COPD Assessment Test Variability as a Predictor of Exacerbations in Patients with COPD: A Prospective Study Using Virtual Assistant Monitoring

Le résumé fourni par la source

Although short-term CAT score variability has been associated with COPD exacerbations, evidence supporting its integration into virtual assistant-based monitoring remains limited. The aim of this study was to assess the short-term variability in CAT scores using a virtual management tool in patients with COPD and to examine its relationship with the prediction of exacerbations. A prospective observational single-center study was conducted in patients aged >35 with COPD and an exacerbator phenotype (≥2 outpatient exacerbations or ≥1 hospitalization in the previous year). Follow-up was performed weekly, including the CAT questionnaire, through calls were conducted by the virtual assistant Tucuvi Health Manager® during a period of one year. A total of 106 patients were included; 64.2% were male, with a mean (SD) age of 66.8 (7.9) years and a median FEV1(%) of 38.8 (30.1–45.0). The mean follow-up was 25 weeks [IQR 20–35], during which 73 (69%) patients experienced ≥1 exacerbation, accounting for 192 episodes (26% of which required hospitalization). A total of 2550 CAT scores were collected. The mean baseline CAT score was 17.9 (7.6), and scores increased significantly during exacerbations compared with non-exacerbation periods and according to exacerbation severity. Each one-point increase in absolute CAT score was associated with an 11.1% increase in the odds of exacerbation (OR: 1.11; 95% CI: 1.08–1.13), while a weekly variation of ≥2 points (MCID) was associated with more than sixfold higher odds (OR: 6.30; 95% CI: 4.28–9.27). The optimal cut-off for CAT change in the prediction of exacerbation risk was 3 points with a ROC curve of 0.773. Weekly deterioration in CAT scores was associated with increased exacerbation risk in patients with COPD. Short-term increases in CAT may serve as clinically useful warning signals for exacerbation.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.