The patient demographics for pulmonary function testing usually includes patient identifiers such as:

• Patient / Hospital ID number
• Last name
• First name
• Date of birth

It also consists of the anthropometric measurements required for reference equations. At a minimum these are:

• Height
• Gender
• Age
• Ethnicity

But can also include:

• Weight
• BMI

The primary factor used in most reference equations is height. This is because there is a relatively linear relationship between a subject’s height and their lung capacity. The linearity of this relationship however, is subject both to the normal distribution of values (usually considered to be a bell-shaped curve) and differences between ethnicity-based populations. It should also be noted that individuals at the extreme ends of the normal distribution of height (very short or very tall) are poorly represented in study populations and their reference values are therefore extrapolated with uncertain accuracy.

Gender is likely the second most important factor and this is due to fact that within any given population males have larger lungs than females.  In addition male lung function declines faster with increasing age than does female.

Age is a factor because lung volumes and expiratory flow rates tend to peak in early adulthood, approximately between the ages of 18 and 25, and declines thereafter. Age appropriate reference equations should therefore be used, but during the developmental stages of infancy, childhood and adolescence the relationship between height and lung capacity is not as linear as it is in adulthood. This is partly due to the fact that the transition between infancy, childhood, adolescence and adulthood does not follow a strict chronological timetable nor are these transitions necessarily related to height or weight. At the other end of the spectrum, the reference values for the elderly, particularly those over 80 years of age, tend to be poorly represented in study populations and their reference values are therefore extrapolated with uncertain accuracy.

Difference in lung capacities between different ethnicities were noted quite early in the study of human lung function. Caucasians are considered to have the largest lung capacity for a specific height; Asians approximately 6% less than this and Blacks approximately 12% less. These apparent differences between ethnicities must be tempered first by the fact that reference equations within a single ethnicity (Caucasian) show a range of values that is the same magnitude as the average difference between Caucasians and Blacks. Second, although the reason for these differences are likely partly genetic, they are also affected by the environment and diet during an individual’s developmental stages, and these factors are rapidly changing across the world. Third, there is no clear definition for ethnicity which makes ascribing a given individual to a specific ethnicity difficult.

Since the anthropometric demographics have a direct effect on normal reference values however, errors, which are most commonly data entry errors, will affect the interpretation of test results. Specifically, entry errors involving:

Patient / Hospital ID:

May cause results to be attached to a different subject’s records. When this occurs test system software may populate other demographic fields (height, age, ethnicity, etc.) with incorrect information. Trended results will also be mis-matched causing errors when previous results are compared.

Date of birth:

An incorrect age will cause reference values to be incorrectly calculated, leading to either under- or over-estimation of the percent predicted and LLN.

Height:

Height is supposed to be measured with a subject’s shoes removed, standing straight with their back against the wall. If the subject is unable to stand straight due to musculoskeletal issues (kyphoscoliosis, wheelchair-bound) then height can be estimated from arm-span.

Height can be mis-entered or mis-measured. In the USA, height is usually measured as feet and inches but this must be converted to inches and this leads to mis-entries such as 5 feet 7 inches as 57 inches or 6 feet 2 inches as 62 inches. If shoes are not removed height may be overestimated. If arm-span is not measured in individuals unable to stand straight then height may be underestimated. If a testing location lacks the ability or the willingness to measure height and instead requests this information from the patient, the patient may mis-remember or guess.

Gender:

Gender can be entered incorrectly but there can also be data management issues with transgender individuals. An individual’s gender during developmental stages determines lung capacity and this does not change after gender reassignment. Data management systems however, may force the entry of an incorrect gender, or the individual themselves may supply their reassigned gender, rather than their original gender. Although a patient’s name can be discordant with the entered gender, many names are not gender-specific.

Ethnicity:

As mentioned, there are limits to the ability of ethnicity to generate the correct reference values. Even when an individual’s ethnicity may be apparent, either to the staff performing the test or to the individual themselves, since there are no clear definition for ethnicity nor specific guidelines for assigning ethnicity, ethnicity may be mis-assigned.

Weight and BMI:

Although weight and BMI can be informative about a subject’s body habitus, these values are almost never a factor in reference equations.  Errors in weight and BMI therefore tend to have little effect on the interpretation of test results other than to misinform the reviewer about the patient’s physical status.

Summary:

	For this Error:	Reference values, LLN will be:	Percent Predicted will be:
Age	Entered age is younger than actual age	↑	↓
	Entered age is older than actual age.	↓	↑
Height	Entered height is shorter than actual height	↓	↑
	Entered height is taller than actual height	↑	↓
Gender	Entered as male, should be female	↑	↓
	Entered as female, should be male	↓	↑
Ethnicity	Entered as Caucasian, should be other	↑	↓
	Entered as Other, should be Caucasian	↓	↑

Recommendation:

A subject’s demographics should be reviewed and any apparent discrepancies noted, investigated and corrected before the test results themselves are reviewed. In addition, knowledge of a subject’s anthroprometic demographics sets certain expectations about the overall appearance of the test results and this itself may help to direct the review and interpretation of test results.

 

PFT Interpretation by Richard Johnston is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License