Patient characteristics: extension or Observation

[pieter-edelman-nictiz]

There are several zibs that could be considered characteristics of the patient, like language proficiency, eduction level, life stance, nationality, marital status, etc. These zibs could potentially by modeled in two ways:

• As a characteristic of the patient (or contact person, practitioner), directly in the relevant resource
• As a distinct Observation

Looking at the FHIR core spec and different IG's, there doesn't seem to be a clear preference for one of these two ways, both methods are used. Some things to note:

1. For some of these concepts, a native FHIR element or core extension exists.
2. For some of these concepts, additional information is or could be relevant (like duration) and so an Observation is more logical.
3. For some of these Observations, one could make the case that there's a use case to query them independently. However, it's quite hard to reason about this in advance.

For reason number 2, our preference is to use Observations for this kind of information, unless a specific core element or extension is available (or unless there's a good reason to do things in another way). This preference needs to be written down in the profiling guidelines.

[pieter-edelman-nictiz]

This has been added to the profiling guidelines through https://bits.nictiz.nl/browse/MM-4835