set micro-F1 was 90.6% (95% CI 90.2-91.0) internally and 42.6% (42.0-43.2) externally; the next-highest observed estimates were 90.3% and 25.5%. InterpretationIn both hospital test sets, Clinico had the highest observed estimates of principal-diagnosis agreement and complete code-set micro-F1 among evaluated methods. Its longitudinal adjudication process links decisions to source evidence, allowing them to be checked against the record. Prospective studies should assess its effects in clinical workflows using independently adjudicated references. FundingNone.
ER -