EXPEDITING LUPUS CLASSIFICATION OF AT-RISK INDIVIDUALS USING NOVEL TECHNOLOGY: OUTCOMES OF A PILOT STUDY
Bibliographic record
Abstract
PV217 / #472 Poster Topic: AS23 - SLE-Diagnosis, Manifestations, & Outcomes Background/Purpose Individuals at risk of developing systemic lupus erythematosus (SLE) often go through a difficult diagnostic journey and may receive conflicting diagnoses from multiple medical providers over time. This study seeks to utilize a novel technology-based program employing virtual and digital care models to determine if the time taken for accurate SLE classification of at-risk individuals could be shortened from the current 5 to 7 years. Methods Study participants were digitally recruited through a publicly available web portal designed for visitors to evaluate their risk of developing rheumatic connective tissue diseases with the Connective Tissue Disease Screening Questionnaire (CSQ). Individuals identified as “possible” (SLE-CSQ=3) or “probable” (SLE-CSQ≥4) risk of SLE were recruited to consent and participate in the study. Medical records (MRs) were obtained and reviewed for a stated SLE diagnosis and/or ICD-10 code M32.9 or related codes. Participants without an apparent diagnosis were eligible to move forward in the study in a sequential digital/virtual diagnostic protocol. They first completed a telehealth evaluation by a primary care physician (PCP) and mobile phlebotomy sample procurement for a predetermined panel of standard and lupus-associated laboratory tests. Laboratory test results, MRs, and PCP evaluation findings were made available to a community rheumatologist (CR) who completed a rheumatology-focused telehealth session. Finally, a tertiary care rheumatologist (TCR) specializing in SLE reviewed all study information and completed a telehealth session. The CR and TCR completed a classification form for each participant that included ACR 1997 or EULAR/ACR 2019 SLE classification criteria. Results The study target was 100 consented participants. In the first 60 days of the study, 108 participants that qualified by the CSQ and signed the informed consent were enrolled. The study population consisted of 95% females with mean age (SD) of 36 (6) years with 85% white, 7% black and 8% other races/ethnicities. MRs were requested for 102 that provided physician contacts; 81 were received. MRs review identified 67 with no previous SLE diagnosis and 14 with SLE diagnosis. Of the 67 who qualified, 39 completed the entire process and were evaluated by their PCP, CR, and TCR. Seven of 39 (18%) met SLE classification (ACR 1997 score range 4-6; EULAR/ACR 2019 score =13), 18 (46%) were classified as incomplete SLE, and 14 (36%) had no current indication of SLE. For those that met SLE classification, the time from date of consent to classification was mean (SD) 371 (43) days, with a range of 326 to 463 days. Conclusions A major goal of this virtual/digital study program was to shorten time to accurately diagnose SLE classification from the typical 5 to 7 years. For the 18% who met classification, the mean time to accurate diagnosis was 1 year and 6 days. For the 46% with incomplete SLE and the 36% with no current indication of SLE, the program may have the potential to shorten time to accurate classification as these participants are prospectively followed. The digital and virtual care technologies applied in this study program combined with currently available laboratory tests demonstrate the potential to effectively classify SLE in a remote care model. Acknowledgments: This study was sponsored by Progentec and funding was provided by GSK (GSK 219884). GSK was provided with the opportunity to review a preliminary version of this abstract for factual accuracy, but the authors are solely responsible for final content and interpretation.
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 | 0.004 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".