Sex disparities in the phenotype at diagnosis of Sjögren's disease: artificial intelligence-driven characterisation in 17,416 patients
Bibliographic record
Abstract
OBJECTIVES: Sjögren disease (SjD) predominantly affects females, but the early disease presentation in male patients remains poorly characterised due to historically small sample sizes. The aim of this study was to investigate sex‑based differences in the clinical phenotype at diagnosis of SjD and identify predictors of patient sex using a large international cohort and AI‑enhanced analysis. METHODS: Cross-sectional analysis of an anonymised dataset comprising 17,416 worldwide patients fulfilling the 2002/2016 classification criteria (Sjögren Big Data Registry). We stratified the dataset by sex and conducted a comparative analysis of baseline glandular and systemic involvement, organ-specific ESSDAI domains, and immunological profiles. Multivariate logistic regression models were developed, adjusting for epidemiological confounders (age and ethnicity) to identify predictors of sex classification. We used a generative AI (OpenAI's GPT-4o model) environment with Python (version 3.9) and the pandas (1.4.3), numpy (1.21.5), and matplotlib (3.5.1) libraries. All analyses adhered to GDPR standards, with anonymized patient data and strictly controlled secure environments. RESULTS: The cohort included 1,161 (6.67%) men and 16,255 (93.33%) women, with a mean age at diagnosis of 51.11 years (SD=14.45). Men showed a higher mean age at diagnosis (54.09 vs. 51.42 years in women; t=6.08, p<0.0001), a higher average ESSDAI score (7.65 vs. 5.93; t=7.91, p<0.0001) and higher frequencies in severe DAS categories (i.e. high activity 20% vs. 12% in women, χ² = 81.15, p<0.0001). The epidemiologically-adjusted logistic regression model (pseudo R-squared value of 0.026) identified statistical significance for age (coefficient =0.009, p=0.024; each additional year in age increased the likelihood of being female by 1.4%), ethnicity (coefficient=0.579, HR=1.78, p=0.004), ocular dryness (coefficient=-0.607, HR=0.54, p<0.001), and systemic activity in the glandular (coefficient=0.359, HR=1.43, p=0.006) and pulmonary (coefficient=0.445, HR=1.56, p=0.004) ESSDAI domains. CONCLUSIONS: Male SjD patients present a distinct, more systemic phenotype at diagnosis. Awareness of sex‑specific features can improve early recognition and tailored management.
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 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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".