Article Text

Download PDFPDF

Paediatric respiratory disease

Statistics from Altmetric.com

P69 CAN CHANGES IN SPIROMETRY OVER TIME BE INTERPRETED USING CROSS-SECTIONAL REFERENCE DATA?

1S. Stanojevic, 1T. J. Cole, 1A. Wade, 1C. Bastardo, 2A. Prasad, 3P. H. Quanjer, 1J. Stocks. 1UCL Institute of Child Health, London, UK, 2Great Ormond Street Hospital, London, UK, 3Sophia Children’s Hospital, Rotterdam, The Netherlands

Background: Serial spirometry measurement can facilitate the identification of individuals with rapid lung function decline. In clinical practice, however, repeated measurements are commonly interpreted using cross-sectional reference data, despite well recognised limitations (Van Pelt et al. AJRCCM 1994).

Aim: To assess adjustments for the correlated nature of repeated measurements in order to interpret changes over time using cross-sectional reference data.

Methods: Normal lung growth can be viewed as roughly a constant Z-score (adjusted for height, age and sex) over time, with the uncertainty around the change normally distributed around a mean change of 0 Z-scores. The standard deviation of the change depends on the correlation (r) between the two measurements (Cole, Human growth in context, 1998). The adjusted Z-score for change (Zc) adjusts the observed Z-score change for the baseline Z-score and the interval between measurements. Spirometry data from 938 healthy, white subjects (aged 5–18 years) measured twice over time intervals 0.1 and 5 years, were collated to estimate the occasion-to-occasion correlation coefficient (r) for FEV1 Z-score (Stanojevic et al. AJRCCM 2008). Measurements less than one year apart had an r value of 0.90, such that normal within-subject variability in health after one year is 0 ± 0.88 Z-scores. Changes in spirometry were also assessed after a one-year interval in 98 children with cystic fibrosis (aged 4–12 years) to determine the clinical implications of using the Zc.

Results: Based on a cross-sectional cut-off of <−1.64 Z-scores, 32/98 patients had an abnormal result at baseline, of whom 21 remained abnormal after one …

View Full Text

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.