Background and objective: Fibrotic interstitial lung disease (F-ILD) has high mortality, necessitating multidimensional staging to assess severity, predict prognosis, and guide treatment. However, the gender-age-physiology (GAP) index lacks patient-reported outcome measures (PROMs) and exercise test parameters, and traditional methods may miss complex variable interactions. The study aimed to develop a risk model of mortality in F-ILD using routine PROMs and exercise test parameters, and to compare the predictive performance with GAP staging system.
Methods: Between December 2018 and December 2022, 246 patients from the ILD Prospective Registry with at least 1 year of follow-up (excluding death) were enrolled. Baseline assessments included GAP, 1MSTS, 6MWD, mMRC, SGRQ, and SF36. A tree-based risk model was developed and validated with a 2:1 split. Kaplan-Meier curves and Cox regression were used to estimate survival and hazard ratios, with statistical significance set at p < 0.05.
Results: The very-high-risk group (SGRQ > 34.5, 1MSTS < 21) had a significantly higher hazard ratio (HR = 5.13; 95% CI: 2.54-10.35) compared to the non-very-high-risk group. Their 1-, 2-, and 5-year mortality rates were 28.87%, 47.18%, and 82.42%, respectively, similar to GAP stage III but with narrower confidence intervals.
Conclusion: This study developed a prognostic model for F-ILD combining PROMs and functional exercise data. The model, which includes the SGRQ, 6MWD, and 1MSTS, outperforms GAP staging, offering benefits like reduced patient fatigue and improved monitoring. Multicenter studies are needed to validate these findings.
Trial registration: The study was registered on ClinicalTrials.gov (NCT06476470).
Keywords: 1MSTS; 6MWD; GAP; SGRQ; fibrotic interstitial lung disease; mortality; tree‐based learning method.
© 2025 The Author(s). Respirology published by John Wiley & Sons Australia, Ltd on behalf of Asian Pacific Society of Respirology.