Development and validation of an electronic daily control score for asthma (e-DASTHMA): a real-world direct patient data study
Bernardo Sousa-Pinto,Cristina Jácome,Ana Margarida Pereira,Frederico S Regateiro,Rute Almeida,Wienczyslawa Czarlewski,Marek Kulus,Mohamed H Shamji,Louis-Philippe Boulet,Matteo Bonini,Luisa Brussino,G Walter Canonica,Alvaro A Cruz,Bilun Gemicioglu,Tari Haahtela,Maciej Kupczyk,Violeta Kvedariene,Desirée Larenas-Linnemann,Renaud Louis,Marek Niedoszytko,Nhân Pham-Thi,Francesca Puggioni,Jan Romantowski,Joaquin Sastre,Nicola Scichilone,Luis Taborda-Barata,Maria Teresa Ventura,Rafael José Vieira,Ioana Agache,Anna Bedbrook,Karl C Bergmann,Rita Amaral,Luís Filipe Azevedo,Sinthia Bosnic-Anticevich,Guy Brusselle,Roland Buhl,Lorenzo Cecchi,Denis Charpin,Claudia Chaves Loureiro,Frédéric de Blay,Stefano Del Giacco,Philippe Devillier,Ewa Jassem,Guy Joos,Marek Jutel,Ludger Klimek,Piotr Kuna,Daniel Laune,Jorge Luna Pech,Mika Makela,Mario Morais-Almeida,Rachel Nadif,Hugo E Neffen,Ken Ohta,Nikolaos G Papadopoulos,Alberto Papi,Benoit Pétré,Oliver Pfaar,Daniela Rivero Yeverino,Carlos Robalo Cordeiro,Nicolas Roche,Ana Sá-Sousa,Boleslaw Samolinski,Aziz Sheikh,Charlotte Suppli Ulrik,Omar S Usmani,Arunas Valiulis,Olivier Vandenplas,Pedro Vieira-Marques,Arzu Yorgancioglu,Torsten Zuberbier,Josep M Anto,João A Fonseca,Jean Bousquet
DOI: https://doi.org/10.1016/S2589-7500(23)00020-1
Abstract:Background: Validated questionnaires are used to assess asthma control over the past 1-4 weeks from reporting. However, they do not adequately capture asthma control in patients with fluctuating symptoms. Using the Mobile Airways Sentinel Network for airway diseases (MASK-air) app, we developed and validated an electronic daily asthma control score (e-DASTHMA). Methods: We used MASK-air data (freely available to users in 27 countries) to develop and assess different daily control scores for asthma. Data-driven control scores were developed based on asthma symptoms reported by a visual analogue scale (VAS) and self-reported asthma medication use. We included the daily monitoring data from all MASK-air users aged 16-90 years (or older than 13 years to 90 years in countries with a lower age of digital consent) who had used the app in at least 3 different calendar months and had reported at least 1 day of asthma medication use. For each score, we assessed construct validity, test-retest reliability, responsiveness, and accuracy. We used VASs on dyspnoea and work disturbance, EQ-5D-VAS, Control of Allergic Rhinitis and Asthma Test (CARAT), CARAT asthma, and Work Productivity and Activity Impairment: Allergy Specific (WPAI:AS) questionnaires as comparators. We performed an internal validation using MASK-air data from Jan 1 to Oct 12, 2022, and an external validation using a cohort of patients with physician-diagnosed asthma (the INSPIRERS cohort) who had had their diagnosis and control (Global Initiative for Asthma [GINA] classification) of asthma ascertained by a physician. Findings: We studied 135 635 days of MASK-air data from 1662 users from May 21, 2015, to Dec 31, 2021. The scores were strongly correlated with VAS dyspnoea (Spearman correlation coefficient range 0·68-0·82) and moderately correlated with work comparators and quality-of-life-related comparators (for WPAI:AS work, we observed Spearman correlation coefficients of 0·59-0·68). They also displayed high test-retest reliability (intraclass correlation coefficients range 0·79-0·95) and moderate-to-high responsiveness (correlation coefficient range 0·69-0·79; effect size measures range 0·57-0·99 in the comparison with VAS dyspnoea). The best-performing score displayed a strong correlation with the effect of asthma on work and school activities in the INSPIRERS cohort (Spearman correlation coefficients 0·70; 95% CI 0·61-0·78) and good accuracy for the identification of patients with uncontrolled or partly controlled asthma according to GINA (area under the receiver operating curve 0·73; 95% CI 0·68-0·78). Interpretation: e-DASTHMA is a good tool for the daily assessment of asthma control. This tool can be used as an endpoint in clinical trials as well as in clinical practice to assess fluctuations in asthma control and guide treatment optimisation. Funding: None.