Predicting Trajectories of Long COVID in Adult Women: The Critical Role of Causal Disentanglement
Organizations: National Library of Medicine, Bethesda, MD · Stevens Institute of Technology, Hoboken, NJ · Columbia University, NYC, NY · The George Washington university, Washington, DC ·
Abstract
Early prediction of Post-Acute Sequelae of SARS-CoV-2 severity is a critical challenge for women's health, particularly given the diagnostic overlap between PASC and common hormonal transitions such as menopause. Identifying and accounting for these confounding factors is essential for accurate long-term trajectory prediction. We conducted a retrospective study of 1,155 women (mean age 61) from the NIH RECOVER dataset. By integrating static clinical profiles with four weeks of longitudinal wearable data (monitoring cardiac activity and sleep), we developed a causal network based on a Large Language Model to predict future PASC scores. Our framework achieved a precision of 86.7% in clinical severity prediction. Our causal attribution analysis demonstrate the model's ability to differentiate between active pathology and baseline noise: direct indicators such as breathlessness and malaise reached maximum saliency (1.00), while confounding factors like menopause and diabetes were successfully suppressed with saliency scores below 0.27.
Figures & tables
| Metric | XGBoost | Causal Inference Network |
|---|---|---|
| MSE | ||
| RMSE | ||
| MAE | ||
| Accuracy | ||
| Precision |
| Clinical Category | Representative Tokens | Max Saliency | Clinical Rationale |
|---|---|---|---|
| Active Pathology | breath, malaise, refreshing | 1.0000 | Direct indicators of pulmonary and systemic dysfunction. |
| Cognitive Symptoms | fog, brain, insomnia | 0.54 – 0.74 | Identifies hallmark Neuro-PASC clusters. |
| Baseline Noise | diabetes, menopause | 0.27 | Successfully filtered as non-causal confounding data. |
| Administrative | that, to, by | 0.17 | Ignored as low-information linguistic fillers. |
| Characteristic | Category / Diagnosis | Total Cohort (N=1,155) |
| Age (Years) | ||
| Median (IQR) | 61 (56, 67) | |
| Range (Min, Max) | 37, 89 | |
| Race and Ethnicity, N (%) | ||
| White / Native Hawaiian | 743 (64.32%) | |
| Hispanic / Latino | 173 (14.97%) |
Clinical Outcome Prediction
Clinical outcome prediction uses patient records, imaging, signals, and biomarkers to estimate risks, disease trajectories, treatment response, and survival.
See all 265 papers