Diarrhoea is the primary cause of nintedanib discontinuation in idiopathic pulmonary fibrosis. We investigated predictors in 100 prospective patients; 46% experienced diarrhoea, defined as common terminology criteria for adverse events V.5.0 grade ≥1. In a multivariable model (area under the curve 0.9056), lower standardised body surface area (adjusted OR 0.20, 95% CI 0.06 to 0.55, p=0.004) emerged as a significant independent predictor. A linear mixed model showed that dose adjustments managed symptoms without compromising 12-month forced vital capacity or diffusing capacity of the lungs for carbon monoxide trajectories (p>0.05). Small body size influences drug tolerability, suggesting that simple anthropometric measures can aid in baseline risk stratification to improve adherence.
