Article Text

Download PDFPDF

What defines abnormal lung function?
  1. Martin R Miller
  1. Department of Medicine, University Hospital Birmingham NHS Trust, Birmingham B29 6JD, UK; martin.miller{at}

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.

A major concern with the GOLD criteria for defining chronic obstructive pulmonary disease (COPD) is that the use of 70% as a fixed cut-off for forced expiratory volume in 1 s (FEV1) will add a significant number of false positives compared with the use of a true lower limit of normal.1 2 The recent paper in Thorax by Mannino et al3 argues that this false positive rate is acceptable because they found that these false positive subjects had an increased hazard of death when compared with those with FEV1 >70%. This finding is to be expected because a group with a less good level of lung function is being compared with a group with better lung function, even though both groups are within the normal range. For example, in the Copenhagen City Heart Study data, if the 8101 subjects whose FEV1% in standardised residuals (SR) was >0 (ie, above predicted) are compared with the 1876 subjects whose FEV1%SR was between 0 and −0.5, the latter have a hazard ratio (HR) for death of 1.09 (95% CI 1.02 to 1.17) (Dr Peter Lange, personal communication). If, instead, those above predicted are compared with the 1292 subjects with FEV1%SR between −0.5 and −1.0, then the latter have a HR of 1.24 (95% CI 1.14 to 1.34). So these 3168 normal subjects have results for FEV1% well within the normal range but have an increased HR for death when compared with that part of the population with the best lung function. This effect is even more true for FEV1 itself.

Is it then justifiable to label an asymptomatic individual with a disease on the basis of spirometric parameters that are within the accepted normal range just because they have an increased risk of death? If this argument is further developed, then male sex is also a disease since life expectancy in men is lower than in women.

Labelling an individual as having a disease can have a bad psychological effect and, if there is as yet no proven treatment for the presumed condition, this is an even more unacceptable state of affairs. Disease has been defined as “an impairment of health or a condition of abnormal functioning” ( and as “a disorder of structure or function in a human, especially one that produces specific symptoms or that affects a specific part” (Oxford English Dictionary). The current GOLD definition of COPD does not meet these requirements for defining this disease and must be changed. Researchers may need to find another term to describe their point of interest rather than labelling normal individuals incorrectly as having the disease COPD.


View Abstract


  • Competing interests: None.