There are a number of ways to assess the ‘normality’ of a spirometry result.

The normality (or abnormality) is most commonly determined as a percentage of predicted (%pred) or as a standardised residual (SR). Both of these observations however, require a normal range, otherwise the volumes and flows generated by the subject are arbitrary.

The upper limit of normal (ULN) and lower limit (LLN) define’s the margin of normality for a particular subject. Reference data (previously termed ‘predicted values’) are generated by observing large numbers of the population across different ages, height and gender.

Reference values and normal ranges for spirometry are calculated from the following information:

• Age
• Height
• Gender
• Ethnicity

The most commonly used is the percentage of the mean, i.e 20% above and 20% below. Evidently, this assumes all subjects, of all ages, and of all genders will be normal should their values fall between these two fixed, and rather arbitrary figures.

The other way is to use regression analysis. This is the comparison between the change in a spirometric value, such as the FEV1, with another variable such as age. In other words, it considers the change in normality of the FEV1 at each different age group.

Just think, as we grow our thoracic cage becomes larger and the lungs develop to fill the space. As our lungs age and become less elastic, the FEV1 value will decline – this decline is indeed ‘normal’ and the influence of age on the ‘normal range’ will be greater in the older populations. Figure 5.20 illustrates how FEV1 declines normally as we age.

Even in a population of normal subjects, there are a small group of outliers at the upper and lower limit which are considered outside of the normal range. A value of 1.645 standardised residuals above and below the predicted value encompasses 90% of the normal population; the remaining 5% above and 5% below this normal range are technically considered abnormal.