I have recently been reading on the topic of composition in photography. There very quickly developed a dichotomy in my mind - the need for discipline and yet, at the same time, freedom. The discipline is required to find an image and then to identify the rules by which the composition should be structured. The creativity is in identifying when the rules should be broken, and in using the rules to enhance the message that you wish to get across, to trigger a response from the viewer. The breaking of the rules becomes a part of the message itself.
This got me thinking about the interpretation of lung function test results. We have the rules, i.e. ATS/ERS recommended standards. We also have a very dirty system in which few, if any, patients actually fit the ideal model of full and consistent cooperation. We will, on occasion, manage to get test results that satisfy the long list of ATS/ERS criteria for acceptable tests but we will also get many who cannot produce data that satisfy all the criteria for a variety of reasons.
Does failure to get acceptable data mean the test results need to be discarded? We need to evaluate the data we have against the clinical question being asked. For example, a string of good forced expiratory efforts, with variable inspirations, and which, when aligned to RV, show repeatable flows from about the mid point of the best inspiration but no evidence of concavity, could allow the deduction that there is unlikely to be airway collapse. This would allow a conclusion that there is unlikely to be significant airflow obstruction despite not having acceptable FEV1's or FVC's. This could well provide a useful answer to the clinical question "Is there airflow obstruction present?"
I think we need to be careful not to follow the rules blindly and recognise when the message requires the rules to be broken. I am not suggesting we need to invoke creativity into our interpretation, but rather that we recognise the need to see the "standards" for what they are - committee derived recommendations. I see the role of an interpretation as providing an answer to the clinical question being asked. That is why it is so important that we are given a clear clinical question to answer when we are asked to perform lung function tests. Only when we know the question being asked, can we make decisions regarding whether the data we are obtain can indeed provide an answer to the question or whether we should simply say "No acceptable data able to be obtained."
This point of view demands a number of other things in consequence. In particular we, as scientists, need to pursue active involvement in clinical education activity within the lab/department and also better flexibility in our reporting software so that we can better manage the data that we include in our reports e.g. report an FEV1 but not an FVC and the other data derived from it.
We need the rules, but also need to know when to break the rules in the interests of delivering a message. In the meantime at least though, hold the creativity!