Longitudinal spirometry data precision and the detectable excess rate of decline.
Hnizdo-E; Yu-L; Glindmeyer-HW
Proc Am Thorac Soc 2005 May; 2(Abstracts)::A410
The efficacy of decision-making based on longitudinal spirometry depends critically on the precision of the available data, which is determined by the size of within-person standard deviation Sw. We investigated how the size of Sw monitored over time on a group of workers determines the detectable excess decline in FEV1. 12,729 workers from 11 manufacturing plants participated in annual spirometry from 1987 to 2001. Testing was done by technicians trained in ATS guidelines. We "monitored" retrospectively yearly values of the plant-specific Sw using consecutive pairwise measurements. We estimated the plant-specific within-person standard deviation Sm from all longitudinal data by the mixed model. We estimated the detectable excess rate of decline based on the longitudinal lower limits of normal accounting for the size of Sw and duration of follow-up D. The average plant-specific values of Sw correlated with Sm (r=0.90). The plant-specific Sm values ranged 130179 ml. For a person i, the value Sw(i) reached 300 m1. Figure shows the detectable excess rate of decline in relation to the size of Sw and D. By monitoring the data precision on a group, one can determine the detectable rate of decline that can be considered excessive for given Sw and D. [figure1] The detectable excess rate of decline in FEV) (m1/yr), by within-person standard deviation and years of follow-up, for yearly testing.
Spirometry; Workers; Workplace-monitoring; Monitors; Monitoring-systems; Statistical-analysis; Surveillance
Abstract; Conference/Symposia Proceedings
Issue of Publication
Disease and Injury: Asthma and Chronic Obstructive Pulmonary Disease; Construction
Proceedings of the American Thoracic Society. 2005 ATS International Conference, May 20-25, 2005, San Diego, California