O8 Performance of the EULAR/ACR 2019 classification criteria for systemic lupus erythematosus in men, ethnicities, and early disease
Bibliographic record
Abstract
Background Supported by both the ACR and EULAR, the EULAR/ACR 2019 Classification Criteria for SLE employ positive ANA (ever) as an entry criterion and use a weighted scheme with values ranging from 2 to 10, for a classification cut-off of 10. Criteria items are attributed to SLE only if there is no more likely alternative diagnosis in the individual patients. Items are organized in domains, and only the highest ranking item within a domain is counted. These criteria have been validated in a cohort of 696 SLE patients and 574 non-SLE patients from a total of 21 centers, reaching an overall sensitivity of 96.1% and a specificity of 93.4%. To at least estimate the performance in groups underrepresented in the validation cohort of this transatlantic project, we analyzed this cohort for patient subsets with regard to sex, ethnicity, and disease duration. Methods The full EULAR/ACR 2019 classification criteria validation cohort was analyzed for female (n=1,098) and male (n=172) patients, Asian (n=118), Black (n=68), Hispanic (n=124) and White (n=941) patients, and patients with an SLE duration of less than 1 year (n=34), one to less than 3 years (n=196), 3 to less than 5 years (n=157), and 5 or more years (n=879). Sensitivity and specificity were calculated for the EULAR/ACR 2019 criteria, the SLICC 2012 criteria and the ACR 1997 criteria each. Results As shown in table 1, most of the point estimates for sensitivity and specificity in subsets lay within the 95% confidence intervals of the sensitivity and specificity of the EULAR/ACR 2019 criteria validation. In particular, sensitivity and specificity for all ethnic groups were within the confidence intervals or even higher. Formally, the sensitivity was slightly lower for male patients, corresponding to a higher specificity, but the male 95% confidence intervals (0.86–0.98 for sensitivity, 0.90–0.99 for specificity) overlapped. While sensitivity appeared independent of disease duration from year 1 on, sensitivity was only 89% in the first year of disease, identical to the SLICC criteria (89%) and numerically higher than the ACR criteria (56%), but all confidence intervals overlapped. Conclusion While not all subgroups of SLE patients in the validation cohort are of adequate size to fully explore the sensitivity and specificity of the EULAR/ACR 2019 SLE classification criteria in the respective subsets, the point estimates of sensitivity and specificity suggest that the new criteria perform at least reasonably well in all ethnic groups, in men and in early disease. Nevertheless, sensitivity and specificity should be independently validated in larger groups of Asian, Black and Hispanic patients, male patients and in early disease.
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 | 0.008 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".