Real-time modelling and visualisation of individual FEV1 trends in cystic fibrosis.

INTRODUCTION

Interpreting lung function trends in cystic fibrosis (CF) is challenging due to biological variability, irregular measurement intervals, and difficulties in establishing an individual baseline. Reliance on visual inspection alone may misclassify true change and delay recognition of clinically meaningful decline.

METHODS

We developed a real-time graphical application, integrated into the Swedish CF registry, which captures spirometry measurements collected in routine clinical care. Individual lung function trends were visualised using locally estimated scatterplot smoothing (LOESS), and annual trends were estimated using linear mixed-effects models fitted to continuously updated data from the Swedish CF cohort. Each individual's lung function trend was compared with an age-based reference group, which included all individuals aged 5-17 years for paediatric assessments and all individuals within ±5 years of the patient's current age for adults.

RESULTS

The application displays observed lung function values alongside dynamic LOESS curves, individual linear trends, and the population mean trend. Clinicians can estimate and visualise individual linear trends with 95% confidence intervals over the past 1, 3, or 5 years. The tool also provides estimated values, Z-scores, and deviations from each individual's estimated baseline, facilitating detailed assessment of both short- and long-term changes in lung function.

CONCLUSION

Integrating real-time FEV1 analysis into the Swedish CF registry provides clinicians and patients with a practical tool for continuous evaluation of short- and long-term lung function trends, facilitating the monitoring of disease progression and supporting clinical decision-making.