Use of forced vital capacity and forced expiratory volume in 1 second quality criteria for determining a valid test

The 2005 American Thoracic Society (ATS)/European Respiratory Society (ERS) spirometry guidelines define valid tests as having three acceptable blows and a repeatable forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1). The aim of this study was to determine how reviewer and computer-determined ATS/ERS quality could affect population reference values for FVC and FEV1. Spirometry results from 7777 normal subjects aged 8-80 years (NHANES (National Health and Nutrition Examination Survey) III) were assigned quality grades A to F for FVC and FEV1 by a computer and one reviewer (reviewer 1). Results from a subgroup of 1466 Caucasian adults (aged 19-80 years ) were reviewed by two additional reviewers. Mean deviations from NHANES III predicted for FVC and FEV1 were examined by quality grade (A to F). Reviewer 1 rejected (D and F grade) 5.2% of the 7777 test sessions and the computer rejected ∼16%, primarily due to end-of-test (EOT) failures. Within the subgroup, the computer rejected 11.5% of the results and the three reviewers rejected 3.7-5.9%. Average FEV1 and FVC were minimally influenced by grades A to C allocated by reviewer 1. Quality assessment of individual blows including EOT assessments should primarily be used as an aid to good quality during testing rather than for subsequently disregarding data. Reconsideration of EOT criteria and its application, and improved grading standards and training in over-reading are required. Present EOT criteria results in the exclusion of too many subjects while having minimal impact on predicted values.