COPD Phenotype Estimation Model
The following on-line application is an implementation of the model described in: Pistolesi M, Camiciottoli G, Paoletti M et al. Identification of a predominant COPD phenotype in clinical practice. Respiratory Medicine 2008;102:367-376.
Please insert the requested data and press "Calculate":
 
Clinical history data:
Sputum (0-occasional, 1-chronic, 2-purulent)
Physical examination data:
Adventitious breath sounds (0/1) Chest hyperresonance (0/1)
Functional data:
FEV1/VC (%)
Chest Radiography data:
Increased lung volume (0/1) Reduced lung density (0/1)
Increased vascular markings (0/1) Bronchial wall thickening (0/1)
Model Output
   
Instructions: Please insert variables derived from physical examination, clinical history and chest radiography as 1 when present or 0 when absent. (A radiographic variable is considered to be present when at least one of the related specific findings was detected). In the case of FEV1/VC (ratio of forced expiratory volume in the first second over slow vital capacity) enter the actual value of the continuous variable (use comma for decimals).


Results interpretation: The statistical model output reflects the specific predominant mechanism of airflow limitation for this patient. The model was trained over a database of hundreds of patients as described in: Pistolesi M, Camiciottoli G, Paoletti M et al. Identification of a predominant COPD phenotype in clinical practice. Respiratory Medicine 2008;102:367-376. Studying the probability density curves of the classification scores of the learning set used to develop the model the found cut-off value  was at a score of 0.56.
This value is used to subdivide COPD patients into two subsets: patients with a score higher than 0.56 (Group A) and patients with a score equal or lower than 0.56 (Group B).
The comparison between the two subsets usually shows that they differ significantly for a great number of categorical and continuous variables which largely confer to Group A the characteristics of Type A (emphysematous) clinical presentation of chronic airway obstruction and to the Group B the characteristics of Type B (bronchial) clinical presentation of chronic airway obstruction, as described by Burrows and colleagues.
Values higher or lower  from the cutoff score are related to unbalanced patients, values near 0,56 are typical in balanced patients.

Please use the application with data collected from COPD patients only.
 
  

COPD phenotype estimation model, CLIP-COPD project
Copyright (c) 2008-2009, Università degli Studi di Firenze
All Rights Reserved
  
 Contact webmaster: Ing. Matteo Paoletti, E-mail: ing.paoletti at mpspace.com