Asthma diagnosis: a comparison of established diagnostic guidelines in adults with respiratory symptoms

Background Considerable variability exists between asthma diagnostic guidelines. We tested the performance characteristics of the European Respiratory Society (ERS), the National Institute for Health and Care Excellence (NICE) and the Global Initiative for Asthma (GINA) guidelines for the diagnosis of asthma in adults. Methods In this prospective observational study (ISRCTN-11676160, - 11676160, May 2019-June - June 2022), participants referred from primary care with clinician-suspected asthma underwent comprehensive investigation including: spirometry, bronchodilator reversibility, fractional exhaled nitric oxide, peak expiratory fl ow variability, bronchial challenge testing with methacholine and mannitol, and responsiveness to inhaled corticosteroid therapy. Results were reviewed by a panel of asthma specialists to determine asthma diagnosis (reference standard) and compared to each diagnostic test and the ERS, NICE and GINA diagnostic algorithms (index tests). The sensitivity, specificity, fi city, positive predictive and negative predictive values were calculated. Findings One hundred and forty adults were enrolled and 118 given a definitive fi nitive diagnostic outcome [75 female; mean (SD) age 36 (12) years; 70 (59%) with asthma] and included in the analysis. Sensitivity of individual tests was poor (15-62%), - 62%), but they provided good specificity fi city at the most stringent thresholds (range: 88-100%). - 100%). The sensitivity/ specificity fi city of ERS, NICE and GINA was 81/85%, 41/100% and 47/100%, respectively. Concordance between guidelines was only moderate (Cohen's ' s Kappa 0.45-0.51). - 0.51). Interpretation Current guidelines for the diagnosis of asthma in adults provide either excellent specificity fi city but low sensitivity (GINA and NICE) or only reasonable sensitivity and specificity fi city (ERS). All guidelines therefore have limitations with regards to their clinical application; new guidelines are needed but should be tested prospectively before roll out.