In our source data, we have multiple specialities for a given provider. Seeing as how the PROVIDER_ID column is meant to be unique in the PROVIDER table, what is the best approach to store the multiple entries for each clinician? I thought of using the OBSERVATION table. However that is used for patients generally. How about the FACT_RELATIONSHIP table to indicate the secondary specialities for each provider? Anyone have useful suggestions?
Yes, that’s a known problem. It usually is a an administrative artefact. A surgeon doesn’t usually also practice ophthalmology. They can barely keep up with one specialty. So, try to find either the most frequently or the most advanced and fine-grained specialty and use that one.
We are experiencing the same problem. We have physicians with multiple specialties.
For example pediatrics, pediatric gastroenterology, pediatric pulmonology, adult pulmonology.
Is there any chance that the problem has been resolved since the last response to this thread?
@robyn.rubin: Isn’t the resolution to use the least common denominator, after taking out some rare outliers? For example, this guy may get a very occasional adult patient, but otherwise treats children. The common denominator would be “pediatrics”. Makes sense?