Longitudinal reliability of unsupervised remote spirometry: a systematic review

Unsupervised remote spirometry enables frequent, home-based lung function monitoring, but the reliability of measurements performed without professional oversight remains uncertain. This systematic review synthesised longitudinal evidence on the reliability of unsupervised remote spirometry. MEDLINE, Embase, CENTRAL, and PubMed were searched from inception to May 11, 2026 for longitudinal studies reporting repeated unsupervised spirometry measurements, with or without comparison to supervised clinical spirometry. Primary outcomes were test–retest reliability and repeatability, assessed using coefficients of variation (COV) and intraclass correlation coefficients (ICC). Risk of bias was assessed using QUADAS-2. COV and ICC estimates were summarised descriptively, with exploratory correlation analyses undertaken for COV. 45 studies met inclusion criteria. 25 studies reported COV for unsupervised remote spirometry, with median values of 6.2% for FEV 1 (range 1.7–18.6%) and 5.9% for FVC (range 1.2–22.7%). Measurement variability was mostly low to moderate across respiratory cohorts, including ILD, cystic fibrosis, asthma and COPD, but higher in neuromuscular populations. Exploratory analyses showed no association between COV and study duration or mean age, while higher FEV 1 and FVC variability was observed in more recent and larger studies. Individual ICC estimates ranged from 0.65 to 1.00 for FEV 1 and from 0.85 to 1.00 for FVC/vital capacity, although few studies reported ICCs and substantial heterogeneity limited direct comparison. Few studies reported longitudinal reliability for supervised clinical spirometry. Unsupervised remote spirometry may provide reliable longitudinal lung function measurements in selected populations, but reliability varies by disease context. Current evidence suggests a role as a complementary monitoring tool rather than a replacement for supervised clinical spirometry.