Differences in spirometry interpretation algorithms: influence on decision making among primary-care physicians

• Published in: NPJ primary care respiratory medicine 2015-03, Vol.25 (1), p.15008-15008, Article 15008
• Main Authors: He, Xiao-Ou, D’Urzo, Anthony, Jugovic, Pieter, Jhirad, Reuven, Sehgal, Prateek, Lilly, Evan
• Format: Article
• Language: English
• Subjects: 631/114; 631/378/2649/1409; 692/699/1785; 692/700/139; Algorithms; Asthma - diagnosis; Asthma - physiopathology; Decision Making; Diagnosis, Differential; Forced Expiratory Volume; Humans; Internal Medicine; Medicine; Medicine & Public Health; Physicians, Primary Care; Pneumology/Respiratory System; Primary Care Medicine; Pulmonary Disease, Chronic Obstructive - diagnosis; Pulmonary Disease, Chronic Obstructive - physiopathology; Spirometry; Thoracic Surgery; Vital Capacity
• ISSN: 2055-1010; 2055-1010
• DOI: 10.1038/npjpcrm.2015.8

Background: Spirometry is recommended for the diagnosis of asthma and chronic obstructive pulmonary disease (COPD) in international guidelines and may be useful for distinguishing asthma from COPD. Numerous spirometry interpretation algorithms (SIAs) are described in the literature, but no studies highlight how different SIAs may influence the interpretation of the same spirometric data. Aims: We examined how two different SIAs may influence decision making among primary-care physicians. Methods: Data for this initiative were gathered from 113 primary-care physicians attending accredited workshops in Canada between 2011 and 2013. Physicians were asked to interpret nine spirograms presented twice in random sequence using two different SIAs and touch pad technology for anonymous data recording. Results: We observed differences in the interpretation of spirograms using two different SIAs. When the pre-bronchodilator FEV 1 /FVC (forced expiratory volume in one second/forced vital capacity) ratio was >0.70, algorithm 1 led to a ‘normal’ interpretation (78% of physicians), whereas algorithm 2 prompted a bronchodilator challenge revealing changes in FEV 1 that were consistent with asthma, an interpretation selected by 94% of physicians. When the FEV 1 /FVC ratio was 12% and 200 ml, 76% suspected asthma and 10% suspected COPD using algorithm 1, whereas 74% suspected asthma versus COPD using algorithm 2 across five separate cases. The absence of a post-bronchodilator FEV 1 /FVC decision node in algorithm 1 did not permit consideration of possible COPD. Conclusions: This study suggests that differences in SIAs may influence decision making and lead clinicians to interpret the same spirometry data differently. Respiratory medicine: Interpreting asthma Variations among algorithms used to interpret ‘blow’ tests for diagnosis of asthma and lung disease may be skewing test results. The researchers, led by Anthony D'Urzo from the University of Toronto in Canada, had 113 primary care physicians make diagnoses from nine sets of blow test or spirogram results using two different spirogram interpretation algorithms (SIAs). They found for a particular case with impaired blow test results, one SIA resulted in a ‘normal’ diagnosis by 78% of physicians, while the other resulted in a diagnosis of ‘consistent with asthma’ by 94% of doctors. The findings suggest a need to standardise the algorithms in order to minimise differ