Concordance and discordance in disease severity classification between clinician judgments and cognitive testing scores for Alzheimer’s disease in the United States Veterans Affairs Healthcare System
Bibliographic record
Abstract
Abstract Background The current standard of practice for assessment of Alzheimer’s disease (AD) severity in the Department of Veterans Affairs (VA) Healthcare System is primarily based on clinical examination and cognitive testing. We aimed to assess the consistency between clinician judgements of AD severity compared with that classified by cognitive test scores in the VA system. Method We identified medical notes for patients with mild, moderate, or severe AD from the VA electronic healthcare records (EHR) database between March 2008 and October 2021, using the natural language processing technology. Mini‐Mental Status Exam (MMSE) and Montreal Cognitive Assessment (MoCA) scores from the corresponding medical notes with physician judgments were classified and used for comparisons of severity levels. A binary variable was coded to indicate concordance or discordance for each paired comparison. Result We screened clinical notes from 51,809 AD patients and analyzed 8,888 clinical notes that had either an MMSE or MoCA test score from 5,150 AD patients (3.5% female; average age at 78.1±9.4 years). Clinician judgements and test‐score‐based classification of AD severity were concordant in 53.2% of comparisons (weighted Kappa = 0.39, 95% CI:0.38‐0.41, p = 0.009); whereas clinician judgments were less severe in 25.7% and more severe in 21.1% comparisons, relative to the test‐score‐based severity level. Clinician judgments were concordant with 54.1%, 51.8%, and 53.0% cognitive test‐score‐based classifications, of mild, moderate, and severe AD, respectively. Concordance was observed in 55.9% clinician judgments involving internist, 53.0% involving psychiatrist or neurologist, and 51.7% involving psychologist specialties, and in 50.4% involving nurse practitioners. Among discordant severity classifications, clinician judgments tended to be less severe than test scores (Figure). Conclusion Examination of VA EHR reveals that nearly one half of clinician judgments of AD severity are discordant with severity classification based on the MMSE or MoCA test scores. Discordance is more pronounced in the moderate relative to mild and severe AD categories but does not vary substantially over clinicians’ specialty background. Research will aim to find plausible explanations for the variation of concordance vs discordance by further examining the relative contributions of patient, clinician, and system characteristics in the process of AD clinical assessments.
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 | 0.023 | 0.077 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.000 |
| 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 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".