Development and Validation of an Algorithm for Quality Grading of Pediatric Spirometry: A Quality Improvement Initiative

Rationale: Current spirometry quality grading for individuals 7 years and older include within-test repeatability thresholds up to 250 ml, which may be inappropriately wide for children. Objectives:1) To develop, internally validate, and implement a quality grading algorithm for forced expiratory volume in 1 second (FEV₁) and forced vital capacity (FVC) for school-aged children and 2) to compare the algorithm with the one proposed by the American Thoracic Society (ATS). Methods: We conducted a review of existing algorithms and obtained expert input. A pediatric quality grading algorithm was drafted and modified in an iterative process until consensus was achieved, with the main difference from current criteria being tighter volume repeatability for the pediatric quality grading. Four pulmonary function technicians evaluated the interrater agreement of the algorithm in a blinded fashion on an unselected consecutive sample of 87 prebronchodilator spirometry tests, and the grades were compared with those from the ATS algorithm in the same sample of spirometry tests. The algorithm was then implemented into the workflow of the pulmonary function laboratory. Results: For FEV₁ and FVC, the interrater agreement values for the pediatric algorithm were 92% and 83%, respectively. When the ATS algorithm was used, 75.9% (n = 66) and 63.2% (n = 55) of subjects achieved a grade of A for FEV₁ and FVC; when the pediatric algorithm was used, 69.0% (n = 60) and 58.6% (n = 51) met grade A criteria. There was a more uniform distribution of tests for the pediatric algorithm across grades B through F for FEV₁ and FVC, and no grade C tests were observed for the ATS grading algorithm. A total of 2,464 tests graded prospectively by using the pediatric algorithm showed a median (interquartile range) FEV₁ and FVC repeatability within 29 ml (13-57 ml) and 34 ml (15-66 ml), respectively. Most subjects received a grade of A for FEV₁ (81.1%) and FVC (71.6%), performing a repeatable spirometry test to within 100 ml. Conclusions: A quality grading algorithm that uses smaller ranges of expired volumes to define repeatability is feasible and may be more appropriate in a pediatric pulmonary function laboratory.