* Medicines Evaluation Unit, North West Lung Centre, The University of Manchester, Wythenshawe Hospital, Manchester, UK # Dept of Mathematics, Manchester University, Manchester, UK CORRESPONDENCE: Kay Roy, Research Registrar, Medicines Evaluation Unit, North West Lung Centre, Wythenshawe Hospital, Manchester, UK Multiple flow rates Fe NO data two-compartment mathematical modelling can estimate NO airway wall concentration (Caw NO ), airway wall diffusing capacity (Daw NO ), alveolar concentration (Calv NO ) and maximal NO flux (Jaw NO ). To compare modelling based on linear, non linear and mixed linear and non linear analyses in COPD. Fe NO was measured using the Niox analyser (Aerocrine) at flow rates: 10, 30, 50, 100 and 200ml/s in 50 COPD patients and the data applied to 4 different methods. All methods showed that current smoking reduced Caw NO . Jaw NO data differed between methods (Table 1). All the methods showed that current smoking did not affect Calv NO or Daw NO . Comparison of the methods showed that Calv NO and Daw NO data were significantly different between all methods, Jaw NO was different for most between method comparisons, while there was agreement between all the methods for Caw NO . Smoking in COPD reduces Caw NO , but not Calv NO and Daw NO . Jaw NO, Calv NO and Daw NO data are model dependent parameters. Caw NO findings were model independent, and hence the most robust modelled parameter. View this table: in this window in a new window   Table 1—