Enhancing description of hospital-conditions with ICD-11 cluster coding: Better codes for monitoring and prevention
Bibliographic record
Abstract
IntroductionExposure to health care events sometimes has unintended and undesired consequences. Health care and complications arising in the course of care are diverse and complex. Representing them comprehensively in information systems is challenging, and presently beyond the bounds of practicality for routine administrative information systems that include ICD coded data. Objectives and ApproachThe ICD-11 conceptual model for hospital-acquired conditions has 3 components: 1) harm to patient 2) cause or source of harm and 3) mode or mechanism. A key feature of the Quality and Safety (Q\&S) code-set in ICD-11 is that a cluster of codes is required to represent an event or injury. Use of the term ‘cluster’ is novel in ICD-11 and so is the extent and the requirement for post-coordination. The cluster required to code a Q\&S case has three codes, one for each of the three components of the model given above. ResultsThe first component, ‘harm’, is represented by an ICD–11 diagnosis code, from any chapter of the classification. Q\&S causes or sources of harm fall into 4 types that capture events caused by substances (drugs and medicaments, etc.), procedures, devices, and a mix of other types of causes (e.g. problems associated with transfusions, incorrect diagnosis, etc.). Q\&S ‘mode or mechanism’ refers to the main way in which the ‘cause’ leads to the ‘harm’ and are specific to the type of ‘cause’ (Table 1). Table 1 - Examples of corresponding Q\&S Mode or Mechanism Cause or Source of Harm Mode or Mechanism Substance Overdose, under-dose, wrong substance. Procedure Accidental perforation of an organ during a procedure. Device Dislodgement. Malfunction. Other cause Mismatched blood. Patient dropped during transfer from OR table. Conclusion/ImplicationsThis new conceptual model for coding healthcare-related harm, dependent on the clustering of codes, has great potential to improve the clinical detail of adverse event descriptions, and the overall quality of coded health data, for better monitoring and strategies for prevention.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.018 | 0.098 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.007 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.006 | 0.007 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.018 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".