Article Text

Download PDFPDF
Feasible and simple exclusion criteria for pulmonary reference populations
  1. Ane Johannessen1,
  2. Ernst R Omenaas1,
  3. Geir Egil Eide2,
  4. Per S Bakke3,
  5. Amund Gulsvik3
  1. 1Centre for Clinical Research, Haukeland University Hospital, Bergen and Institute of Medicine, University of Bergen, Norway
  2. 2Centre for Clinical Research, Haukeland University Hospital, Bergen and Section for Epidemiology and Medical Statistics, Department of Public Health and Primary Health Care, University of Bergen, Norway
  3. 3Department of Thoracic Medicine, Haukeland University Hospital, Bergen and Institute of Medicine, University of Bergen, Norway
  1. Correspondence to:
    Dr Ane Johannessen
    Centre for Clinical Research, Haukeland University Hospital, N-5021 Bergen, Norway; Ane.Johannessen{at}


Background: International guidelines recommend that pulmonary reference populations consist of never-smokers without respiratory diseases or symptoms, but the diseases and symptoms are not clearly specified. The present study aimed to identify simple exclusion criteria for defining pulmonary reference populations.

Methods: Based on a random sample from a general population (the parent population), 2358 subjects aged 26–82 years performed spirometric tests. From this sample, subjects were stepwise excluded according to self-reported obstructive lung diseases, symptoms and smoking history. Four increasingly more healthy respiratory reference populations were formed. Prediction equations for the median and lower limit of normal lung function were derived using quantile regression analysis.

Results: Subjects without self-reported obstructive lung diseases or the cardinal respiratory symptoms of breathlessness, cough or wheeze (population B), never-smokers without cardinal symptoms (population C) and never-smokers without any respiratory symptoms (population D) constituted 50% (n = 1184), 23% (n = 539) and 14% (n = 331) of the parent population (population A), respectively. The largest discrepancy between prediction equations was found between the parent population and the population without cardinal respiratory symptoms (population B) (p<0.05). Minor changes in the reference equations were also seen when excluding ever-smokers (population C). There was no additional change with exclusion of other respiratory symptoms (population D). Age-related decline in lung function was steepest in the parent population.

Conclusions: Obstructive lung diseases, smoking history, breathlessness, cough and wheeze are optimal exclusion criteria for a pulmonary reference population. Further validation of the exclusion criteria identified in this study is recommended with identical wording in other and larger multinational populations.

  • FEV1, forced expiratory volume in 1 s
  • FVC, forced vital capacity
  • LLN, lower limit of normal

Statistics from

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.


  • Published Online First 27 March 2007

  • The Hordaland County Cohort Study was funded from the Royal Norwegian Council for Scientific and Industrial Research and the Norwegian Research Council.

  • Competing interests: None declared.