Lung function in healthy never smoking adults: reference values and lower limits of normal of a Swiss population. Issue 3 (March 1996)

• Record Type: Journal Article
• Authors: Brändli, O; Schindler, C; Künzli, N; Keller, R; Perruchoud, A P
• Is Part Of: Thorax. Volume 51:Issue 3(1996)
• Page Start: 277
• Page End: 283
• Publication Date: 1996-03
• Subjects: Chest -- Diseases -- Periodicals; Thorax; Chest -- Diseases; Periodicals; Periodicals; 617.54
• DOI: 10.1136/thx.51.3.277
• Languages: English
• ISSNs: 0040-6376

Abstract

Abstract : BACKGROUND: Reference values and definitions of "normal" are prerequisites for population screening and classification of lung diseases. The aim of this study was to calculate reference values for never smoking Caucasian adults. METHODS: In the SAPALDIA cross sectional study respiratory health was assessed in a random sample of 9651 subjects, aged 18-60 years, from eight areas of Switzerland. Lung function was measured according to ATS criteria including quality control. In 3157 healthy never smoking adults without respiratory symptoms the mean values and fifth percentiles of lung function variables were calculated. For each sex, logarithms of lung function were regressed against age, age squared, and the logarithm of height. Residuals were used to estimate fifth percentiles across the age range using a technique not requiring normality or homoscedasticity of residuals. RESULTS: Most lung function variables were non-linear with age and showed an increase in early adulthood and an accelerated decline thereafter. The reference values for forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) were higher than those of the European Community for Coal and Steel and those from North America. The prediction equations for lower fifth percentile values defined a stable proportion of subjects outside this limit whereas alternative methods for estimating the fifth percentile showed a loss of sensitivity with age. CONCLUSIONS: The reference equations for mean values of spirometric indices and their lower fifth percentiles gave an improved and unbiased lower limit of normal. The higher mean values may in part be due to the strictly selected population, quality control procedures, cohort effects, and altitude, and are not explained by the statistical model used.