We have just published cTNM & pTNM as CLUSTER archetypes. They have been on a journey over time CLUSTER class to OBSERVATION and back again to CLUSTER, with the intent of ensuring that the TNM will move with a cancer diagnosis archetype if exchanged etc. The TNM is effectively regarded as a qualifier of the diagnosis, intended to be inserted into the ‘Specific details’ SLOT in the EVAL.problem_diagnosis.
When we modelled it as a standalone OBSERVATION it meant that we needed to add the Diagnosis in to the TNM representation in order to provide context, especially if the patient had more than one cancer, and then it got more complicated if these diverged etc. So we made the pragmatic decision to go back to CLUSTER, largely on the advice of experienced anatomical pathologist @sabine.leh.
We don’t currently have a formal ‘date of assertion/staging’ or similar within the TNM archetypes. As the record is updated, the addition of the CLUSTER will effectively record when it was added to the record, and the date that the EVAL.problem_diagnosis will/should also be updated simultaneously.
I wonder if it is reasonable to record the ISS as a CLUSTER, and use it in the same place as the TNM archetypes? That SLOT is intended to " include structured detail about the grading or staging of the diagnosis; diagnostic criteria, classification criteria or formal severity assessments such as Common Terminology Criteria for Adverse Events."
I’m curious about your inclusion of the staging criteria as categories as justification for the final stage/grade, which makes it more like a score, and we usually model scales or scores as a reliably repeatable OBSERVATION. Or do we only need to record the grading/conclusion/assessment, based on a diverse number of criteria, each of which would possibly be recorded elsewhere in lab test results, but fit better within the EVALUATION model (albeit nested within the EVAL as a CLUSTER like TNM).
I tend to favour the CLUSTER within the EVAL.problem_diagnosis approach. In this way we could include the ISS stage alongside the TNM assessment for a given cancer diagnosis, plus potentially other staging models as they might be developed.
I’d also consider recording only the stage itself - reflecting the reference within the archetype. Do you need to record the result categories of each of the criteria in the record rather than the actual results?
Tricky question. My 2c…