Coding diagnoses and procedures using a high‐quality clinical database instead of a medical record review
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
A discharge abstract must be completed for each hospitalization. The most time-consuming component of this task is a complete review of the doctors' progress notes to identify and code all diagnoses and procedures. We have developed a clinical database that creates hospital discharge summaries. To compare diagnostic and procedural coding from a clinical database vs. the standard chart review by health records analysts (HRA). All patients admitted and discharged from general medical and surgical services at a teaching hospital in Ontario, Canada. Diagnostic and procedural codes were identified by reviewing discharge summaries generated from a clinical database. Independently, codes were identified by hospital health records analysts using chart review alone. Codes were compared with a gold standard case review conducted by a health records analyst and a doctor. Coding accuracy (percentage of codes in gold standard review) and completeness (percentage of gold standard codes identified). The study included 124 patients (mean length of stay 5.5 days; 66.4% medical patients). The accuracy of the most responsible diagnosis was 68.5% and 62.9% for the database (D) and chart review (C), respectively (P = 0.18). Overall, the database significantly improved the accuracy (D = 78.9% vs. C = 74.5%; P = 0.02) and completeness (D = 63.9% vs. C = 36.7%; P < 0.0001) of diagnostic coding. Although completeness of procedural coding was similar (D = 5.4% vs. C = 64.2%; P = NS), accuracy decreased with the database (D = 70.3% vs. C = 92.2%; P < 0.0001). Mean resource intensity weightings calculated from the codes (D = 1.3 vs. C = 1.4; P = NS) were similar. Coding from a clinical database may circumvent the need for HRAs to review doctors' progress notes, while maintaining the quality of coding in the discharge abstract.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.200 | 0.623 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.003 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 it